[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Emacs-diffs] master accb7b7 1/3: Merge remote-tracking branch 'origin/m
|
From: |
Tom Tromey |
|
Subject: |
[Emacs-diffs] master accb7b7 1/3: Merge remote-tracking branch 'origin/master' into feature/bignum |
|
Date: |
Sat, 11 Aug 2018 15:47:34 -0400 (EDT) |
branch: master
commit accb7b7ecc19f85c2750ded1046a464bc73c6a52
Merge: f822a25 53483df
Author: Tom Tromey <address@hidden>
Commit: Tom Tromey <address@hidden>
Merge remote-tracking branch 'origin/master' into feature/bignum
---
.gitignore | 1 +
admin/MAINTAINERS | 14 +-
admin/find-gc.el | 2 +-
admin/merge-gnulib | 9 +-
build-aux/config.guess | 114 +-
build-aux/config.sub | 10 +-
configure.ac | 7 +-
doc/emacs/dired.texi | 5 +
doc/emacs/files.texi | 2 +
doc/emacs/fixit.texi | 14 +-
doc/emacs/indent.texi | 21 +-
doc/emacs/maintaining.texi | 16 +-
doc/emacs/misc.texi | 4 +-
doc/emacs/mule.texi | 19 +-
doc/emacs/regs.texi | 4 +-
doc/lispref/debugging.texi | 115 +-
doc/lispref/edebug.texi | 12 +-
doc/lispref/eval.texi | 3 -
doc/lispref/files.texi | 8 +-
doc/lispref/frames.texi | 10 +-
doc/lispref/lists.texi | 23 +-
doc/lispref/minibuf.texi | 2 +-
doc/lispref/numbers.texi | 36 +-
doc/lispref/sequences.texi | 8 +-
doc/lispref/strings.texi | 27 +-
doc/lispref/threads.texi | 23 +-
doc/misc/efaq.texi | 4 +-
doc/misc/ert.texi | 8 +-
doc/misc/flymake.texi | 2 +-
doc/misc/tramp.texi | 76 +-
etc/NEWS | 155 +-
etc/NEWS.26 | 8 +
etc/emacs.service | 2 +-
etc/themes/adwaita-theme.el | 4 -
etc/themes/deeper-blue-theme.el | 4 -
etc/themes/dichromacy-theme.el | 4 -
etc/themes/leuven-theme.el | 1 -
etc/themes/light-blue-theme.el | 4 -
etc/themes/manoj-dark-theme.el | 4 -
etc/themes/misterioso-theme.el | 4 -
etc/themes/tango-dark-theme.el | 4 -
etc/themes/tango-theme.el | 4 -
etc/themes/tsdh-dark-theme.el | 4 -
etc/themes/tsdh-light-theme.el | 5 -
etc/themes/wheatgrass-theme.el | 4 -
etc/themes/whiteboard-theme.el | 4 -
etc/themes/wombat-theme.el | 4 -
lib-src/Makefile.in | 8 +-
lib-src/etags.c | 2 +-
lib-src/ntlib.c | 67 +
lib-src/pop.c | 8 +-
lib/Makefile.in | 8 +-
lib/gnulib.mk.in | 170 +-
lib/ieee754.in.h | 222 +
lib/regcomp.c | 3944 ++++++++++++++++++
lib/regex.c | 81 +
lib/regex.h | 658 +++
lib/regex_internal.c | 1740 ++++++++
lib/regex_internal.h | 911 +++++
lib/regexec.c | 4324 ++++++++++++++++++++
lib/unistd.in.h | 4 +-
lib/warn-on-use.h | 64 +-
lisp/auth-source.el | 2 +-
lisp/bookmark.el | 7 +-
lisp/calendar/todo-mode.el | 202 +-
lisp/char-fold.el | 2 +-
lisp/comint.el | 19 +-
lisp/cus-theme.el | 63 +-
lisp/custom.el | 257 +-
lisp/dired-aux.el | 37 +-
lisp/dired.el | 3 +
lisp/emacs-lisp/backtrace.el | 916 +++++
lisp/emacs-lisp/cl-macs.el | 5 +-
lisp/emacs-lisp/cl-print.el | 301 +-
lisp/emacs-lisp/debug.el | 463 +--
lisp/emacs-lisp/easy-mmode.el | 28 +-
lisp/emacs-lisp/edebug.el | 276 +-
lisp/emacs-lisp/ert.el | 41 +-
lisp/emacs-lisp/lisp-mode.el | 26 +-
lisp/emacs-lisp/map-ynp.el | 18 +-
lisp/emacs-lisp/rx.el | 32 +-
lisp/emacs-lisp/subr-x.el | 12 +-
lisp/env.el | 6 +-
lisp/epg-config.el | 27 +-
lisp/eshell/em-dirs.el | 3 +
lisp/eshell/em-ls.el | 1 +
lisp/eshell/em-unix.el | 10 +
lisp/eshell/esh-cmd.el | 51 +-
lisp/eshell/esh-ext.el | 1 +
lisp/files.el | 22 +-
lisp/format.el | 2 +
lisp/gnus/gnus-art.el | 6 +
lisp/gnus/gnus-sum.el | 8 +-
lisp/ielm.el | 14 +-
lisp/imenu.el | 13 +-
lisp/indent.el | 9 +-
lisp/international/fontset.el | 9 +-
lisp/international/mule-cmds.el | 9 +-
lisp/international/mule-conf.el | 9 +-
lisp/international/mule-diag.el | 3 +-
lisp/international/quail.el | 20 +-
lisp/jsonrpc.el | 88 +-
lisp/language/thai.el | 4 +-
lisp/ldefs-boot.el | 874 ++--
lisp/mh-e/mh-comp.el | 100 +-
lisp/mh-e/mh-funcs.el | 2 +
lisp/mh-e/mh-identity.el | 27 +
lisp/net/rcirc.el | 2 +-
lisp/net/rlogin.el | 8 +-
lisp/net/soap-client.el | 21 +-
lisp/net/tramp-gvfs.el | 37 +-
lisp/net/tramp-sh.el | 6 +-
lisp/net/tramp.el | 12 +-
lisp/net/trampver.el | 2 +-
lisp/profiler.el | 6 +-
lisp/progmodes/cc-engine.el | 12 +-
lisp/progmodes/elisp-mode.el | 2 +-
lisp/progmodes/grep.el | 4 +-
lisp/progmodes/hideif.el | 2 +-
lisp/progmodes/python.el | 95 +-
lisp/progmodes/subword.el | 4 +-
lisp/register.el | 16 +-
lisp/scroll-bar.el | 24 +-
lisp/shadowfile.el | 478 +--
lisp/shell.el | 4 +-
lisp/simple.el | 12 +-
lisp/subr.el | 31 +-
lisp/term.el | 20 +-
lisp/term/tty-colors.el | 14 +-
lisp/textmodes/flyspell.el | 19 +-
lisp/textmodes/ispell.el | 5 +-
lisp/textmodes/reftex-vars.el | 6 +-
lisp/thingatpt.el | 2 +-
lisp/vc/add-log.el | 77 +-
lisp/vc/diff-mode.el | 35 +
lisp/vc/diff.el | 5 +-
lisp/vc/log-edit.el | 6 +-
lisp/w32-fns.el | 2 +-
lisp/wdired.el | 51 +-
m4/builtin-expect.m4 | 49 +
m4/eealloc.m4 | 31 +
m4/extern-inline.m4 | 16 +-
m4/glibc21.m4 | 34 +
m4/gnulib-comp.m4 | 34 +
m4/ieee754-h.m4 | 21 +
m4/mbstate_t.m4 | 41 +
m4/regex.m4 | 300 ++
nt/inc/ms-w32.h | 12 +-
nt/inc/sys/socket.h | 5 +
src/Makefile.in | 2 +-
src/alloc.c | 65 +-
src/callint.c | 2 +-
src/casetab.c | 3 +-
src/character.c | 10 +-
src/chartab.c | 4 +-
src/dbusbind.c | 4 +-
src/deps.mk | 9 +-
src/dired.c | 2 +-
src/dispextern.h | 2 +-
src/editfns.c | 68 +-
src/emacs.c | 19 +-
src/eval.c | 20 +-
src/fileio.c | 35 +-
src/floatfns.c | 7 -
src/fns.c | 103 +-
src/frame.c | 27 +-
src/gnutls.c | 38 +-
src/image.c | 2 +-
src/intervals.c | 2 +-
src/intervals.h | 2 +-
src/json.c | 11 +-
src/keyboard.c | 47 +-
src/lisp.h | 14 +-
src/lread.c | 18 +-
src/menu.c | 6 +-
src/nsfns.m | 4 +-
src/nsmenu.m | 2 +-
src/nsselect.m | 6 +-
src/print.c | 48 +-
src/process.c | 33 +-
src/{regex.c => regex-emacs.c} | 2751 +++----------
src/regex-emacs.h | 197 +
src/regex.h | 654 ---
src/search.c | 36 +-
src/syntax.c | 11 +-
src/terminal.c | 2 +-
src/textprop.c | 22 +-
src/thread.c | 43 +-
src/thread.h | 9 +-
src/w32cygwinx.c | 37 +-
src/w32term.c | 2 +-
src/xdisp.c | 55 +-
src/xfaces.c | 4 +-
src/xfns.c | 4 +-
src/xmenu.c | 4 +-
src/xselect.c | 4 +-
src/xwidget.c | 2 +-
test/lisp/auth-source-tests.el | 20 +
test/lisp/calendar/todo-mode-tests.el | 190 +-
test/lisp/custom-tests.el | 87 +
test/lisp/emacs-lisp/backtrace-tests.el | 436 ++
test/lisp/emacs-lisp/cl-print-tests.el | 178 +-
.../edebug-resources/edebug-test-code.el | 9 +-
test/lisp/emacs-lisp/edebug-tests.el | 29 +-
test/lisp/emacs-lisp/ert-tests.el | 2 +-
test/lisp/emacs-lisp/lisp-mode-tests.el | 23 +
test/lisp/emacs-lisp/package-tests.el | 31 +-
test/lisp/emacs-lisp/subr-x-tests.el | 47 +
test/lisp/epg-tests.el | 38 +-
test/lisp/filenotify-tests.el | 4 +-
test/lisp/net/secrets-tests.el | 6 +-
test/lisp/net/tramp-archive-tests.el | 18 +-
test/lisp/net/tramp-tests.el | 102 +-
test/lisp/progmodes/compile-tests.el | 46 +-
test/lisp/shadowfile-tests.el | 945 +++++
test/lisp/wdired-tests.el | 129 +
test/src/editfns-tests.el | 8 +-
test/src/fns-tests.el | 11 +
test/src/{regex-tests.el => regex-emacs-tests.el} | 6 +-
test/src/thread-tests.el | 125 +-
220 files changed, 20451 insertions(+), 5111 deletions(-)
diff --git a/.gitignore b/.gitignore
index d3712b0..26fe4bb 100644
--- a/.gitignore
+++ b/.gitignore
@@ -57,6 +57,7 @@ lib/execinfo.h
lib/fcntl.h
lib/getopt.h
lib/getopt-cdefs.h
+lib/ieee754.h
lib/inttypes.h
lib/libgnu.a
lib/limits.h
diff --git a/admin/MAINTAINERS b/admin/MAINTAINERS
index 753a676..10633a8 100644
--- a/admin/MAINTAINERS
+++ b/admin/MAINTAINERS
@@ -37,7 +37,7 @@ Kenichi Handa
Mule
Stefan Monnier
- src/regex.c
+ src/regex-emacs.c
src/syntax.c
src/keymap.c
font-lock/jit-lock/syntax
@@ -61,7 +61,7 @@ Michael Albinus
lisp/net/tramp*.el
lisp/url/url-tramp.el
doc/misc/tramp*.texi
- test/lisp/net/tramp-tests.el
+ test/lisp/net/tramp*-tests.el
test/lisp/url/url-tramp-tests.el
D-Bus
@@ -210,11 +210,21 @@ Paul Eggert
Michael Albinus
src/inotify.c
lisp/autorevert.el
+ lisp/files.el (file-name-non-special)
lisp/eshell/em-tramp.el
+ lisp/net/ange-ftp.el
lisp/notifications.el
+ lisp/shadowfile.el
test/lisp/autorevert-tests.el
+ test/lisp/files-tests.el (file-name-non-special)
+ test/lisp/shadowfile-tests.el
test/src/inotify-test.el
+ Secret Service API in
+ lisp/auth-source.el
+ doc/misc/auth.texi
+ test/lisp/auth-source-tests.el
+
Nicolas Petton
lisp/emacs-lisp/subr-x.el
lisp/arc-mode.el
diff --git a/admin/find-gc.el b/admin/find-gc.el
index fb56403..e8cc113 100644
--- a/admin/find-gc.el
+++ b/admin/find-gc.el
@@ -57,7 +57,7 @@ Each entry has the form (FUNCTION . FUNCTIONS-IT-CALLS).")
"keymap.c" "sysdep.c" "buffer.c" "filelock.c"
"insdel.c" "marker.c" "minibuf.c" "fileio.c"
"dired.c" "cmds.c" "casefiddle.c"
- "indent.c" "search.c" "regex.c" "undo.c"
+ "indent.c" "search.c" "regex-emacs.c" "undo.c"
"alloc.c" "data.c" "doc.c" "editfns.c"
"callint.c" "eval.c" "fns.c" "print.c" "lread.c"
"syntax.c" "unexcoff.c"
diff --git a/admin/merge-gnulib b/admin/merge-gnulib
index 39dfaee..abb1929 100755
--- a/admin/merge-gnulib
+++ b/admin/merge-gnulib
@@ -35,9 +35,9 @@ GNULIB_MODULES='
fcntl fcntl-h fdatasync fdopendir
filemode filevercmp flexmember fpieee fstatat fsusage fsync
getloadavg getopt-gnu gettime gettimeofday gitlog-to-changelog
- ignore-value intprops largefile lstat
+ ieee754-h ignore-value intprops largefile lstat
manywarnings memrchr minmax mkostemp mktime nstrftime
- pipe2 pselect pthread_sigmask putenv qcopy-acl readlink readlinkat
+ pipe2 pselect pthread_sigmask putenv qcopy-acl readlink readlinkat regex
sig2str socklen stat-time std-gnu11 stdalign stddef stdio
stpcpy strtoimax symlink sys_stat sys_time
tempname time time_r time_rz timegm timer-time timespec-add timespec-sub
@@ -46,11 +46,12 @@ GNULIB_MODULES='
'
AVOIDED_MODULES='
- close dup fchdir fstat
- malloc-posix msvc-inval msvc-nothrow
+ btowc close dup fchdir fstat langinfo lock
+ malloc-posix mbrtowc mbsinit msvc-inval msvc-nothrow nl_langinfo
openat-die opendir raise
save-cwd select setenv sigprocmask stat stdarg stdbool
threadlib tzset unsetenv utime utime-h
+ wchar wcrtomb wctype-h
'
GNULIB_TOOL_FLAGS='
diff --git a/build-aux/config.guess b/build-aux/config.guess
index 2b79f6d..d4fb321 100755
--- a/build-aux/config.guess
+++ b/build-aux/config.guess
@@ -2,7 +2,7 @@
# Attempt to guess a canonical system name.
# Copyright 1992-2018 Free Software Foundation, Inc.
-timestamp='2018-07-06'
+timestamp='2018-08-02'
# This file is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by
@@ -84,8 +84,6 @@ if test $# != 0; then
exit 1
fi
-trap 'exit 1' 1 2 15
-
# CC_FOR_BUILD -- compiler used by this script. Note that the use of a
# compiler to aid in system detection is discouraged as it requires
# temporary files to be created and, as you can see below, it is a
@@ -96,34 +94,39 @@ trap 'exit 1' 1 2 15
# Portable tmp directory creation inspired by the Autoconf team.
-set_cc_for_build='
-trap "exitcode=\$?; (rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null) &&
exit \$exitcode" 0 ;
-trap "rm -f \$tmpfiles 2>/dev/null; rmdir \$tmp 2>/dev/null; exit 1" 1 2 13 15
;
-: ${TMPDIR=/tmp} ;
- { tmp=`(umask 077 && mktemp -d "$TMPDIR/cgXXXXXX") 2>/dev/null` && test -n
"$tmp" && test -d "$tmp" ; } ||
- { test -n "$RANDOM" && tmp=$TMPDIR/cg$$-$RANDOM && (umask 077 && mkdir $tmp
2>/dev/null) ; } ||
- { tmp=$TMPDIR/cg-$$ && (umask 077 && mkdir $tmp 2>/dev/null) && echo
"Warning: creating insecure temp directory" >&2 ; } ||
- { echo "$me: cannot create a temporary directory in $TMPDIR" >&2 ; exit 1 ; }
;
-dummy=$tmp/dummy ;
-tmpfiles="$dummy.c $dummy.o $dummy.rel $dummy" ;
-case ${CC_FOR_BUILD-},${HOST_CC-},${CC-} in
- ,,) echo "int x;" > "$dummy.c" ;
- for c in cc gcc c89 c99 ; do
- if ($c -c -o "$dummy.o" "$dummy.c") >/dev/null 2>&1 ; then
- CC_FOR_BUILD="$c"; break ;
- fi ;
- done ;
- if test x"$CC_FOR_BUILD" = x ; then
- CC_FOR_BUILD=no_compiler_found ;
- fi
- ;;
- ,,*) CC_FOR_BUILD=$CC ;;
- ,*,*) CC_FOR_BUILD=$HOST_CC ;;
-esac ; set_cc_for_build= ;'
+tmp=
+# shellcheck disable=SC2172
+trap 'test -z "$tmp" || rm -fr "$tmp"' 1 2 13 15
+trap 'exitcode=$?; test -z "$tmp" || rm -fr "$tmp"; exit $exitcode' 0
+
+set_cc_for_build() {
+ : "${TMPDIR=/tmp}"
+ # shellcheck disable=SC2039
+ { tmp=`(umask 077 && mktemp -d "$TMPDIR/cgXXXXXX") 2>/dev/null` && test -n
"$tmp" && test -d "$tmp" ; } ||
+ { test -n "$RANDOM" && tmp=$TMPDIR/cg$$-$RANDOM && (umask 077 && mkdir
"$tmp" 2>/dev/null) ; } ||
+ { tmp=$TMPDIR/cg-$$ && (umask 077 && mkdir "$tmp" 2>/dev/null) && echo
"Warning: creating insecure temp directory" >&2 ; } ||
+ { echo "$me: cannot create a temporary directory in $TMPDIR" >&2 ; exit
1 ; }
+ dummy=$tmp/dummy
+ case ${CC_FOR_BUILD-},${HOST_CC-},${CC-} in
+ ,,) echo "int x;" > "$dummy.c"
+ for driver in cc gcc c89 c99 ; do
+ if ($driver -c -o "$dummy.o" "$dummy.c") >/dev/null 2>&1 ;
then
+ CC_FOR_BUILD="$driver"
+ break
+ fi
+ done
+ if test x"$CC_FOR_BUILD" = x ; then
+ CC_FOR_BUILD=no_compiler_found
+ fi
+ ;;
+ ,,*) CC_FOR_BUILD=$CC ;;
+ ,*,*) CC_FOR_BUILD=$HOST_CC ;;
+ esac
+}
# This is needed to find uname on a Pyramid OSx when run in the BSD universe.
# (address@hidden 1994-08-24)
-if (test -f /.attbin/uname) >/dev/null 2>&1 ; then
+if test -f /.attbin/uname ; then
PATH=$PATH:/.attbin ; export PATH
fi
@@ -138,7 +141,7 @@ Linux|GNU|GNU/*)
# We could probably try harder.
LIBC=gnu
- eval "$set_cc_for_build"
+ set_cc_for_build
cat <<-EOF > "$dummy.c"
#include <features.h>
#if defined(__UCLIBC__)
@@ -199,7 +202,7 @@ case
"$UNAME_MACHINE:$UNAME_SYSTEM:$UNAME_RELEASE:$UNAME_VERSION" in
os=netbsdelf
;;
arm*|i386|m68k|ns32k|sh3*|sparc|vax)
- eval "$set_cc_for_build"
+ set_cc_for_build
if echo __ELF__ | $CC_FOR_BUILD -E - 2>/dev/null \
| grep -q __ELF__
then
@@ -389,20 +392,15 @@ case
"$UNAME_MACHINE:$UNAME_SYSTEM:$UNAME_RELEASE:$UNAME_VERSION" in
echo i386-pc-auroraux"$UNAME_RELEASE"
exit ;;
i86pc:SunOS:5.*:* | i86xen:SunOS:5.*:*)
- eval "$set_cc_for_build"
- SUN_ARCH=i386
- # If there is a compiler, see if it is configured for 64-bit objects.
- # Note that the Sun cc does not turn __LP64__ into 1 like gcc does.
- # This test works for both compilers.
- if [ "$CC_FOR_BUILD" != no_compiler_found ]; then
- if (echo '#ifdef __amd64'; echo IS_64BIT_ARCH; echo '#endif') | \
- (CCOPTS="" $CC_FOR_BUILD -E - 2>/dev/null) | \
- grep IS_64BIT_ARCH >/dev/null
- then
- SUN_ARCH=x86_64
- fi
- fi
- echo "$SUN_ARCH"-pc-solaris2"`echo "$UNAME_RELEASE"|sed -e 's/[^.]*//'`"
+ UNAME_REL="`echo "$UNAME_RELEASE" | sed -e 's/[^.]*//'`"
+ case `isainfo -b` in
+ 32)
+ echo i386-pc-solaris2"$UNAME_REL"
+ ;;
+ 64)
+ echo x86_64-pc-solaris2"$UNAME_REL"
+ ;;
+ esac
exit ;;
sun4*:SunOS:6*:*)
# According to config.sub, this is the proper way to canonicalize
@@ -482,7 +480,7 @@ case
"$UNAME_MACHINE:$UNAME_SYSTEM:$UNAME_RELEASE:$UNAME_VERSION" in
echo clipper-intergraph-clix"$UNAME_RELEASE"
exit ;;
mips:*:*:UMIPS | mips:*:*:RISCos)
- eval "$set_cc_for_build"
+ set_cc_for_build
sed 's/^ //' << EOF > "$dummy.c"
#ifdef __cplusplus
#include <stdio.h> /* for printf() prototype */
@@ -579,7 +577,7 @@ EOF
exit ;;
*:AIX:2:3)
if grep bos325 /usr/include/stdio.h >/dev/null 2>&1; then
- eval "$set_cc_for_build"
+ set_cc_for_build
sed 's/^ //' << EOF > "$dummy.c"
#include <sys/systemcfg.h>
@@ -660,7 +658,7 @@ EOF
esac
fi
if [ "$HP_ARCH" = "" ]; then
- eval "$set_cc_for_build"
+ set_cc_for_build
sed 's/^ //' << EOF > "$dummy.c"
#define _HPUX_SOURCE
@@ -700,7 +698,7 @@ EOF
esac
if [ "$HP_ARCH" = hppa2.0w ]
then
- eval "$set_cc_for_build"
+ set_cc_for_build
# hppa2.0w-hp-hpux* has a 64-bit kernel and a compiler generating
# 32-bit code. hppa64-hp-hpux* has the same kernel and a compiler
@@ -726,7 +724,7 @@ EOF
echo ia64-hp-hpux"$HPUX_REV"
exit ;;
3050*:HI-UX:*:*)
- eval "$set_cc_for_build"
+ set_cc_for_build
sed 's/^ //' << EOF > "$dummy.c"
#include <unistd.h>
int
@@ -840,6 +838,17 @@ EOF
*:BSD/OS:*:*)
echo "$UNAME_MACHINE"-unknown-bsdi"$UNAME_RELEASE"
exit ;;
+ arm*:FreeBSD:*:*)
+ UNAME_PROCESSOR=`uname -p`
+ set_cc_for_build
+ if echo __ARM_PCS_VFP | $CC_FOR_BUILD -E - 2>/dev/null \
+ | grep -q __ARM_PCS_VFP
+ then
+ echo "${UNAME_PROCESSOR}"-unknown-freebsd"`echo
${UNAME_RELEASE}|sed -e 's/[-(].*//'`"-gnueabi
+ else
+ echo "${UNAME_PROCESSOR}"-unknown-freebsd"`echo
${UNAME_RELEASE}|sed -e 's/[-(].*//'`"-gnueabihf
+ fi
+ exit ;;
*:FreeBSD:*:*)
UNAME_PROCESSOR=`/usr/bin/uname -p`
case "$UNAME_PROCESSOR" in
@@ -922,7 +931,7 @@ EOF
echo "$UNAME_MACHINE"-unknown-linux-"$LIBC"
exit ;;
arm*:Linux:*:*)
- eval "$set_cc_for_build"
+ set_cc_for_build
if echo __ARM_EABI__ | $CC_FOR_BUILD -E - 2>/dev/null \
| grep -q __ARM_EABI__
then
@@ -971,7 +980,7 @@ EOF
echo "$UNAME_MACHINE"-unknown-linux-"$LIBC"
exit ;;
mips:Linux:*:* | mips64:Linux:*:*)
- eval "$set_cc_for_build"
+ set_cc_for_build
sed 's/^ //' << EOF > "$dummy.c"
#undef CPU
#undef ${UNAME_MACHINE}
@@ -1285,7 +1294,7 @@ EOF
exit ;;
*:Darwin:*:*)
UNAME_PROCESSOR=`uname -p` || UNAME_PROCESSOR=unknown
- eval "$set_cc_for_build"
+ set_cc_for_build
if test "$UNAME_PROCESSOR" = unknown ; then
UNAME_PROCESSOR=powerpc
fi
@@ -1358,6 +1367,7 @@ EOF
# "uname -m" is not consistent, so use $cputype instead. 386
# is converted to i386 for consistency with other x86
# operating systems.
+ # shellcheck disable=SC2154
if test "$cputype" = 386; then
UNAME_MACHINE=i386
else
diff --git a/build-aux/config.sub b/build-aux/config.sub
index c95acc6..52eb02e 100755
--- a/build-aux/config.sub
+++ b/build-aux/config.sub
@@ -2,7 +2,7 @@
# Configuration validation subroutine script.
# Copyright 1992-2018 Free Software Foundation, Inc.
-timestamp='2018-07-03'
+timestamp='2018-07-25'
# This file is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by
@@ -307,7 +307,7 @@ case $1 in
os=mach
;;
vsta)
- basic_machine=i386-unknown
+ basic_machine=i386-pc
os=vsta
;;
isi68 | isi)
@@ -371,7 +371,7 @@ case $1 in
os=sysv4
;;
netbsd386)
- basic_machine=i386-unknown
+ basic_machine=i386-pc
os=netbsd
;;
netwinder)
@@ -739,6 +739,7 @@ case $basic_machine in
| mipsr5900-* | mipsr5900el-* \
| mipstx39-* | mipstx39el-* \
| mmix-* \
+ | moxie-* \
| mt-* \
| msp430-* \
| nds32-* | nds32le-* | nds32be-* \
@@ -1263,9 +1264,6 @@ case $basic_machine in
pmac | pmac-mpw)
basic_machine=powerpc-apple
;;
- *-unknown)
- # Make sure to match an already-canonicalized machine name.
- ;;
*)
echo Invalid configuration \`"$1"\': machine
\`"$basic_machine"\' not recognized 1>&2
exit 1
diff --git a/configure.ac b/configure.ac
index 2708091..3d0a787 100644
--- a/configure.ac
+++ b/configure.ac
@@ -5483,9 +5483,10 @@ echo
if test "$HAVE_NS" = "yes"; then
echo
- AS_ECHO(["You must run \"${MAKE-make} install\" in order to test the built
application.
-The installed application will go to nextstep/Emacs.app and can be
-run or moved from there."])
+ AS_ECHO(["Run '${MAKE-make}' to build Emacs, then run 'src/emacs' to test
it.
+Run '${MAKE-make} install' in order to build an application bundle.
+The application will go to nextstep/Emacs.app and can be run or moved
+from there."])
if test "$EN_NS_SELF_CONTAINED" = "yes"; then
echo "The application will be fully self-contained."
else
diff --git a/doc/emacs/dired.texi b/doc/emacs/dired.texi
index 007a943..1b03a39 100644
--- a/doc/emacs/dired.texi
+++ b/doc/emacs/dired.texi
@@ -1468,6 +1468,11 @@ rotation is lossless, and uses an external utility called
directory's name, and creates that directory. It signals an error if
the directory already exists.
address@hidden dired-create-empty-file
+ The command (@code{dired-create-empty-file}) reads a
+file name, and creates that file. It signals an error if
+the file already exists.
+
@cindex searching multiple files via Dired
@kindex M-s a C-s @r{(Dired)}
@kindex M-s a M-C-s @r{(Dired)}
diff --git a/doc/emacs/files.texi b/doc/emacs/files.texi
index f0a11fd..a7cc57e 100644
--- a/doc/emacs/files.texi
+++ b/doc/emacs/files.texi
@@ -881,6 +881,8 @@ You can answer ``no'' to bypass copying of this file, this
time. If
you want to cancel the shadowing permanently for a certain file, use
@address@hidden shadow-cancel}} to eliminate or change the shadow file group.
+File Shadowing is not available on MS Windows.
+
@node Time Stamps
@subsection Updating Time Stamps Automatically
@cindex time stamps
diff --git a/doc/emacs/fixit.texi b/doc/emacs/fixit.texi
index ec26a35..8277278 100644
--- a/doc/emacs/fixit.texi
+++ b/doc/emacs/fixit.texi
@@ -427,11 +427,15 @@ dictionary.
@cindex mode, Flyspell
@findex flyspell-mode
Flyspell mode is a minor mode that performs automatic spell-checking
-as you type. When it finds a word that it does not recognize, it
-highlights that word. Type @kbd{M-x flyspell-mode} to toggle Flyspell
-mode in the current buffer. To enable Flyspell mode in all text mode
-buffers, add @code{flyspell-mode} to @code{text-mode-hook}.
address@hidden
+of the text you type as you type it. When it finds a word that it
+does not recognize, it highlights that word. Type @kbd{M-x
+flyspell-mode} to toggle Flyspell mode in the current buffer. To
+enable Flyspell mode in all text mode buffers, add
address@hidden to @code{text-mode-hook}. @xref{Hooks}. Note
+that, as Flyspell mode needs to check each word across which you move,
+it will slow down cursor motion and scrolling commands. It also
+doesn't automatically check the text you didn't type or move across;
+use @code{flyspell-region} or @code{flyspell-buffer} for that.
@findex flyspell-correct-word
@findex flyspell-auto-correct-word
diff --git a/doc/emacs/indent.texi b/doc/emacs/indent.texi
index b38e858..bf43909 100644
--- a/doc/emacs/indent.texi
+++ b/doc/emacs/indent.texi
@@ -60,9 +60,9 @@ repositioned to the first non-whitespace character on the
line.
@node Indentation Commands
@section Indentation Commands
-Apart from the @key{TAB} (@code{indent-for-tab-command}) command,
-Emacs provides a variety of commands to perform indentation in other
-ways.
+Apart from the @address@hidden (@code{indent-for-tab-command})
+command, Emacs provides a variety of commands to perform indentation
+in other ways.
@table @kbd
@item C-M-o
@@ -113,8 +113,8 @@ appears after the newline that is deleted. @xref{Fill
Prefix}.
@item C-M-\
@kindex C-M-\
@findex indent-region
-Indent all the lines in the region, as though you had typed @key{TAB}
-at the beginning of each line (@code{indent-region}).
+Indent all the lines in the region, as though you had typed
address@hidden@key{TAB}} at the beginning of each line (@code{indent-region}).
If a numeric argument is supplied, indent every line in the region to
that column number.
@@ -128,11 +128,12 @@ in the region, moving the affected lines as a rigid unit.
If called with no argument, the command activates a transient mode for
adjusting the indentation of the affected lines interactively. While
-this transient mode is active, typing @key{LEFT} or @key{RIGHT}
-indents leftward and rightward, respectively, by one space. You can
-also type @address@hidden or @address@hidden to indent leftward
-or rightward to the next tab stop (@pxref{Tab Stops}). Typing any
-other key disables the transient mode, and resumes normal editing.
+this transient mode is active, typing @address@hidden or
address@hidden@key{RIGHT}} indents leftward and rightward, respectively, by one
+space. You can also type @address@hidden or @address@hidden to
+indent leftward or rightward to the next tab stop (@pxref{Tab Stops}).
+Typing any other key disables the transient mode, and resumes normal
+editing.
If called with a prefix argument @var{n}, this command indents the
lines forward by @var{n} spaces (without enabling the transient mode).
diff --git a/doc/emacs/maintaining.texi b/doc/emacs/maintaining.texi
index 024fd97..b31cacf 100644
--- a/doc/emacs/maintaining.texi
+++ b/doc/emacs/maintaining.texi
@@ -1655,10 +1655,20 @@ not just to the next change log entry. You can also use
log files into a buffer in Change Log Mode, preserving the date
ordering of entries.
address@hidden add-log-dont-create-changelog-file
Version control systems are another way to keep track of changes in
-your program and keep a change log. In the VC log buffer, typing
address@hidden C-a} (@code{log-edit-insert-changelog}) inserts the relevant
-change log entry, if one exists. @xref{Log Buffer}.
+your program and keep a change log. Many projects that use a VCS don't
+keep a separate versioned change log file nowadays, so you may wish to
+avoid having such a file in the repository. If the value of
address@hidden is address@hidden, commands
+like @kbd{C-x 4 a} (@code{add-change-log-entry-other-window}) will
+record changes in a suitably named temporary buffer instead of a file,
+if such a file does not already exist.
+
+Whether you have a change log file or use a temporary buffer for
+change logs, you can type @kbd{C-c C-a}
+(@code{log-edit-insert-changelog}) in the VC Log buffer to insert the
+relevant change log entries, if they exist. @xref{Log Buffer}.
@node Format of ChangeLog
@subsection Format of ChangeLog
diff --git a/doc/emacs/misc.texi b/doc/emacs/misc.texi
index 3d34414..236cb07 100644
--- a/doc/emacs/misc.texi
+++ b/doc/emacs/misc.texi
@@ -1026,8 +1026,8 @@ Move backward across one shell command, but not beyond
the current line
Ask the shell for its working directory, and update the Shell buffer's
default directory. @xref{Directory Tracking}.
address@hidden M-x send-invisible @key{RET} @var{text} @key{RET}
address@hidden send-invisible
address@hidden M-x comint-send-invisible @key{RET} @var{text} @key{RET}
address@hidden comint-send-invisible
Send @var{text} as input to the shell, after reading it without
echoing. This is useful when a shell command runs a program that asks
for a password.
diff --git a/doc/emacs/mule.texi b/doc/emacs/mule.texi
index 401c83d..6c0c5b2 100644
--- a/doc/emacs/mule.texi
+++ b/doc/emacs/mule.texi
@@ -156,12 +156,19 @@ system encodes the character safely and with a single byte
(@pxref{Coding Systems}). If the character's encoding is longer than
one byte, Emacs shows @samp{file ...}.
- As a special case, if the character lies in the range 128 (0200
-octal) through 159 (0237 octal), it stands for a raw byte that
-does not correspond to any specific displayable character. Such a
-character lies within the @code{eight-bit-control} character set,
-and is displayed as an escaped octal character code. In this case,
address@hidden =} shows @samp{part of display ...} instead of @samp{file}.
address@hidden eight-bit character set
address@hidden raw bytes
+ On rare occasions, Emacs encounters @dfn{raw bytes}: single bytes
+whose values are in the range 128 (0200 octal) through 255 (0377
+octal), which Emacs cannot interpret as part of a known encoding of
+some non-ASCII character. Such raw bytes are treated as if they
+belonged to a special character set @code{eight-bit}; Emacs displays
+them as escaped octal codes (this can be customized; @pxref{Display
+Custom}). In this case, @kbd{C-x =} shows @samp{raw-byte} instead of
address@hidden In addition, @kbd{C-x =} shows the character codes of
+raw bytes as if they were in the range @code{#x3FFF80..#x3FFFFF},
+which is where Emacs maps them to distinguish them from Unicode
+characters in the range @code{#x0080..#x00FF}.
@cindex character set of character at point
@cindex font of character at point
diff --git a/doc/emacs/regs.texi b/doc/emacs/regs.texi
index 7d16d53..98eed06 100644
--- a/doc/emacs/regs.texi
+++ b/doc/emacs/regs.texi
@@ -80,7 +80,9 @@ information until you store something else in it.
@kindex C-x r j
@findex jump-to-register
The command @kbd{C-x r j @var{r}} switches to the buffer recorded in
-register @var{r}, and moves point to the recorded position. The
+register @var{r}, pushes a mark, and moves point to the recorded
+position. (The mark is not pushed if point was already at the
+recorded position, or in successive calls to the command.) The
contents of the register are not changed, so you can jump to the saved
position any number of times.
diff --git a/doc/lispref/debugging.texi b/doc/lispref/debugging.texi
index 1b1f874..9b3ba6c 100644
--- a/doc/lispref/debugging.texi
+++ b/doc/lispref/debugging.texi
@@ -81,7 +81,8 @@ debugger recursively. @xref{Recursive Editing}.
* Function Debugging:: Entering it when a certain function is called.
* Variable Debugging:: Entering it when a variable is modified.
* Explicit Debug:: Entering it at a certain point in the program.
-* Using Debugger:: What the debugger does; what you see while in it.
+* Using Debugger:: What the debugger does.
+* Backtraces:: What you see while in the debugger.
* Debugger Commands:: Commands used while in the debugger.
* Invoking the Debugger:: How to call the function @code{debug}.
* Internals of Debugger:: Subroutines of the debugger, and global variables.
@@ -392,32 +393,82 @@ this is not what you want, you can either set
@code{eval-expression-debug-on-error} to @code{nil}, or set
@code{debug-on-error} to @code{nil} in @code{debugger-mode-hook}.
+ The debugger itself must be run byte-compiled, since it makes
+assumptions about the state of the Lisp interpreter. These
+assumptions are false if the debugger is running interpreted.
+
address@hidden Backtraces
address@hidden Backtraces
address@hidden backtrace buffer
+
+Debugger mode is derived from Backtrace mode, which is also used to
+show backtraces by Edebug and ERT. (@pxref{Edebug}, and @ref{Top,the
+ERT manual,, ert, ERT: Emacs Lisp Regression Testing}.)
+
address@hidden stack frame
+The backtrace buffer shows you the functions that are executing and
+their argument values. When a backtrace buffer is created, it shows
+each stack frame on one, possibly very long, line. (A stack frame is
+the place where the Lisp interpreter records information about a
+particular invocation of a function.) The most recently called
+function will be at the top.
+
@cindex current stack frame
- The backtrace buffer shows you the functions that are executing and
-their argument values. It also allows you to specify a stack frame by
-moving point to the line describing that frame. (A stack frame is the
-place where the Lisp interpreter records information about a particular
-invocation of a function.) The frame whose line point is on is
-considered the @dfn{current frame}. Some of the debugger commands
-operate on the current frame. If a line starts with a star, that means
-that exiting that frame will call the debugger again. This is useful
-for examining the return value of a function.
-
- If a function name is underlined, that means the debugger knows
-where its source code is located. You can click with the mouse on
-that name, or move to it and type @key{RET}, to visit the source code.
+In a backtrace you can specify a stack frame by moving point to a line
+describing that frame. The frame whose line point is on is considered
+the @dfn{current frame}.
+
+If a function name is underlined, that means Emacs knows where its
+source code is located. You can click with the mouse on that name, or
+move to it and type @key{RET}, to visit the source code. You can also
+type @key{RET} while point is on any name of a function or variable
+which is not underlined, to see help information for that symbol in a
+help buffer, if any exists. The @code{xref-find-definitions} command,
+bound to @key{M-.}, can also be used on any identifier in a backtrace
+(@pxref{Looking Up Identifiers,,,emacs, The GNU Emacs Manual}).
+
+In backtraces, the tails of long lists and the ends of long strings,
+vectors or structures, as well as objects which are deeply nested,
+will be printed as underlined ``...''. You can click with the mouse
+on a ``...'', or type @key{RET} while point is on it, to show the part
+of the object that was hidden. To control how much abbreviation is
+done, customize @code{backtrace-line-length}.
+
+Here is a list of commands for navigating and viewing backtraces:
- The debugger itself must be run byte-compiled, since it makes
-assumptions about how many stack frames are used for the debugger
-itself. These assumptions are false if the debugger is running
-interpreted.
address@hidden @kbd
address@hidden v
+Toggle the display of local variables of the current stack frame.
+
address@hidden p
+Move to the beginning of the frame, or to the beginning
+of the previous frame.
+
address@hidden n
+Move to the beginning of the next frame.
+
address@hidden +
+Add line breaks and indentation to the top-level Lisp form at point to
+make it more readable.
+
address@hidden -
+Collapse the top-level Lisp form at point back to a single line.
+
address@hidden #
+Toggle @code{print-circle} for the frame at point.
+
address@hidden .
+Expand all the forms abbreviated with ``...'' in the frame at point.
+
address@hidden table
@node Debugger Commands
@subsection Debugger Commands
@cindex debugger command list
The debugger buffer (in Debugger mode) provides special commands in
-addition to the usual Emacs commands. The most important use of
+addition to the usual Emacs commands and to the Backtrace mode commands
+described in the previous section. The most important use of
debugger commands is for stepping through code, so that you can see
how control flows. The debugger can step through the control
structures of an interpreted function, but cannot do so in a
@@ -427,6 +478,11 @@ the same function. (To do this, visit the source for the
function and
type @kbd{C-M-x} on its definition.) You cannot use the Lisp debugger
to step through a primitive function.
+Some of the debugger commands operate on the current frame. If a
+frame starts with a star, that means that exiting that frame will call the
+debugger again. This is useful for examining the return value of a
+function.
+
@c FIXME: Add @findex for the following commands? --xfq
Here is a list of Debugger mode commands:
@@ -502,8 +558,6 @@ Display a list of functions that will invoke the debugger
when called.
This is a list of functions that are set to break on entry by means of
@code{debug-on-entry}.
address@hidden v
-Toggle the display of local variables of the current stack frame.
@end table
@node Invoking the Debugger
@@ -624,20 +678,19 @@ of @code{debug} (@pxref{Invoking the Debugger}).
@cindex run time stack
@cindex call stack
This function prints a trace of Lisp function calls currently active.
-This is the function used by @code{debug} to fill up the
address@hidden buffer. It is written in C, since it must have access
-to the stack to determine which function calls are active. The return
-value is always @code{nil}.
+The trace is identical to the one that @code{debug} would show in the
address@hidden buffer. The return value is always nil.
In the following example, a Lisp expression calls @code{backtrace}
explicitly. This prints the backtrace to the stream
@code{standard-output}, which, in this case, is the buffer
@samp{backtrace-output}.
-Each line of the backtrace represents one function call. The line shows
-the values of the function's arguments if they are all known; if they
-are still being computed, the line says so. The arguments of special
-forms are elided.
+Each line of the backtrace represents one function call. The line
+shows the function followed by a list of the values of the function's
+arguments if they are all known; if they are still being computed, the
+line consists of a list containing the function and its unevaluated
+arguments. Long lists or deeply nested structures may be elided.
@smallexample
@group
@@ -654,7 +707,7 @@ forms are elided.
@group
----------- Buffer: backtrace-output ------------
backtrace()
- (list ...computing arguments...)
+ (list 'testing (backtrace))
@end group
(progn ...)
eval((progn (1+ var) (list 'testing (backtrace))))
@@ -685,7 +738,7 @@ example would look as follows:
@group
----------- Buffer: backtrace-output ------------
(backtrace)
- (list ...computing arguments...)
+ (list 'testing (backtrace))
@end group
(progn ...)
(eval (progn (1+ var) (list 'testing (backtrace))))
diff --git a/doc/lispref/edebug.texi b/doc/lispref/edebug.texi
index b9cc1d5..54200b9 100644
--- a/doc/lispref/edebug.texi
+++ b/doc/lispref/edebug.texi
@@ -442,8 +442,16 @@ Redisplay the most recently known expression result in the
echo area
Display a backtrace, excluding Edebug's own functions for clarity
(@code{edebug-backtrace}).
-You cannot use debugger commands in the backtrace buffer in Edebug as
-you would in the standard debugger.
address@hidden, for a description of backtraces
+and the commands which work on them.
+
+If you would like to see Edebug's functions in the backtrace,
+use @kbd{M-x edebug-backtrace-show-instrumentation}. To hide them
+again use @kbd{M-x edebug-backtrace-hide-instrumentation}.
+
+If a backtrace frame starts with @samp{>} that means that Edebug knows
+where the source code for the frame is located. Use @kbd{s} to jump
+to the source code for the current frame.
The backtrace buffer is killed automatically when you continue
execution.
diff --git a/doc/lispref/eval.texi b/doc/lispref/eval.texi
index 4e8b0df..c9401be 100644
--- a/doc/lispref/eval.texi
+++ b/doc/lispref/eval.texi
@@ -507,9 +507,6 @@ Emacs Lisp with a reference to where each is described.
@item setq-default
@pxref{Creating Buffer-Local}
address@hidden track-mouse
address@hidden Tracking}
-
@item unwind-protect
@pxref{Nonlocal Exits}
diff --git a/doc/lispref/files.texi b/doc/lispref/files.texi
index 068cf05..25fabe1 100644
--- a/doc/lispref/files.texi
+++ b/doc/lispref/files.texi
@@ -3005,10 +3005,16 @@ This command creates a directory named @var{dirname}.
If
@var{parents} is address@hidden, as is always the case in an
interactive call, that means to create the parent directories first,
if they don't already exist.
-
@code{mkdir} is an alias for this.
@end deffn
address@hidden Command make-empty-file filename &optional parents
+This command creates an empty file named @var{filename}.
+As @code{make-directory}, this command creates parent directories
+if @var{parents} is address@hidden
+If @var{filename} already exists, this command signals an error.
address@hidden deffn
+
@deffn Command copy-directory dirname newname &optional keep-time parents
copy-contents
This command copies the directory named @var{dirname} to
@var{newname}. If @var{newname} is a directory name,
diff --git a/doc/lispref/frames.texi b/doc/lispref/frames.texi
index 6678644..ba4b931 100644
--- a/doc/lispref/frames.texi
+++ b/doc/lispref/frames.texi
@@ -3373,10 +3373,10 @@ occur. That is useful, because normally you don't want
to track the
mouse forever---only until some other event, such as the release of a
button.
address@hidden track-mouse address@hidden
-This special form executes @var{body}, with generation of mouse motion
-events enabled. Typically, @var{body} would use @code{read-event} to
-read the motion events and modify the display accordingly. @xref{Motion
address@hidden track-mouse address@hidden
+This macro executes @var{body}, with generation of mouse motion events
+enabled. Typically, @var{body} would use @code{read-event} to read
+the motion events and modify the display accordingly. @xref{Motion
Events}, for the format of mouse motion events.
The value of @code{track-mouse} is that of the last form in @var{body}.
@@ -3396,7 +3396,7 @@ on (@pxref{Pointer Shape}). Therefore, Lisp programs
that need the
mouse pointer to retain its original shape during dragging should bind
@code{track-mouse} to the value @code{dragging} at the beginning of
their @var{body}.
address@hidden defspec
address@hidden defmac
The usual purpose of tracking mouse motion is to indicate on the screen
the consequences of pushing or releasing a button at the current
diff --git a/doc/lispref/lists.texi b/doc/lispref/lists.texi
index 57cefea..b7bb3cf 100644
--- a/doc/lispref/lists.texi
+++ b/doc/lispref/lists.texi
@@ -50,16 +50,19 @@ convention; at the level of cons cells, the @sc{car} and
@sc{cdr}
slots have similar properties). Hence, the @sc{cdr} slot of each cons
cell in a list refers to the following cons cell.
address@hidden proper list
@cindex true list
Also by convention, the @sc{cdr} of the last cons cell in a list is
@code{nil}. We call such a @code{nil}-terminated structure a
address@hidden list}. In Emacs Lisp, the symbol @code{nil} is both a
-symbol and a list with no elements. For convenience, the symbol
address@hidden is considered to have @code{nil} as its @sc{cdr} (and also
-as its @sc{car}).
-
- Hence, the @sc{cdr} of a true list is always a true list. The
address@hidden of a nonempty true list is a true list containing all the
address@hidden address@hidden is sometimes also referred to as a
address@hidden list}, but we generally do not use this terminology in this
+manual.}. In Emacs Lisp, the symbol @code{nil} is both a symbol and a
+list with no elements. For convenience, the symbol @code{nil} is
+considered to have @code{nil} as its @sc{cdr} (and also as its
address@hidden).
+
+ Hence, the @sc{cdr} of a proper list is always a proper list. The
address@hidden of a nonempty proper list is a proper list containing all the
elements except the first.
@cindex dotted list
@@ -71,10 +74,10 @@ Pair Notation}). There is one other possibility: some cons
cell's
@sc{cdr} could point to one of the previous cons cells in the list.
We call that structure a @dfn{circular list}.
- For some purposes, it does not matter whether a list is true,
+ For some purposes, it does not matter whether a list is proper,
circular or dotted. If a program doesn't look far enough down the
list to see the @sc{cdr} of the final cons cell, it won't care.
-However, some functions that operate on lists demand true lists and
+However, some functions that operate on lists demand proper lists and
signal errors if given a dotted list. Most functions that try to find
the end of a list enter infinite loops if given a circular list.
@@ -538,7 +541,7 @@ object. The final argument is not copied or converted; it
becomes the
is itself a list, then its elements become in effect elements of the
result list. If the final element is not a list, the result is a
dotted list since its final @sc{cdr} is not @code{nil} as required
-in a true list.
+in a proper list (@pxref{Cons Cells}).
@end defun
Here is an example of using @code{append}:
diff --git a/doc/lispref/minibuf.texi b/doc/lispref/minibuf.texi
index 889b64a..d091787 100644
--- a/doc/lispref/minibuf.texi
+++ b/doc/lispref/minibuf.texi
@@ -2199,7 +2199,7 @@ function @code{read-passwd}.
@defun read-passwd prompt &optional confirm default
This function reads a password, prompting with @var{prompt}. It does
not echo the password as the user types it; instead, it echoes
address@hidden for each character in the password. If you want to apply
address@hidden for each character in the password. If you want to apply
another character to hide the password, let-bind the variable
@code{read-hide-char} with that character.
diff --git a/doc/lispref/numbers.texi b/doc/lispref/numbers.texi
index d9fb432..89205f9 100644
--- a/doc/lispref/numbers.texi
+++ b/doc/lispref/numbers.texi
@@ -213,7 +213,7 @@ least one digit after any decimal point in a floating-point
number;
@samp{1500.} is an integer, not a floating-point number.
Emacs Lisp treats @code{-0.0} as numerically equal to ordinary zero
-with respect to @code{equal} and @code{=}. This follows the
+with respect to numeric comparisons like @code{=}. This follows the
@acronym{IEEE} floating-point standard, which says @code{-0.0} and
@code{0.0} are numerically equal even though other operations can
distinguish them.
@@ -227,8 +227,20 @@ infinity and negative infinity as floating-point values.
It also
provides for a class of values called NaN, or ``not a number'';
numerical functions return such values in cases where there is no
correct answer. For example, @code{(/ 0.0 0.0)} returns a address@hidden
-Although NaN values carry a sign, for practical purposes there is no other
-significant difference between different NaN values in Emacs Lisp.
+A NaN is never numerically equal to any value, not even to itself.
+NaNs carry a sign and a significand, and non-numeric functions treat
+two NaNs as equal when their
+signs and significands agree. Significands of NaNs are
+machine-dependent, as are the digits in their string representation.
+
+ When NaNs and signed zeros are involved, non-numeric functions like
address@hidden, @code{equal}, @code{sxhash-eql}, @code{sxhash-equal} and
address@hidden determine whether values are indistinguishable, not
+whether they are numerically equal. For example, when @var{x} and
address@hidden are the same NaN, @code{(equal x y)} returns @code{t} whereas
address@hidden(= x y)} uses numeric comparison and returns @code{nil};
+conversely, @code{(equal 0.0 -0.0)} returns @code{nil} whereas
address@hidden(= 0.0 -0.0)} returns @code{t}.
Here are read syntaxes for these special floating-point values:
@@ -358,11 +370,15 @@ if so, @code{nil} otherwise. The argument must be a
number.
@cindex comparing numbers
To test numbers for numerical equality, you should normally use
address@hidden, not @code{eq}. There can be many distinct floating-point
-and large integer objects with the same numeric value. If you use
address@hidden to compare them, then you test whether two values are the
-same @emph{object}. By contrast, @code{=} compares only the numeric
-values of the objects.
address@hidden instead of non-numeric comparison predicates like @code{eq},
address@hidden and @code{equal}. Distinct floating-point and large
+integer objects can be numerically equal. If you use @code{eq} to
+compare them, you test whether they are the same @emph{object}; if you
+use @code{eql} or @code{equal}, you test whether their values are
address@hidden In contrast, @code{=} uses numeric
+comparison, and sometimes returns @code{t} when a non-numeric
+comparison would return @code{nil} and vice versa. @xref{Float
+Basics}.
In Emacs Lisp, each small integer is a unique Lisp object.
Therefore, @code{eq} is equivalent to @code{=} where small integers are
@@ -829,7 +845,7 @@ reproducing the same pattern moved over.
bits in @var{integer1} to the left @var{count} places, or to the right
if @var{count} is negative, bringing zeros into the vacated bits. If
@var{count} is negative, @code{lsh} shifts zeros into the leftmost
-(most-significant) bit, producing a positive result even if
+(most-significant) bit, producing a nonnegative result even if
@var{integer1} is negative. Contrast this with @code{ash}, below.
Here are two examples of @code{lsh}, shifting a pattern of bits one
@@ -1167,7 +1183,7 @@ returns a NaN.
@defun expt x y
This function returns @var{x} raised to power @var{y}. If both
-arguments are integers and @var{y} is positive, the result is an
+arguments are integers and @var{y} is nonnegative, the result is an
integer; in this case, overflow causes truncation, so watch out.
If @var{x} is a finite negative number and @var{y} is a finite
non-integer, @code{expt} returns a NaN.
diff --git a/doc/lispref/sequences.texi b/doc/lispref/sequences.texi
index b98889e..6a6f4d5 100644
--- a/doc/lispref/sequences.texi
+++ b/doc/lispref/sequences.texi
@@ -63,7 +63,8 @@ But it is possible to add elements to the list, or remove
elements.
@defun sequencep object
This function returns @code{t} if @var{object} is a list, vector,
-string, bool-vector, or char-table, @code{nil} otherwise.
+string, bool-vector, or char-table, @code{nil} otherwise. See also
address@hidden below.
@end defun
@defun length sequence
@@ -479,7 +480,8 @@ built-in sequence types, @code{seq-length} behaves like
@code{length}.
@defun seqp object
This function returns address@hidden if @var{object} is a sequence
(a list or array), or any additional type of sequence defined via
address@hidden generic functions.
address@hidden generic functions. This is an extensible variant of
address@hidden
@example
@group
@@ -1355,7 +1357,7 @@ each initialized to @var{object}.
@defun vconcat &rest sequences
@cindex copying vectors
This function returns a new vector containing all the elements of
address@hidden The arguments @var{sequences} may be true lists,
address@hidden The arguments @var{sequences} may be proper lists,
vectors, strings or bool-vectors. If no @var{sequences} are given,
the empty vector is returned.
diff --git a/doc/lispref/strings.texi b/doc/lispref/strings.texi
index 026ba74..3558f17 100644
--- a/doc/lispref/strings.texi
+++ b/doc/lispref/strings.texi
@@ -922,7 +922,8 @@ Functions}). Thus, strings are enclosed in @samp{"}
characters, and
@item %o
@cindex integer to octal
Replace the specification with the base-eight representation of an
-unsigned integer. The object can also be a nonnegative floating-point
+integer. Negative integers are formatted in a platform-dependent
+way. The object can also be a nonnegative floating-point
number that is formatted as an integer, dropping any fraction, if the
integer does not exceed machine limits.
@@ -935,7 +936,8 @@ formatted as an integer, dropping any fraction.
@itemx %X
@cindex integer to hexadecimal
Replace the specification with the base-sixteen representation of an
-unsigned integer. @samp{%x} uses lower case and @samp{%X} uses upper
+integer. Negative integers are formatted in a platform-dependent
+way. @samp{%x} uses lower case and @samp{%X} uses upper
case. The object can also be a nonnegative floating-point number that
is formatted as an integer, dropping any fraction, if the integer does
not exceed machine limits.
@@ -1015,17 +1017,17 @@ numbered or unnumbered format specifications but not
both, except that
After the @samp{%} and any field number, you can put certain
@dfn{flag characters}.
- The flag @samp{+} inserts a plus sign before a positive number, so
+ The flag @samp{+} inserts a plus sign before a nonnegative number, so
that it always has a sign. A space character as flag inserts a space
-before a positive number. (Otherwise, positive numbers start with the
-first digit.) These flags are useful for ensuring that positive
-numbers and negative numbers use the same number of columns. They are
+before a nonnegative number. (Otherwise, nonnegative numbers start with the
+first digit.) These flags are useful for ensuring that nonnegative
+and negative numbers use the same number of columns. They are
ignored except for @samp{%d}, @samp{%e}, @samp{%f}, @samp{%g}, and if
both flags are used, @samp{+} takes precedence.
The flag @samp{#} specifies an alternate form which depends on
the format in use. For @samp{%o}, it ensures that the result begins
-with a @samp{0}. For @samp{%x} and @samp{%X}, it prefixes the result
+with a @samp{0}. For @samp{%x} and @samp{%X}, it prefixes nonzero results
with @samp{0x} or @samp{0X}. For @samp{%e} and @samp{%f}, the
@samp{#} flag means include a decimal point even if the precision is
zero. For @samp{%g}, it always includes a decimal point, and also
@@ -1108,6 +1110,17 @@ shows only the first three characters of the
representation for
precision is what the local library functions of the @code{printf}
family produce.
address@hidden formatting numbers for rereading later
+ If you plan to use @code{read} later on the formatted string to
+retrieve a copy of the formatted value, use a specification that lets
address@hidden reconstruct the value. To format numbers in this
+reversible way you can use @samp{%s} and @samp{%S}, to format just
+integers you can also use @samp{%d}, and to format just nonnegative
+integers you can also use @samp{#x%x} and @samp{#o%o}. Other formats
+may be problematic; for example, @samp{%d} and @samp{%g} can mishandle
+NaNs and can lose precision and type, and @samp{#x%x} and @samp{#o%o}
+can mishandle negative integers. @xref{Input Functions}.
+
@node Case Conversion
@section Case Conversion in Lisp
@cindex upper case
diff --git a/doc/lispref/threads.texi b/doc/lispref/threads.texi
index f05af49..58a3a91 100644
--- a/doc/lispref/threads.texi
+++ b/doc/lispref/threads.texi
@@ -75,8 +75,8 @@ thread, @code{nil} otherwise.
@defun thread-join thread
Block until @var{thread} exits, or until the current thread is
-signaled. If @var{thread} has already exited, this returns
-immediately.
+signaled. It returns the result of the @var{thread} function. If
address@hidden has already exited, this returns immediately.
@end defun
@defun thread-signal thread error-symbol data
@@ -87,6 +87,15 @@ thread, then this just calls @code{signal} immediately.
Otherwise,
If @var{thread} was blocked by a call to @code{mutex-lock},
@code{condition-wait}, or @code{thread-join}; @code{thread-signal}
will unblock it.
+
+Since signal handlers in Emacs are located in the main thread, a
+signal must be propagated there in order to become visible. The
+second @code{signal} call let the thread die:
+
address@hidden
+(thread-signal main-thread 'error data)
+(signal 'error data)
address@hidden example
@end defun
@defun thread-yield
@@ -127,15 +136,21 @@ Return a list of all the live thread objects. A new list
is returned
by each invocation.
@end defun
address@hidden main-thread
+This variable keeps the main thread Emacs is running, or @code{nil} if
+Emacs is compiled without thread support.
address@hidden defvar
+
When code run by a thread signals an error that is unhandled, the
thread exits. Other threads can access the error form which caused
the thread to exit using the following function.
address@hidden thread-last-error
address@hidden thread-last-error &optional cleanup
This function returns the last error form recorded when a thread
exited due to an error. Each thread that exits abnormally overwrites
the form stored by the previous thread's error with a new value, so
-only the last one can be accessed.
+only the last one can be accessed. If @var{cleanup} is
address@hidden, the stored form is reset to @code{nil}.
@end defun
@node Mutexes
diff --git a/doc/misc/efaq.texi b/doc/misc/efaq.texi
index 903c56c..5b432d5 100644
--- a/doc/misc/efaq.texi
+++ b/doc/misc/efaq.texi
@@ -2988,7 +2988,7 @@ Emacs compiled on a 64-bit machine can handle much larger
buffers.
@cindex Shell buffer, echoed commands and @samp{^M} in
@cindex Echoed commands in @code{shell-mode}
-Try typing @kbd{M-x shell-strip-ctrl-m @key{RET}} while in @code{shell-mode} to
+Try typing @kbd{M-x comint-strip-ctrl-m @key{RET}} while in @code{shell-mode}
to
make them go away. If that doesn't work, you have several options:
For @code{tcsh}, put this in your @file{.cshrc} (or @file{.tcshrc})
@@ -3041,7 +3041,7 @@ characters from the buffer by adding this to your
@file{.emacs} init
file:
@smalllisp
-(add-hook 'comint-output-filter-functions 'shell-strip-ctrl-m)
+(add-hook 'comint-output-filter-functions #'comint-strip-ctrl-m)
@end smalllisp
On a related note: if your shell is echoing your input line in the shell
diff --git a/doc/misc/ert.texi b/doc/misc/ert.texi
index 82e0e27..6a34f5c 100644
--- a/doc/misc/ert.texi
+++ b/doc/misc/ert.texi
@@ -273,9 +273,11 @@ moving point to it and typing @address@hidden jumps to its
definition.
@cindex backtrace of a failed test
Pressing @kbd{r} re-runs the test near point on its own. Pressing
@kbd{d} re-runs it with the debugger enabled. @kbd{.} jumps to the
-definition of the test near point (@address@hidden has the same effect if
-point is on the name of the test). On a failed test, @kbd{b} shows
-the backtrace of the failure.
+definition of the test near point (@address@hidden has the same effect
+if point is on the name of the test). On a failed test, @kbd{b} shows
+the backtrace of the failure. @xref{Debugging,, Backtraces, elisp,
+GNU Emacs Lisp Reference Manual}, for more information about
+backtraces.
@kindex address@hidden, in ert results buffer}
@kbd{l} shows the list of @code{should} forms executed in the test.
diff --git a/doc/misc/flymake.texi b/doc/misc/flymake.texi
index bdefd40..bda7e14 100644
--- a/doc/misc/flymake.texi
+++ b/doc/misc/flymake.texi
@@ -4,7 +4,7 @@
@set VERSION 1.0
@set UPDATED June 2018
@settitle GNU Flymake @value{VERSION}
address@hidden ../emacs/docstyle.texi
address@hidden docstyle.texi
@syncodeindex pg cp
@syncodeindex vr cp
@syncodeindex fn cp
diff --git a/doc/misc/tramp.texi b/doc/misc/tramp.texi
index d8f4b41..55c21b7 100644
--- a/doc/misc/tramp.texi
+++ b/doc/misc/tramp.texi
@@ -544,9 +544,9 @@ file system), @address@hidden,user@@host,/path/to/file}} and
@cindex method @option{gdrive}
@cindex @option{gdrive} method
@cindex google drive
address@hidden method @option{owncloud}
address@hidden @option{owncloud} method
address@hidden nextcloud
address@hidden method @option{nextcloud}
address@hidden @option{nextcloud} method
address@hidden owncloud
GVFS-based methods include also @acronym{GNOME} Online Accounts, which
support the @option{Files} service. These are the Google Drive file
@@ -554,7 +554,7 @@ system, and the OwnCloud/NextCloud file system. The file
name syntax
is here always
@address@hidden,john.doe@@gmail.com,/path/to/file}}
(@samp{john.doe@@gmail.com} stands here for your Google Drive
-account), or @address@hidden,user@@host#8081,/path/to/file}}
+account), or @address@hidden,user@@host#8081,/path/to/file}}
(@samp{8081} stands for the port number) for OwnCloud/NextCloud files.
@@ -1094,6 +1094,10 @@ syntax requires a leading volume (share) name, for
example:
based on standard protocols, such as address@hidden @option{davs} does the
same
but with SSL encryption. Both methods support the port numbers.
+Paths being part of the WebDAV volume to be mounted by GVFS, as it is
+common for OwnCloud or NextCloud file names, are not supported by
+these methods. See method @option{nextcloud} for handling them.
+
@item @option{gdrive}
@cindex method @option{gdrive}
@cindex @option{gdrive} method
@@ -1110,13 +1114,13 @@ Since Google Drive uses cryptic blob file names
internally,
could produce unexpected behavior in case two files in the same
directory have the same @code{display-name}, such a situation must be avoided.
address@hidden @option{owncloud}
address@hidden @option{nextcloud}
@cindex @acronym{GNOME} Online Accounts
address@hidden method @option{owncloud}
address@hidden @option{owncloud} method
address@hidden nextcloud
address@hidden method @option{nextcloud}
address@hidden @option{nextcloud} method
address@hidden owncloud
-As the name indicates, the method @option{owncloud} allows you to
+As the name indicates, the method @option{nextcloud} allows you to
access OwnCloud or NextCloud hosted files and directories. Like the
@option{gdrive} method, your credentials must be populated in your
@command{Online Accounts} application outside Emacs. The method
@@ -1135,7 +1139,7 @@ that for security reasons refuse @command{ssh}
connections.
@defopt tramp-gvfs-methods
This user option is a list of external methods for address@hidden By default,
this list includes @option{afp}, @option{dav}, @option{davs},
address@hidden, @option{owncloud} and @option{sftp}. Other methods
address@hidden, @option{nextcloud} and @option{sftp}. Other methods
to include are @option{ftp}, @option{http}, @option{https} and
@option{smb}. These methods are not intended to be used directly as
GVFS based method. Instead, they are added here for the benefit of
@@ -1238,7 +1242,7 @@ improvement is not always true.
@cindex default user
@defopt tramp-default-user
address@hidden file name can omit the user name part since
+A @value{tramp} file name can omit the user name part since
@value{tramp} substitutes the currently logged-in user name. However
this substitution can be overridden with @code{tramp-default-user}.
For example:
@@ -1453,7 +1457,7 @@ support this command.
@subsection Tunneling with ssh
-With ssh, you could use the @code{ProxyCommand} entry in the
+With ssh, you could use the @code{ProxyCommand} entry in
@file{~/.ssh/config}:
@example
@@ -1589,12 +1593,12 @@ A function dedicated to @file{/etc/hosts} for host
names.
@item @code{tramp-parse-passwd}
@findex tramp-parse-passwd
-A function which parses @file{/etc/passwd} files for user names.
+A function which parses @file{/etc/passwd} for user names.
@item @code{tramp-parse-etc-group}
@findex tramp-parse-etc-group
-A function which parses @file{/etc/group} files for group names.
+A function which parses @file{/etc/group} for group names.
@item @code{tramp-parse-netrc}
@findex tramp-parse-netrc
@@ -2194,7 +2198,7 @@ of the secretfile is now owned by the user logged in from
When @code{backup-directory-alist} is @code{nil} (the default), such
problems do not occur.
-To ``turn off'' the backup feature for @value{tramp} files and stop
+To ``turn off'' the backup feature for remote files and stop
@value{tramp} from saving to the backup directory, use this:
@lisp
@@ -2256,12 +2260,11 @@ The backup file name of
@vindex auto-save-file-name-transforms
Just as for backup files, similar issues of file naming affect
-auto-saving @value{tramp} files. Auto-saved files are saved in the
-directory specified by the user option
address@hidden By default this is set to
-the local temporary directory. But in some versions of Debian
-GNU/Linux, this points to the source directory where the Emacs was
-compiled. Reset such values to a valid directory.
+auto-saving remote files. Auto-saved files are saved in the directory
+specified by the user option @code{auto-save-file-name-transforms}.
+By default this is set to the local temporary directory. But in some
+versions of Debian GNU/Linux, this points to the source directory
+where the Emacs was compiled. Reset such values to a valid directory.
Set @code{auto-save-file-name-transforms} to @code{nil} to save
auto-saved files to the same directory as the original file.
@@ -2765,8 +2768,8 @@ hard-coded, fixed name. Note that using @code{:0} for
X11 display name
here will not work as expected.
An alternate approach is specify @code{ForwardX11 yes} or
address@hidden yes} in the file @file{~/.ssh/config} on the
-local host.
address@hidden yes} in @file{~/.ssh/config} on the local
+host.
@subsection Running @code{shell} on a remote host
@@ -3446,6 +3449,19 @@ source "address@hidden@}/.iterm2_shell_integration.bash"
@end group
@end example
+And finally, bash's readline should not use key bindings like
address@hidden to commands. Disable this in your @file{~/.inputrc}:
+
address@hidden
address@hidden
+$if term=dumb
+# Don't bind Control-J or it messes up @value{tramp}.
+$else
+"\C-j": next-history
+$endif
address@hidden group
address@hidden example
+
@item
Echoed characters after login
@@ -3582,13 +3598,13 @@ When testing, ensure the remote shell is the same shell
How to get notified after @value{tramp} completes file transfers?
Make Emacs beep after reading from or writing to the remote host with
-the following code in @file{~/.emacs} file.
+the following code in @file{~/.emacs}.
@lisp
@group
(defadvice tramp-handle-write-region
(after tramp-write-beep-advice activate)
- "Make tramp beep after writing a file."
+ "Make @value{tramp} beep after writing a file."
(interactive)
(beep))
@end group
@@ -3596,7 +3612,7 @@ the following code in @file{~/.emacs} file.
@group
(defadvice tramp-handle-do-copy-or-rename-file
(after tramp-copy-beep-advice activate)
- "Make tramp beep after copying a file."
+ "Make @value{tramp} beep after copying a file."
(interactive)
(beep))
@end group
@@ -3604,7 +3620,7 @@ the following code in @file{~/.emacs} file.
@group
(defadvice tramp-handle-insert-file-contents
(after tramp-insert-beep-advice activate)
- "Make tramp beep after inserting a file."
+ "Make @value{tramp} beep after inserting a file."
(interactive)
(beep))
@end group
@@ -3642,7 +3658,7 @@ then set them with a hook as follows:
@item
-Why is @file{~/.sh_history} file on the remote host growing?
+Why is @file{~/.sh_history} on the remote host growing?
@vindex tramp-histfile-override
@vindex address@hidden, environment variable}
@@ -3663,7 +3679,7 @@ undesired results when using @command{bash} as remote
shell.
Another approach is to disable @value{tramp}'s handling of the
@env{HISTFILE} at all by setting @code{tramp-histfile-override} to
@code{nil}. In this case, saving history could be turned off by
-putting this shell code in the @file{.bashrc} or @file{.kshrc} file:
+putting this shell code in @file{.bashrc} or @file{.kshrc}:
@example
@group
@@ -3680,7 +3696,7 @@ fi
@end example
For @option{ssh}-based method, add the following line to your
address@hidden/.ssh/environment} file:
address@hidden/.ssh/environment}:
@example
HISTFILE=/dev/null
diff --git a/etc/NEWS b/etc/NEWS
index 2be4fe9..081325a 100644
--- a/etc/NEWS
+++ b/etc/NEWS
@@ -34,6 +34,13 @@ functions 'json-serialize', 'json-insert',
'json-parse-string', and
'json-parse-buffer' are typically much faster than their Lisp
counterparts from json.el.
+** The etags program now uses the C library's regular expression matcher
+when possible, and a compatible regex substitute otherwise. This will
+let developers maintain Emacs's own regex code without having to also
+support other programs. The new configure option '--without-included-regex'
+forces etags to use the C library's regex matcher even if the regex
+substitute ordinarily would be used to work around compatibility problems.
+
** Emacs has been ported to the -fcheck-pointer-bounds option of GCC.
This causes Emacs to check bounds of some arrays addressed by its
internal pointers, which can be helpful when debugging the Emacs
@@ -84,14 +91,24 @@ work right without some adjustment:
- you can use the new 'package-quickstart' so activation of packages does not
need to pay attention to 'package-load-list' or 'package-user-dir' any more.
+---
+** Emacs now notifies systemd when startup finishes or shutdown begins.
+Units that are ordered after 'emacs.service' will only be started
+after Emacs has finished initialization and is ready for use.
+(If your Emacs is installed in a non-standard location and you copied the
+emacs.service file to eg ~/.config/systemd/user/, you will need to copy
+the new version of the file again.)
+
* Changes in Emacs 27.1
++++
+** The function 'read-passwd' uses '*' as default character to hide passwords.
+
---
** New variable 'xft-ignore-color-fonts'.
Default t means don't try to load color fonts when using Xft, as they
often cause crashes. Set it to nil if you really need those fonts.
-(Bug#30874)
---
** The new option 'tooltip-resize-echo-area' avoids truncating tooltip text
@@ -178,6 +195,9 @@ This triggers to search the program on the remote host as
indicated by
* Editing Changes in Emacs 27.1
++++
+** New command 'make-empty-file'.
+
---
** New variable 'x-wait-for-event-timeout'.
This controls how long Emacs will wait for updates to the graphical
@@ -215,6 +235,29 @@ navigation and editing of large files.
* Changes in Specialized Modes and Packages in Emacs 27.1
++++
+** Dired
+
+*** New command 'dired-create-empty-file'.
+
+** Change Logs and VC
+
+*** Recording ChangeLog entries doesn't require an actual file.
+If a ChangeLog file doesn't exist, and if the new variable
+'add-log-dont-create-changelog-file' is non-nil (which is the
+default), commands such as 'C-x 4 a' will add log entries to a
+suitable named temporary buffer. (An existing ChangeLog file will
+still be used if it exists.) Set the variable to nil to get the
+previous behavior of always creating a buffer that visits a ChangeLog
+file.
+
+** diff-mode
+*** Hunks are now automatically refined by default
+To disable it, set the new defcustom 'diff-font-lock-refine' to nil.
+
+*** File headers can be shortened, mimicking Magit's diff format
+To enable it, set the new defcustom 'diff-font-lock-prettify to t.
+
** Browse-url
*** The function 'browse-url-emacs' can now visit a URL in selected window.
@@ -224,6 +267,10 @@ shown in the currently selected window.
** Comint
+++
+*** 'send-invisible' is now an obsolete alias for `comint-send-invisible'
+Also, 'shell-strip-ctrl-m' is declared obsolete.
+
++++
*** 'C-c .' (comint-insert-previous-argument) no longer interprets '&'.
This feature caused problems when '&&' was present in the previous
command. Since this command emulates 'M-.' in Bash and zsh, neither
@@ -240,6 +287,11 @@ better emulate 'M-.' in both Bash and zsh, since the
former counts
from the beginning of the arguments, while the latter counts from the
end.
+** Term
+
+---
+*** 'term-read-noecho' is now obsolete, use 'read-passwd' instead.
+
** Flymake
+++
@@ -371,6 +423,13 @@ bound to 'C-c C-f'.
when escaping text and in addition all numeric entities when
unescaping text.
+** Python mode
+
+---
+*** Python mode supports three different font lock decoration levels.
+The maximum level is used by default; customize
+'font-lock-maximum-decoration' to tone down the decoration.
+
** Dired
+++
@@ -417,6 +476,14 @@ the shift key.
*** Isearch now remembers the regexp-based search mode for words/symbols
and case-sensitivity together with search strings in the search ring.
+** Debugger
+
++++
+*** The Lisp Debugger is now based on 'backtrace-mode'.
+Backtrace mode adds fontification and commands for changing the
+appearance of backtrace frames. See the node "Backtraces" in the Elisp
+manual for documentation of the new mode and its commands.
+
** Edebug
+++
@@ -426,14 +493,27 @@ using the new variables 'edebug-behavior-alist',
'edebug-new-definition-function'. Edebug's behavior can be changed
globally or for individual definitions.
++++
+*** Edebug's backtrace buffer now uses 'backtrace-mode'.
+Backtrace mode adds fontification, links and commands for changing the
+appearance of backtrace frames. See the node "Backtraces" in the Elisp
+manual for documentation of the new mode and its commands.
+
+The binding of 'd' in Edebug's keymap is now 'edebug-pop-to-backtrace'
+which replaces 'edebug-backtrace'. Consequently Edebug's backtrace
+windows now behave like those of the Lisp Debugger and of ERT, in that
+when they appear they will be the selected window.
+
+The new 'backtrace-goto-source' command, bound to 's', works in
+Edebug's backtraces on backtrace frames whose source code has
+been instrumented by Edebug.
+
** Enhanced xterm support
*** New variable 'xterm-set-window-title' controls whether Emacs sets
the XTerm window title. This feature is experimental and is disabled
by default.
-** Gamegrid
-
** grep
+++
@@ -450,6 +530,14 @@ The abbreviation can be disabled by the new option
*** New variable 'ert-quiet' allows to make ERT output in batch mode
less verbose by removing non-essential information.
++++
+*** ERT's backtrace buffer now uses 'backtrace-mode'.
+Backtrace mode adds fontification and commands for changing the
+appearance of backtrace frames. See the node "Backtraces" in the Elisp
+manual for documentation of the new mode and its commands.
+
+** Gamegrid
+
---
*** Gamegrid now determines its default glyph size based on display
dimensions, instead of always using 16 pixels. As a result, Tetris,
@@ -520,7 +608,7 @@ process. It now accepts signals specified either by name
or by its number.
** Tramp
+++
-*** New connection method "owncloud", which allows to access OwnCloud
+*** New connection method "nextcloud", which allows to access OwnCloud
or NextCloud hosted files and directories.
+++
@@ -534,6 +622,11 @@ are obsoleted in GVFS.
*** The user option 'tramp-ignored-file-name-regexp' allows to disable
Tramp for some look-alike remote file names.
+** Register
+---
+*** The return value of method 'register-val-describe' includes the
+names of buffers shown by the windows of a window configuration.
+
---
** The options.el library has been removed.
It was obsolete since Emacs 22.1, replaced by customize.
@@ -542,7 +635,6 @@ It was obsolete since Emacs 22.1, replaced by customize.
Use of built-in libgnutls based functionality (described in the Emacs
GnuTLS manual) is recommended instead.
-
** Message
+++
@@ -588,6 +680,17 @@ If this option is non-nil, messages appended to an output
file by the
selects the messages to summarize with a regexp that matches the
sender of the current message.
+** Threads
+
++++
+*** New variable 'main-thread' holds Emacs's main thread.
+This is handy in Lisp programs that run on a non-main thread and want
+to signal the main thread, e.g., when they encounter an error.
+
++++
+*** 'thread-join' returns the result of the finished thread now.
+
+
* New Modes and Packages in Emacs 27.1
+++
@@ -605,6 +708,13 @@ transport strategies as well as a separate API to use
them. A
transport implementation for process-based communication, such as is
used by the Language Server Protocol (LSP), is readily available.
++++
+** Backtrace mode improves viewing of Elisp backtraces.
+Backtrace mode adds pretty printing, fontification and ellipsis
+expansion to backtrace buffers produced by the Lisp debugger, Edebug
+and ERT. See the node "Backtraces" in the Elisp manual for
+documentation of the new mode and its commands.
+
* Incompatible Lisp Changes in Emacs 27.1
@@ -645,7 +755,7 @@ as new-style, bind the new variable
'force-new-style-backquotes' to t.
** When formatting a floating-point number as an octal or hexadecimal
integer, Emacs now signals an error if the number is too large for the
-implementation to format (Bug#30408).
+implementation to format.
---
** Some functions and variables obsolete since Emacs 22 have been removed:
@@ -694,13 +804,15 @@ however applications should instead call
'display-buffer-in-side-window'
is backwards-compatible with versions of Emacs in which the old function
exists. See the node "Displaying Buffers in Side Windows" in the ELisp
manual for more details.
+
* Lisp Changes in Emacs 27.1
+++
** New function 'proper-list-p'.
Given a proper list as argument, this predicate returns its length;
-otherwise, it returns nil.
+otherwise, it returns nil. 'format-proper-list-p' is now an obsolete
+alias for the new function.
+++
** Emacs Lisp integers can be of arbitrary precision. The new
@@ -758,6 +870,15 @@ between two strings.
** 'print-quoted' now defaults to t, so if you want to see
(quote x) instead of 'x you will have to bind it to nil where applicable.
++++
+** Numbers formatted via %o or %x may now be formatted as signed integers.
+This avoids problems in calls like (read (format "#x%x" -1)), and is
+more compatible with bignums, a planned feature. To get this
+behavior, set the experimental variable binary-as-unsigned to nil,
+and if the new behavior breaks your code please email
address@hidden Because %o and %x can now format signed
+integers, they now support the + and space flags.
+
** To avoid confusion caused by "smart quotes", the reader signals an
error when reading Lisp symbols which begin with one of the following
quotation characters: ‘’‛“”‟〞＂＇. A symbol beginning with such a
@@ -782,12 +903,12 @@ The new variable 'comment-use-syntax-ppss' can be set to
nil to recover the old
behavior if needed.
** The 'server-name' and 'server-socket-dir' variables are set when a
-socket has been passed to Emacs (Bug#24218).
+socket has been passed to Emacs.
---
** The 'file-system-info' function is now available on all platforms.
instead of just Microsoft platforms. This fixes a 'get-free-disk-space'
-bug on OS X 10.8 and later (Bug#28639).
+bug on OS X 10.8 and later.
+++
** 'memory-limit' now returns a better estimate of memory consumption.
@@ -804,12 +925,22 @@ changes and the change hooks are time consuming.
remote systems, which support this check.
+++
+** 'eql', 'make-hash-table', etc. now treat NaNs consistently.
+Formerly, some of these functions ignored signs and significands of
+NaNs. Now, all these functions treat NaN signs and significands as
+significant. For example, (eql 0.0e+NaN -0.0e+NaN) now returns nil
+because the two NaNs have different signs; formerly it returned t.
+Also, Emacs now reads and prints NaN significands; e.g., if X is a
+NaN, (format "%s" X) now returns "0.0e+NaN", "1.0e+NaN", etc.,
+depending on X's significand.
+
++++
** The function 'make-string' accepts an additional optional argument.
If the optional third argument is non-nil, 'make-string' will produce
a multibyte string even if its second argument is an ASCII character.
** (format "%d" X) no longer mishandles a floating-point number X that
-does not fit in a machine integer (Bug#30408).
+does not fit in a machine integer.
** New JSON parsing and serialization functions 'json-serialize',
'json-insert', 'json-parse-string', and 'json-parse-buffer'. These
@@ -844,10 +975,6 @@ higher-level functions.
some years back. It now respects 'imagemagick-types-inhibit' as a way
to disable that.
-+++
-** The new function 'read-answer' accepts either long or short answers
-depending on the new customizable variable 'read-answer-short'.
-
** The function 'load' now behaves correctly when loading modules.
Specifically, it puts the module name into 'load-history', prints
loading messages if requested, and protects against recursive loads.
diff --git a/etc/NEWS.26 b/etc/NEWS.26
index 58b4c4a..a766e96 100644
--- a/etc/NEWS.26
+++ b/etc/NEWS.26
@@ -85,12 +85,20 @@ it now shows the global revision number, in the form of its
changeset
hash value. To get back the previous behavior, customize the new
option 'vc-hg-symbolic-revision-styles' to the value '("{rev}")'.
+---
+** shadowfile.el has been rewritten to support Tramp file names.
+
* New Modes and Packages in Emacs 26.2
* Incompatible Lisp Changes in Emacs 26.2
+---
+** shadowfile config files have changed their syntax.
+Existing files "~/.emacs.d/shadows" and "~/.emacs.d/shadow_todo" must
+be removed prior using the changed 'shadow-*' commands.
+
* Lisp Changes in Emacs 26.2
diff --git a/etc/emacs.service b/etc/emacs.service
index b29177b..dbcb6bc 100644
--- a/etc/emacs.service
+++ b/etc/emacs.service
@@ -7,7 +7,7 @@ Description=Emacs text editor
Documentation=info:emacs man:emacs(1) https://gnu.org/software/emacs/
[Service]
-Type=simple
+Type=notify
ExecStart=emacs --fg-daemon
ExecStop=emacsclient --eval "(kill-emacs)"
Environment=SSH_AUTH_SOCK=%t/keyring/ssh
diff --git a/etc/themes/adwaita-theme.el b/etc/themes/adwaita-theme.el
index b376153..415db8a 100644
--- a/etc/themes/adwaita-theme.el
+++ b/etc/themes/adwaita-theme.el
@@ -99,8 +99,4 @@ default look of the Gnome 3 desktop.")
`(diff-added ((,class (:bold t :foreground "#4E9A06"))))
`(diff-removed ((,class (:bold t :foreground "#F5666D"))))))
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; adwaita-theme.el ends here
diff --git a/etc/themes/deeper-blue-theme.el b/etc/themes/deeper-blue-theme.el
index c6aa175..0700f4f 100644
--- a/etc/themes/deeper-blue-theme.el
+++ b/etc/themes/deeper-blue-theme.el
@@ -110,8 +110,4 @@
(provide-theme 'deeper-blue)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; deeper-blue-theme.el ends here
diff --git a/etc/themes/dichromacy-theme.el b/etc/themes/dichromacy-theme.el
index 793209c..bfced43 100644
--- a/etc/themes/dichromacy-theme.el
+++ b/etc/themes/dichromacy-theme.el
@@ -122,8 +122,4 @@ Ansi-Color faces are included.")
(provide-theme 'dichromacy)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; dichromacy-theme.el ends here
diff --git a/etc/themes/leuven-theme.el b/etc/themes/leuven-theme.el
index 5c0d19c..c3c6665 100644
--- a/etc/themes/leuven-theme.el
+++ b/etc/themes/leuven-theme.el
@@ -708,7 +708,6 @@ Semantic, and Ansi-Color faces are included -- and much
more...")
;; time-stamp-format: "%:y%02m%02d.%02H%02M"
;; time-stamp-start: "Version: "
;; time-stamp-end: "$"
-;; no-byte-compile: t
;; End:
;;; leuven-theme.el ends here
diff --git a/etc/themes/light-blue-theme.el b/etc/themes/light-blue-theme.el
index 9935c56..ba00db6 100644
--- a/etc/themes/light-blue-theme.el
+++ b/etc/themes/light-blue-theme.el
@@ -61,8 +61,4 @@
(provide-theme 'light-blue)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; light-blue-theme.el ends here
diff --git a/etc/themes/manoj-dark-theme.el b/etc/themes/manoj-dark-theme.el
index fe61441..ddcaa0b 100644
--- a/etc/themes/manoj-dark-theme.el
+++ b/etc/themes/manoj-dark-theme.el
@@ -700,8 +700,4 @@ jarring angry fruit salad look to reduce eye fatigue.")
(provide-theme 'manoj-dark)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; manoj-dark.el ends here
diff --git a/etc/themes/misterioso-theme.el b/etc/themes/misterioso-theme.el
index 42e448d..6c1eec0 100644
--- a/etc/themes/misterioso-theme.el
+++ b/etc/themes/misterioso-theme.el
@@ -103,8 +103,4 @@
(provide-theme 'misterioso)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; misterioso-theme.el ends here
diff --git a/etc/themes/tango-dark-theme.el b/etc/themes/tango-dark-theme.el
index 3b6eeb7..dae77a5 100644
--- a/etc/themes/tango-dark-theme.el
+++ b/etc/themes/tango-dark-theme.el
@@ -170,8 +170,4 @@ Semantic, and Ansi-Color faces are included.")
(provide-theme 'tango-dark)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; tango-dark-theme.el ends here
diff --git a/etc/themes/tango-theme.el b/etc/themes/tango-theme.el
index a7a79c0..4fe2480 100644
--- a/etc/themes/tango-theme.el
+++ b/etc/themes/tango-theme.el
@@ -154,8 +154,4 @@ Semantic, and Ansi-Color faces are included.")
(provide-theme 'tango)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; tango-theme.el ends here
diff --git a/etc/themes/tsdh-dark-theme.el b/etc/themes/tsdh-dark-theme.el
index 287fef8..c216750 100644
--- a/etc/themes/tsdh-dark-theme.el
+++ b/etc/themes/tsdh-dark-theme.el
@@ -144,8 +144,4 @@
(provide-theme 'tsdh-dark)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; tsdh-dark-theme.el ends here
diff --git a/etc/themes/tsdh-light-theme.el b/etc/themes/tsdh-light-theme.el
index 17a86fd..ce9d1a2 100644
--- a/etc/themes/tsdh-light-theme.el
+++ b/etc/themes/tsdh-light-theme.el
@@ -106,9 +106,4 @@ Used and created by Tassilo Horn.")
(provide-theme 'tsdh-light)
-
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; tsdh-light-theme.el ends here
diff --git a/etc/themes/wheatgrass-theme.el b/etc/themes/wheatgrass-theme.el
index 9585e3a..8d34c28 100644
--- a/etc/themes/wheatgrass-theme.el
+++ b/etc/themes/wheatgrass-theme.el
@@ -83,8 +83,4 @@ of green, brown, and blue.")
(provide-theme 'wheatgrass)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; wheatgrass-theme.el ends here
diff --git a/etc/themes/whiteboard-theme.el b/etc/themes/whiteboard-theme.el
index 5db0ddd..fe46cb0 100644
--- a/etc/themes/whiteboard-theme.el
+++ b/etc/themes/whiteboard-theme.el
@@ -100,8 +100,4 @@
(provide-theme 'whiteboard)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; whiteboard-theme.el ends here
diff --git a/etc/themes/wombat-theme.el b/etc/themes/wombat-theme.el
index 583b8dc..00f29bb 100644
--- a/etc/themes/wombat-theme.el
+++ b/etc/themes/wombat-theme.el
@@ -102,8 +102,4 @@ are included.")
(provide-theme 'wombat)
-;; Local Variables:
-;; no-byte-compile: t
-;; End:
-
;;; wombat-theme.el ends here
diff --git a/lib-src/Makefile.in b/lib-src/Makefile.in
index fa37d8e..b2b9017 100644
--- a/lib-src/Makefile.in
+++ b/lib-src/Makefile.in
@@ -361,13 +361,9 @@ TAGS: etags${EXEEXT} ${tagsfiles}
../lib/libgnu.a: $(config_h)
$(MAKE) -C ../lib all
-regex.o: $(srcdir)/../src/regex.c $(srcdir)/../src/regex.h $(config_h)
- $(AM_V_CC)$(CC) -c $(CPP_CFLAGS) $<
-
-
-etags_deps = ${srcdir}/etags.c regex.o $(NTLIB) $(config_h)
+etags_deps = ${srcdir}/etags.c $(NTLIB) $(config_h)
etags_cflags = -DEMACS_NAME="\"GNU Emacs\"" -DVERSION="\"${version}\"" -o $@
-etags_libs = regex.o $(NTLIB) $(LOADLIBES)
+etags_libs = $(NTLIB) $(LOADLIBES)
etags${EXEEXT}: ${etags_deps}
$(AM_V_CCLD)$(CC) ${ALL_CFLAGS} $(etags_cflags) $< $(etags_libs)
diff --git a/lib-src/etags.c b/lib-src/etags.c
index b3b4575..ee50670 100644
--- a/lib-src/etags.c
+++ b/lib-src/etags.c
@@ -6401,7 +6401,7 @@ add_regex (char *regexp_pattern, language *lang)
*patbuf = zeropattern;
if (ignore_case)
{
- static char lc_trans[UCHAR_MAX + 1];
+ static unsigned char lc_trans[UCHAR_MAX + 1];
int i;
for (i = 0; i < UCHAR_MAX + 1; i++)
lc_trans[i] = c_tolower (i);
diff --git a/lib-src/ntlib.c b/lib-src/ntlib.c
index 9551285..4ca521d 100644
--- a/lib-src/ntlib.c
+++ b/lib-src/ntlib.c
@@ -31,6 +31,10 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#include <ctype.h>
#include <sys/timeb.h>
#include <mbstring.h>
+#include <locale.h>
+
+#include <nl_types.h>
+#include <langinfo.h>
#include "ntlib.h"
@@ -423,3 +427,66 @@ sys_open (const char * path, int oflag, int mode)
{
return _open (path, oflag, mode);
}
+
+/* Emulation of nl_langinfo that supports only CODESET.
+ Used in Gnulib regex.c. */
+char *
+nl_langinfo (nl_item item)
+{
+ switch (item)
+ {
+ case CODESET:
+ {
+ /* Shamelessly stolen from Gnulib's nl_langinfo.c, modulo
+ CPP directives. */
+ static char buf[2 + 10 + 1];
+ char const *locale = setlocale (LC_CTYPE, NULL);
+ char *codeset = buf;
+ size_t codesetlen;
+ codeset[0] = '\0';
+
+ if (locale && locale[0])
+ {
+ /* If the locale name contains an encoding after the
+ dot, return it. */
+ char *dot = strchr (locale, '.');
+
+ if (dot)
+ {
+ /* Look for the possible @... trailer and remove it,
+ if any. */
+ char *codeset_start = dot + 1;
+ char const *modifier = strchr (codeset_start, '@');
+
+ if (! modifier)
+ codeset = codeset_start;
+ else
+ {
+ codesetlen = modifier - codeset_start;
+ if (codesetlen < sizeof buf)
+ {
+ codeset = memcpy (buf, codeset_start, codesetlen);
+ codeset[codesetlen] = '\0';
+ }
+ }
+ }
+ }
+ /* If setlocale is successful, it returns the number of the
+ codepage, as a string. Otherwise, fall back on Windows
+ API GetACP, which returns the locale's codepage as a
+ number (although this doesn't change according to what
+ the 'setlocale' call specified). Either way, prepend
+ "CP" to make it a valid codeset name. */
+ codesetlen = strlen (codeset);
+ if (0 < codesetlen && codesetlen < sizeof buf - 2)
+ memmove (buf + 2, codeset, codesetlen + 1);
+ else
+ sprintf (buf + 2, "%u", GetACP ());
+ codeset = memcpy (buf, "CP", 2);
+
+ return codeset;
+ }
+ default:
+ return (char *) "";
+ }
+}
diff --git a/lib-src/pop.c b/lib-src/pop.c
index 10aac95..731f951 100644
--- a/lib-src/pop.c
+++ b/lib-src/pop.c
@@ -30,8 +30,12 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#include "ntlib.h"
#undef _WIN32_WINNT
#define _WIN32_WINNT 0x0501 /* for getaddrinfo stuff */
-#include <winsock2.h>
-#include <ws2tcpip.h>
+#if defined __MINGW32_VERSION && __MINGW32_VERSION >= 5000002L
+# include <windows.h>
+#else
+# include <winsock2.h>
+#endif
+# include <ws2tcpip.h>
#undef getaddrinfo
#define getaddrinfo sys_getaddrinfo
#undef freeaddrinfo
diff --git a/lib/Makefile.in b/lib/Makefile.in
index 201f4b5..b26db27 100644
--- a/lib/Makefile.in
+++ b/lib/Makefile.in
@@ -79,9 +79,15 @@ endif
Makefile: ../config.status $(srcdir)/Makefile.in
$(MAKE) -C .. src/$@
+# Object modules that need not be built for Emacs.
+# Emacs does not need e-regex.o (it has its own regex-emacs.c),
+# and building it would just waste time.
+not_emacs_OBJECTS = regex.o
+
libgnu_a_OBJECTS = $(gl_LIBOBJS) \
$(patsubst %.c,%.o,$(filter %.c,$(libgnu_a_SOURCES)))
-libegnu_a_OBJECTS = $(patsubst %.o,e-%.o,$(libgnu_a_OBJECTS))
+for_emacs_OBJECTS = $(filter-out $(not_emacs_OBJECTS),$(libgnu_a_OBJECTS))
+libegnu_a_OBJECTS = $(patsubst %.o,e-%.o,$(for_emacs_OBJECTS))
$(libegnu_a_OBJECTS) $(libgnu_a_OBJECTS): $(BUILT_SOURCES)
diff --git a/lib/gnulib.mk.in b/lib/gnulib.mk.in
index 3e91738..7ad3908 100644
--- a/lib/gnulib.mk.in
+++ b/lib/gnulib.mk.in
@@ -20,7 +20,134 @@
# the same distribution terms as the rest of that program.
#
# Generated by gnulib-tool.
-# Reproduce by: gnulib-tool --import --lib=libgnu --source-base=lib
--m4-base=m4 --doc-base=doc --tests-base=tests --aux-dir=build-aux
--avoid=close --avoid=dup --avoid=fchdir --avoid=fstat --avoid=malloc-posix
--avoid=msvc-inval --avoid=msvc-nothrow --avoid=openat-die --avoid=opendir
--avoid=raise --avoid=save-cwd --avoid=select --avoid=setenv
--avoid=sigprocmask --avoid=stat --avoid=stdarg --avoid=stdbool
--avoid=threadlib --avoid=tzset --avoid=unsetenv --avoid=utime --avoid=utime-h
-- [...]
+# Reproduce by:
+# gnulib-tool --import \
+# --lib=libgnu \
+# --source-base=lib \
+# --m4-base=m4 \
+# --doc-base=doc \
+# --tests-base=tests \
+# --aux-dir=build-aux \
+# --gnu-make \
+# --makefile-name=gnulib.mk.in \
+# --conditional-dependencies \
+# --no-libtool \
+# --macro-prefix=gl \
+# --no-vc-files \
+# --avoid=btowc \
+# --avoid=close \
+# --avoid=dup \
+# --avoid=fchdir \
+# --avoid=fstat \
+# --avoid=langinfo \
+# --avoid=lock \
+# --avoid=malloc-posix \
+# --avoid=mbrtowc \
+# --avoid=mbsinit \
+# --avoid=msvc-inval \
+# --avoid=msvc-nothrow \
+# --avoid=nl_langinfo \
+# --avoid=openat-die \
+# --avoid=opendir \
+# --avoid=raise \
+# --avoid=save-cwd \
+# --avoid=select \
+# --avoid=setenv \
+# --avoid=sigprocmask \
+# --avoid=stat \
+# --avoid=stdarg \
+# --avoid=stdbool \
+# --avoid=threadlib \
+# --avoid=tzset \
+# --avoid=unsetenv \
+# --avoid=utime \
+# --avoid=utime-h \
+# --avoid=wchar \
+# --avoid=wcrtomb \
+# --avoid=wctype-h \
+# alloca-opt \
+# binary-io \
+# byteswap \
+# c-ctype \
+# c-strcase \
+# careadlinkat \
+# close-stream \
+# count-leading-zeros \
+# count-one-bits \
+# count-trailing-zeros \
+# crypto/md5-buffer \
+# crypto/sha1-buffer \
+# crypto/sha256-buffer \
+# crypto/sha512-buffer \
+# d-type \
+# diffseq \
+# dtoastr \
+# dtotimespec \
+# dup2 \
+# environ \
+# execinfo \
+# explicit_bzero \
+# faccessat \
+# fcntl \
+# fcntl-h \
+# fdatasync \
+# fdopendir \
+# filemode \
+# filevercmp \
+# flexmember \
+# fpieee \
+# fstatat \
+# fsusage \
+# fsync \
+# getloadavg \
+# getopt-gnu \
+# gettime \
+# gettimeofday \
+# gitlog-to-changelog \
+# ieee754-h \
+# ignore-value \
+# intprops \
+# largefile \
+# lstat \
+# manywarnings \
+# memrchr \
+# minmax \
+# mkostemp \
+# mktime \
+# nstrftime \
+# pipe2 \
+# pselect \
+# pthread_sigmask \
+# putenv \
+# qcopy-acl \
+# readlink \
+# readlinkat \
+# regex \
+# sig2str \
+# socklen \
+# stat-time \
+# std-gnu11 \
+# stdalign \
+# stddef \
+# stdio \
+# stpcpy \
+# strtoimax \
+# symlink \
+# sys_stat \
+# sys_time \
+# tempname \
+# time \
+# time_r \
+# time_rz \
+# timegm \
+# timer-time \
+# timespec-add \
+# timespec-sub \
+# unlocked-io \
+# update-copyright \
+# utimens \
+# vla \
+# warnings
MOSTLYCLEANFILES += core *.stackdump
@@ -99,11 +226,13 @@ GETOPT_CDEFS_H = @GETOPT_CDEFS_H@
GETOPT_H = @GETOPT_H@
GFILENOTIFY_CFLAGS = @GFILENOTIFY_CFLAGS@
GFILENOTIFY_LIBS = @GFILENOTIFY_LIBS@
+GLIBC21 = @GLIBC21@
GL_COND_LIBTOOL = @GL_COND_LIBTOOL@
GL_GENERATE_ALLOCA_H = @GL_GENERATE_ALLOCA_H@
GL_GENERATE_BYTESWAP_H = @GL_GENERATE_BYTESWAP_H@
GL_GENERATE_ERRNO_H = @GL_GENERATE_ERRNO_H@
GL_GENERATE_EXECINFO_H = @GL_GENERATE_EXECINFO_H@
+GL_GENERATE_IEEE754_H = @GL_GENERATE_IEEE754_H@
GL_GENERATE_LIMITS_H = @GL_GENERATE_LIMITS_H@
GL_GENERATE_STDALIGN_H = @GL_GENERATE_STDALIGN_H@
GL_GENERATE_STDDEF_H = @GL_GENERATE_STDDEF_H@
@@ -530,6 +659,7 @@ HAVE_WINSOCK2_H = @HAVE_WINSOCK2_H@
HAVE_XSERVER = @HAVE_XSERVER@
HAVE__EXIT = @HAVE__EXIT@
HYBRID_MALLOC = @HYBRID_MALLOC@
+IEEE754_H = @IEEE754_H@
IMAGEMAGICK_CFLAGS = @IMAGEMAGICK_CFLAGS@
IMAGEMAGICK_LIBS = @IMAGEMAGICK_LIBS@
INCLUDE_NEXT = @INCLUDE_NEXT@
@@ -905,6 +1035,7 @@ gameuser = @gameuser@
gl_GNULIB_ENABLED_03e0aaad4cb89ca757653bd367a6ccb7 =
@gl_GNULIB_ENABLED_03e0aaad4cb89ca757653bd367a6ccb7@
gl_GNULIB_ENABLED_2049e887c7e5308faad27b3f894bb8c9 =
@gl_GNULIB_ENABLED_2049e887c7e5308faad27b3f894bb8c9@
gl_GNULIB_ENABLED_260941c0e5dc67ec9e87d1fb321c300b =
@gl_GNULIB_ENABLED_260941c0e5dc67ec9e87d1fb321c300b@
+gl_GNULIB_ENABLED_37f71b604aa9c54446783d80f42fe547 =
@gl_GNULIB_ENABLED_37f71b604aa9c54446783d80f42fe547@
gl_GNULIB_ENABLED_5264294aa0a5557541b53c8c741f7f31 =
@gl_GNULIB_ENABLED_5264294aa0a5557541b53c8c741f7f31@
gl_GNULIB_ENABLED_6099e9737f757db36c47fa9d9f02e88c =
@gl_GNULIB_ENABLED_6099e9737f757db36c47fa9d9f02e88c@
gl_GNULIB_ENABLED_682e609604ccaac6be382e4ee3a4eaec =
@gl_GNULIB_ENABLED_682e609604ccaac6be382e4ee3a4eaec@
@@ -1671,6 +1802,32 @@ EXTRA_libgnu_a_SOURCES += group-member.c
endif
## end gnulib module group-member
+## begin gnulib module ieee754-h
+ifeq (,$(OMIT_GNULIB_MODULE_ieee754-h))
+
+BUILT_SOURCES += $(IEEE754_H)
+
+# We need the following in order to create <ieee754.h> when the system
+# doesn't have one that works with the given compiler.
+ifneq (,$(GL_GENERATE_IEEE754_H))
+ieee754.h: ieee754.in.h $(top_builddir)/config.status
+ $(AM_V_GEN)rm -f address@hidden && \
+ { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */'; \
+ sed -e 's/ifndef _GL_GNULIB_HEADER/if 0/g' \
+ $(srcdir)/ieee754.in.h; \
+ } > address@hidden && \
+ mv -f address@hidden $@
+else
+ieee754.h: $(top_builddir)/config.status
+ rm -f $@
+endif
+MOSTLYCLEANFILES += ieee754.h ieee754.h-t
+
+EXTRA_DIST += ieee754.in.h
+
+endif
+## end gnulib module ieee754-h
+
## begin gnulib module ignore-value
ifeq (,$(OMIT_GNULIB_MODULE_ignore-value))
@@ -1950,6 +2107,17 @@ EXTRA_libgnu_a_SOURCES += at-func.c readlinkat.c
endif
## end gnulib module readlinkat
+## begin gnulib module regex
+ifeq (,$(OMIT_GNULIB_MODULE_regex))
+
+
+EXTRA_DIST += regcomp.c regex.c regex.h regex_internal.c regex_internal.h
regexec.c
+
+EXTRA_libgnu_a_SOURCES += regcomp.c regex.c regex_internal.c regexec.c
+
+endif
+## end gnulib module regex
+
## begin gnulib module root-uid
ifeq (,$(OMIT_GNULIB_MODULE_root-uid))
diff --git a/lib/ieee754.in.h b/lib/ieee754.in.h
new file mode 100644
index 0000000..316ac03
--- /dev/null
+++ b/lib/ieee754.in.h
@@ -0,0 +1,222 @@
+/* Copyright (C) 1992-2018 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU General Public
+ License as published by the Free Software Foundation; either
+ version 3 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ General Public License for more details.
+
+ You should have received a copy of the GNU General Public
+ License along with the GNU C Library; if not, see
+ <http://www.gnu.org/licenses/>. */
+
+#ifndef _IEEE754_H
+
+#define _IEEE754_H 1
+
+#ifndef _GL_GNULIB_HEADER
+/* Ordinary glibc usage. */
+# include <features.h>
+# include <endian.h>
+#else
+/* Gnulib usage. */
+# ifndef __BEGIN_DECLS
+# ifdef __cplusplus
+# define __BEGIN_DECLS extern "C" {
+# define __END_DECLS }
+# else
+# define __BEGIN_DECLS
+# define __END_DECLS
+# endif
+# endif
+# ifndef __FLOAT_WORD_ORDER
+# define __LITTLE_ENDIAN 1234
+# define __BIG_ENDIAN 4321
+# ifdef WORDS_BIGENDIAN
+# define __BYTE_ORDER __BIG_ENDIAN
+# else
+# define __BYTE_ORDER __LITTLE_ENDIAN
+# endif
+# define __FLOAT_WORD_ORDER __BYTE_ORDER
+# endif
+#endif
+
+__BEGIN_DECLS
+
+union ieee754_float
+ {
+ float f;
+
+ /* This is the IEEE 754 single-precision format. */
+ struct
+ {
+#if __BYTE_ORDER == __BIG_ENDIAN
+ unsigned int negative:1;
+ unsigned int exponent:8;
+ unsigned int mantissa:23;
+#endif /* Big endian. */
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+ unsigned int mantissa:23;
+ unsigned int exponent:8;
+ unsigned int negative:1;
+#endif /* Little endian. */
+ } ieee;
+
+ /* This format makes it easier to see if a NaN is a signalling NaN. */
+ struct
+ {
+#if __BYTE_ORDER == __BIG_ENDIAN
+ unsigned int negative:1;
+ unsigned int exponent:8;
+ unsigned int quiet_nan:1;
+ unsigned int mantissa:22;
+#endif /* Big endian. */
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+ unsigned int mantissa:22;
+ unsigned int quiet_nan:1;
+ unsigned int exponent:8;
+ unsigned int negative:1;
+#endif /* Little endian. */
+ } ieee_nan;
+ };
+
+#define IEEE754_FLOAT_BIAS 0x7f /* Added to exponent. */
+
+
+union ieee754_double
+ {
+ double d;
+
+ /* This is the IEEE 754 double-precision format. */
+ struct
+ {
+#if __BYTE_ORDER == __BIG_ENDIAN
+ unsigned int negative:1;
+ unsigned int exponent:11;
+ /* Together these comprise the mantissa. */
+ unsigned int mantissa0:20;
+ unsigned int mantissa1:32;
+#endif /* Big endian. */
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+# if __FLOAT_WORD_ORDER == __BIG_ENDIAN
+ unsigned int mantissa0:20;
+ unsigned int exponent:11;
+ unsigned int negative:1;
+ unsigned int mantissa1:32;
+# else
+ /* Together these comprise the mantissa. */
+ unsigned int mantissa1:32;
+ unsigned int mantissa0:20;
+ unsigned int exponent:11;
+ unsigned int negative:1;
+# endif
+#endif /* Little endian. */
+ } ieee;
+
+ /* This format makes it easier to see if a NaN is a signalling NaN. */
+ struct
+ {
+#if __BYTE_ORDER == __BIG_ENDIAN
+ unsigned int negative:1;
+ unsigned int exponent:11;
+ unsigned int quiet_nan:1;
+ /* Together these comprise the mantissa. */
+ unsigned int mantissa0:19;
+ unsigned int mantissa1:32;
+#else
+# if __FLOAT_WORD_ORDER == __BIG_ENDIAN
+ unsigned int mantissa0:19;
+ unsigned int quiet_nan:1;
+ unsigned int exponent:11;
+ unsigned int negative:1;
+ unsigned int mantissa1:32;
+# else
+ /* Together these comprise the mantissa. */
+ unsigned int mantissa1:32;
+ unsigned int mantissa0:19;
+ unsigned int quiet_nan:1;
+ unsigned int exponent:11;
+ unsigned int negative:1;
+# endif
+#endif
+ } ieee_nan;
+ };
+
+#define IEEE754_DOUBLE_BIAS 0x3ff /* Added to exponent. */
+
+
+union ieee854_long_double
+ {
+ long double d;
+
+ /* This is the IEEE 854 double-extended-precision format. */
+ struct
+ {
+#if __BYTE_ORDER == __BIG_ENDIAN
+ unsigned int negative:1;
+ unsigned int exponent:15;
+ unsigned int empty:16;
+ unsigned int mantissa0:32;
+ unsigned int mantissa1:32;
+#endif
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+# if __FLOAT_WORD_ORDER == __BIG_ENDIAN
+ unsigned int exponent:15;
+ unsigned int negative:1;
+ unsigned int empty:16;
+ unsigned int mantissa0:32;
+ unsigned int mantissa1:32;
+# else
+ unsigned int mantissa1:32;
+ unsigned int mantissa0:32;
+ unsigned int exponent:15;
+ unsigned int negative:1;
+ unsigned int empty:16;
+# endif
+#endif
+ } ieee;
+
+ /* This is for NaNs in the IEEE 854 double-extended-precision format. */
+ struct
+ {
+#if __BYTE_ORDER == __BIG_ENDIAN
+ unsigned int negative:1;
+ unsigned int exponent:15;
+ unsigned int empty:16;
+ unsigned int one:1;
+ unsigned int quiet_nan:1;
+ unsigned int mantissa0:30;
+ unsigned int mantissa1:32;
+#endif
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+# if __FLOAT_WORD_ORDER == __BIG_ENDIAN
+ unsigned int exponent:15;
+ unsigned int negative:1;
+ unsigned int empty:16;
+ unsigned int mantissa0:30;
+ unsigned int quiet_nan:1;
+ unsigned int one:1;
+ unsigned int mantissa1:32;
+# else
+ unsigned int mantissa1:32;
+ unsigned int mantissa0:30;
+ unsigned int quiet_nan:1;
+ unsigned int one:1;
+ unsigned int exponent:15;
+ unsigned int negative:1;
+ unsigned int empty:16;
+# endif
+#endif
+ } ieee_nan;
+ };
+
+#define IEEE854_LONG_DOUBLE_BIAS 0x3fff
+
+__END_DECLS
+
+#endif /* ieee754.h */
diff --git a/lib/regcomp.c b/lib/regcomp.c
new file mode 100644
index 0000000..53eb226
--- /dev/null
+++ b/lib/regcomp.c
@@ -0,0 +1,3944 @@
+/* Extended regular expression matching and search library.
+ Copyright (C) 2002-2018 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+ Contributed by Isamu Hasegawa <address@hidden>.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU General Public
+ License as published by the Free Software Foundation; either
+ version 3 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ General Public License for more details.
+
+ You should have received a copy of the GNU General Public
+ License along with the GNU C Library; if not, see
+ <https://www.gnu.org/licenses/>. */
+
+#ifdef _LIBC
+# include <locale/weight.h>
+#endif
+
+static reg_errcode_t re_compile_internal (regex_t *preg, const char * pattern,
+ size_t length, reg_syntax_t syntax);
+static void re_compile_fastmap_iter (regex_t *bufp,
+ const re_dfastate_t *init_state,
+ char *fastmap);
+static reg_errcode_t init_dfa (re_dfa_t *dfa, size_t pat_len);
+#ifdef RE_ENABLE_I18N
+static void free_charset (re_charset_t *cset);
+#endif /* RE_ENABLE_I18N */
+static void free_workarea_compile (regex_t *preg);
+static reg_errcode_t create_initial_state (re_dfa_t *dfa);
+#ifdef RE_ENABLE_I18N
+static void optimize_utf8 (re_dfa_t *dfa);
+#endif
+static reg_errcode_t analyze (regex_t *preg);
+static reg_errcode_t preorder (bin_tree_t *root,
+ reg_errcode_t (fn (void *, bin_tree_t *)),
+ void *extra);
+static reg_errcode_t postorder (bin_tree_t *root,
+ reg_errcode_t (fn (void *, bin_tree_t *)),
+ void *extra);
+static reg_errcode_t optimize_subexps (void *extra, bin_tree_t *node);
+static reg_errcode_t lower_subexps (void *extra, bin_tree_t *node);
+static bin_tree_t *lower_subexp (reg_errcode_t *err, regex_t *preg,
+ bin_tree_t *node);
+static reg_errcode_t calc_first (void *extra, bin_tree_t *node);
+static reg_errcode_t calc_next (void *extra, bin_tree_t *node);
+static reg_errcode_t link_nfa_nodes (void *extra, bin_tree_t *node);
+static Idx duplicate_node (re_dfa_t *dfa, Idx org_idx, unsigned int
constraint);
+static Idx search_duplicated_node (const re_dfa_t *dfa, Idx org_node,
+ unsigned int constraint);
+static reg_errcode_t calc_eclosure (re_dfa_t *dfa);
+static reg_errcode_t calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa,
+ Idx node, bool root);
+static reg_errcode_t calc_inveclosure (re_dfa_t *dfa);
+static Idx fetch_number (re_string_t *input, re_token_t *token,
+ reg_syntax_t syntax);
+static int peek_token (re_token_t *token, re_string_t *input,
+ reg_syntax_t syntax);
+static bin_tree_t *parse (re_string_t *regexp, regex_t *preg,
+ reg_syntax_t syntax, reg_errcode_t *err);
+static bin_tree_t *parse_reg_exp (re_string_t *regexp, regex_t *preg,
+ re_token_t *token, reg_syntax_t syntax,
+ Idx nest, reg_errcode_t *err);
+static bin_tree_t *parse_branch (re_string_t *regexp, regex_t *preg,
+ re_token_t *token, reg_syntax_t syntax,
+ Idx nest, reg_errcode_t *err);
+static bin_tree_t *parse_expression (re_string_t *regexp, regex_t *preg,
+ re_token_t *token, reg_syntax_t syntax,
+ Idx nest, reg_errcode_t *err);
+static bin_tree_t *parse_sub_exp (re_string_t *regexp, regex_t *preg,
+ re_token_t *token, reg_syntax_t syntax,
+ Idx nest, reg_errcode_t *err);
+static bin_tree_t *parse_dup_op (bin_tree_t *dup_elem, re_string_t *regexp,
+ re_dfa_t *dfa, re_token_t *token,
+ reg_syntax_t syntax, reg_errcode_t *err);
+static bin_tree_t *parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa,
+ re_token_t *token, reg_syntax_t syntax,
+ reg_errcode_t *err);
+static reg_errcode_t parse_bracket_element (bracket_elem_t *elem,
+ re_string_t *regexp,
+ re_token_t *token, int token_len,
+ re_dfa_t *dfa,
+ reg_syntax_t syntax,
+ bool accept_hyphen);
+static reg_errcode_t parse_bracket_symbol (bracket_elem_t *elem,
+ re_string_t *regexp,
+ re_token_t *token);
+#ifdef RE_ENABLE_I18N
+static reg_errcode_t build_equiv_class (bitset_t sbcset,
+ re_charset_t *mbcset,
+ Idx *equiv_class_alloc,
+ const unsigned char *name);
+static reg_errcode_t build_charclass (RE_TRANSLATE_TYPE trans,
+ bitset_t sbcset,
+ re_charset_t *mbcset,
+ Idx *char_class_alloc,
+ const char *class_name,
+ reg_syntax_t syntax);
+#else /* not RE_ENABLE_I18N */
+static reg_errcode_t build_equiv_class (bitset_t sbcset,
+ const unsigned char *name);
+static reg_errcode_t build_charclass (RE_TRANSLATE_TYPE trans,
+ bitset_t sbcset,
+ const char *class_name,
+ reg_syntax_t syntax);
+#endif /* not RE_ENABLE_I18N */
+static bin_tree_t *build_charclass_op (re_dfa_t *dfa,
+ RE_TRANSLATE_TYPE trans,
+ const char *class_name,
+ const char *extra,
+ bool non_match, reg_errcode_t *err);
+static bin_tree_t *create_tree (re_dfa_t *dfa,
+ bin_tree_t *left, bin_tree_t *right,
+ re_token_type_t type);
+static bin_tree_t *create_token_tree (re_dfa_t *dfa,
+ bin_tree_t *left, bin_tree_t *right,
+ const re_token_t *token);
+static bin_tree_t *duplicate_tree (const bin_tree_t *src, re_dfa_t *dfa);
+static void free_token (re_token_t *node);
+static reg_errcode_t free_tree (void *extra, bin_tree_t *node);
+static reg_errcode_t mark_opt_subexp (void *extra, bin_tree_t *node);
+
+/* This table gives an error message for each of the error codes listed
+ in regex.h. Obviously the order here has to be same as there.
+ POSIX doesn't require that we do anything for REG_NOERROR,
+ but why not be nice? */
+
+static const char __re_error_msgid[] =
+ {
+#define REG_NOERROR_IDX 0
+ gettext_noop ("Success") /* REG_NOERROR */
+ "\0"
+#define REG_NOMATCH_IDX (REG_NOERROR_IDX + sizeof "Success")
+ gettext_noop ("No match") /* REG_NOMATCH */
+ "\0"
+#define REG_BADPAT_IDX (REG_NOMATCH_IDX + sizeof "No match")
+ gettext_noop ("Invalid regular expression") /* REG_BADPAT */
+ "\0"
+#define REG_ECOLLATE_IDX (REG_BADPAT_IDX + sizeof "Invalid regular expression")
+ gettext_noop ("Invalid collation character") /* REG_ECOLLATE */
+ "\0"
+#define REG_ECTYPE_IDX (REG_ECOLLATE_IDX + sizeof "Invalid collation
character")
+ gettext_noop ("Invalid character class name") /* REG_ECTYPE */
+ "\0"
+#define REG_EESCAPE_IDX (REG_ECTYPE_IDX + sizeof "Invalid character
class name")
+ gettext_noop ("Trailing backslash") /* REG_EESCAPE */
+ "\0"
+#define REG_ESUBREG_IDX (REG_EESCAPE_IDX + sizeof "Trailing backslash")
+ gettext_noop ("Invalid back reference") /* REG_ESUBREG */
+ "\0"
+#define REG_EBRACK_IDX (REG_ESUBREG_IDX + sizeof "Invalid back reference")
+ gettext_noop ("Unmatched [, [^, [:, [., or [=") /* REG_EBRACK */
+ "\0"
+#define REG_EPAREN_IDX (REG_EBRACK_IDX + sizeof "Unmatched [, [^, [:, [., or
[=")
+ gettext_noop ("Unmatched ( or \\(") /* REG_EPAREN */
+ "\0"
+#define REG_EBRACE_IDX (REG_EPAREN_IDX + sizeof "Unmatched ( or \\(")
+ gettext_noop ("Unmatched \\{") /* REG_EBRACE */
+ "\0"
+#define REG_BADBR_IDX (REG_EBRACE_IDX + sizeof "Unmatched \\{")
+ gettext_noop ("Invalid content of \\{\\}") /* REG_BADBR */
+ "\0"
+#define REG_ERANGE_IDX (REG_BADBR_IDX + sizeof "Invalid content of \\{\\}")
+ gettext_noop ("Invalid range end") /* REG_ERANGE */
+ "\0"
+#define REG_ESPACE_IDX (REG_ERANGE_IDX + sizeof "Invalid range end")
+ gettext_noop ("Memory exhausted") /* REG_ESPACE */
+ "\0"
+#define REG_BADRPT_IDX (REG_ESPACE_IDX + sizeof "Memory exhausted")
+ gettext_noop ("Invalid preceding regular expression") /* REG_BADRPT */
+ "\0"
+#define REG_EEND_IDX (REG_BADRPT_IDX + sizeof "Invalid preceding regular
expression")
+ gettext_noop ("Premature end of regular expression") /* REG_EEND */
+ "\0"
+#define REG_ESIZE_IDX (REG_EEND_IDX + sizeof "Premature end of regular
expression")
+ gettext_noop ("Regular expression too big") /* REG_ESIZE */
+ "\0"
+#define REG_ERPAREN_IDX (REG_ESIZE_IDX + sizeof "Regular expression too
big")
+ gettext_noop ("Unmatched ) or \\)") /* REG_ERPAREN */
+ };
+
+static const size_t __re_error_msgid_idx[] =
+ {
+ REG_NOERROR_IDX,
+ REG_NOMATCH_IDX,
+ REG_BADPAT_IDX,
+ REG_ECOLLATE_IDX,
+ REG_ECTYPE_IDX,
+ REG_EESCAPE_IDX,
+ REG_ESUBREG_IDX,
+ REG_EBRACK_IDX,
+ REG_EPAREN_IDX,
+ REG_EBRACE_IDX,
+ REG_BADBR_IDX,
+ REG_ERANGE_IDX,
+ REG_ESPACE_IDX,
+ REG_BADRPT_IDX,
+ REG_EEND_IDX,
+ REG_ESIZE_IDX,
+ REG_ERPAREN_IDX
+ };
+
+/* Entry points for GNU code. */
+
+/* re_compile_pattern is the GNU regular expression compiler: it
+ compiles PATTERN (of length LENGTH) and puts the result in BUFP.
+ Returns 0 if the pattern was valid, otherwise an error string.
+
+ Assumes the 'allocated' (and perhaps 'buffer') and 'translate' fields
+ are set in BUFP on entry. */
+
+const char *
+re_compile_pattern (const char *pattern, size_t length,
+ struct re_pattern_buffer *bufp)
+{
+ reg_errcode_t ret;
+
+ /* And GNU code determines whether or not to get register information
+ by passing null for the REGS argument to re_match, etc., not by
+ setting no_sub, unless RE_NO_SUB is set. */
+ bufp->no_sub = !!(re_syntax_options & RE_NO_SUB);
+
+ /* Match anchors at newline. */
+ bufp->newline_anchor = 1;
+
+ ret = re_compile_internal (bufp, pattern, length, re_syntax_options);
+
+ if (!ret)
+ return NULL;
+ return gettext (__re_error_msgid + __re_error_msgid_idx[(int) ret]);
+}
+#ifdef _LIBC
+weak_alias (__re_compile_pattern, re_compile_pattern)
+#endif
+
+/* Set by 're_set_syntax' to the current regexp syntax to recognize. Can
+ also be assigned to arbitrarily: each pattern buffer stores its own
+ syntax, so it can be changed between regex compilations. */
+/* This has no initializer because initialized variables in Emacs
+ become read-only after dumping. */
+reg_syntax_t re_syntax_options;
+
+
+/* Specify the precise syntax of regexps for compilation. This provides
+ for compatibility for various utilities which historically have
+ different, incompatible syntaxes.
+
+ The argument SYNTAX is a bit mask comprised of the various bits
+ defined in regex.h. We return the old syntax. */
+
+reg_syntax_t
+re_set_syntax (reg_syntax_t syntax)
+{
+ reg_syntax_t ret = re_syntax_options;
+
+ re_syntax_options = syntax;
+ return ret;
+}
+#ifdef _LIBC
+weak_alias (__re_set_syntax, re_set_syntax)
+#endif
+
+int
+re_compile_fastmap (struct re_pattern_buffer *bufp)
+{
+ re_dfa_t *dfa = bufp->buffer;
+ char *fastmap = bufp->fastmap;
+
+ memset (fastmap, '\0', sizeof (char) * SBC_MAX);
+ re_compile_fastmap_iter (bufp, dfa->init_state, fastmap);
+ if (dfa->init_state != dfa->init_state_word)
+ re_compile_fastmap_iter (bufp, dfa->init_state_word, fastmap);
+ if (dfa->init_state != dfa->init_state_nl)
+ re_compile_fastmap_iter (bufp, dfa->init_state_nl, fastmap);
+ if (dfa->init_state != dfa->init_state_begbuf)
+ re_compile_fastmap_iter (bufp, dfa->init_state_begbuf, fastmap);
+ bufp->fastmap_accurate = 1;
+ return 0;
+}
+#ifdef _LIBC
+weak_alias (__re_compile_fastmap, re_compile_fastmap)
+#endif
+
+static inline void
+__attribute__ ((always_inline))
+re_set_fastmap (char *fastmap, bool icase, int ch)
+{
+ fastmap[ch] = 1;
+ if (icase)
+ fastmap[tolower (ch)] = 1;
+}
+
+/* Helper function for re_compile_fastmap.
+ Compile fastmap for the initial_state INIT_STATE. */
+
+static void
+re_compile_fastmap_iter (regex_t *bufp, const re_dfastate_t *init_state,
+ char *fastmap)
+{
+ re_dfa_t *dfa = bufp->buffer;
+ Idx node_cnt;
+ bool icase = (dfa->mb_cur_max == 1 && (bufp->syntax & RE_ICASE));
+ for (node_cnt = 0; node_cnt < init_state->nodes.nelem; ++node_cnt)
+ {
+ Idx node = init_state->nodes.elems[node_cnt];
+ re_token_type_t type = dfa->nodes[node].type;
+
+ if (type == CHARACTER)
+ {
+ re_set_fastmap (fastmap, icase, dfa->nodes[node].opr.c);
+#ifdef RE_ENABLE_I18N
+ if ((bufp->syntax & RE_ICASE) && dfa->mb_cur_max > 1)
+ {
+ unsigned char buf[MB_LEN_MAX];
+ unsigned char *p;
+ wchar_t wc;
+ mbstate_t state;
+
+ p = buf;
+ *p++ = dfa->nodes[node].opr.c;
+ while (++node < dfa->nodes_len
+ && dfa->nodes[node].type == CHARACTER
+ && dfa->nodes[node].mb_partial)
+ *p++ = dfa->nodes[node].opr.c;
+ memset (&state, '\0', sizeof (state));
+ if (__mbrtowc (&wc, (const char *) buf, p - buf,
+ &state) == p - buf
+ && (__wcrtomb ((char *) buf, __towlower (wc), &state)
+ != (size_t) -1))
+ re_set_fastmap (fastmap, false, buf[0]);
+ }
+#endif
+ }
+ else if (type == SIMPLE_BRACKET)
+ {
+ int i, ch;
+ for (i = 0, ch = 0; i < BITSET_WORDS; ++i)
+ {
+ int j;
+ bitset_word_t w = dfa->nodes[node].opr.sbcset[i];
+ for (j = 0; j < BITSET_WORD_BITS; ++j, ++ch)
+ if (w & ((bitset_word_t) 1 << j))
+ re_set_fastmap (fastmap, icase, ch);
+ }
+ }
+#ifdef RE_ENABLE_I18N
+ else if (type == COMPLEX_BRACKET)
+ {
+ re_charset_t *cset = dfa->nodes[node].opr.mbcset;
+ Idx i;
+
+# ifdef _LIBC
+ /* See if we have to try all bytes which start multiple collation
+ elements.
+ e.g. In da_DK, we want to catch 'a' since "aa" is a valid
+ collation element, and don't catch 'b' since 'b' is
+ the only collation element which starts from 'b' (and
+ it is caught by SIMPLE_BRACKET). */
+ if (_NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES) != 0
+ && (cset->ncoll_syms || cset->nranges))
+ {
+ const int32_t *table = (const int32_t *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB);
+ for (i = 0; i < SBC_MAX; ++i)
+ if (table[i] < 0)
+ re_set_fastmap (fastmap, icase, i);
+ }
+# endif /* _LIBC */
+
+ /* See if we have to start the match at all multibyte characters,
+ i.e. where we would not find an invalid sequence. This only
+ applies to multibyte character sets; for single byte character
+ sets, the SIMPLE_BRACKET again suffices. */
+ if (dfa->mb_cur_max > 1
+ && (cset->nchar_classes || cset->non_match || cset->nranges
+# ifdef _LIBC
+ || cset->nequiv_classes
+# endif /* _LIBC */
+ ))
+ {
+ unsigned char c = 0;
+ do
+ {
+ mbstate_t mbs;
+ memset (&mbs, 0, sizeof (mbs));
+ if (__mbrtowc (NULL, (char *) &c, 1, &mbs) == (size_t) -2)
+ re_set_fastmap (fastmap, false, (int) c);
+ }
+ while (++c != 0);
+ }
+
+ else
+ {
+ /* ... Else catch all bytes which can start the mbchars. */
+ for (i = 0; i < cset->nmbchars; ++i)
+ {
+ char buf[256];
+ mbstate_t state;
+ memset (&state, '\0', sizeof (state));
+ if (__wcrtomb (buf, cset->mbchars[i], &state) != (size_t) -1)
+ re_set_fastmap (fastmap, icase, *(unsigned char *) buf);
+ if ((bufp->syntax & RE_ICASE) && dfa->mb_cur_max > 1)
+ {
+ if (__wcrtomb (buf, __towlower (cset->mbchars[i]), &state)
+ != (size_t) -1)
+ re_set_fastmap (fastmap, false, *(unsigned char *) buf);
+ }
+ }
+ }
+ }
+#endif /* RE_ENABLE_I18N */
+ else if (type == OP_PERIOD
+#ifdef RE_ENABLE_I18N
+ || type == OP_UTF8_PERIOD
+#endif /* RE_ENABLE_I18N */
+ || type == END_OF_RE)
+ {
+ memset (fastmap, '\1', sizeof (char) * SBC_MAX);
+ if (type == END_OF_RE)
+ bufp->can_be_null = 1;
+ return;
+ }
+ }
+}
+
+/* Entry point for POSIX code. */
+/* regcomp takes a regular expression as a string and compiles it.
+
+ PREG is a regex_t *. We do not expect any fields to be initialized,
+ since POSIX says we shouldn't. Thus, we set
+
+ 'buffer' to the compiled pattern;
+ 'used' to the length of the compiled pattern;
+ 'syntax' to RE_SYNTAX_POSIX_EXTENDED if the
+ REG_EXTENDED bit in CFLAGS is set; otherwise, to
+ RE_SYNTAX_POSIX_BASIC;
+ 'newline_anchor' to REG_NEWLINE being set in CFLAGS;
+ 'fastmap' to an allocated space for the fastmap;
+ 'fastmap_accurate' to zero;
+ 're_nsub' to the number of subexpressions in PATTERN.
+
+ PATTERN is the address of the pattern string.
+
+ CFLAGS is a series of bits which affect compilation.
+
+ If REG_EXTENDED is set, we use POSIX extended syntax; otherwise, we
+ use POSIX basic syntax.
+
+ If REG_NEWLINE is set, then . and [^...] don't match newline.
+ Also, regexec will try a match beginning after every newline.
+
+ If REG_ICASE is set, then we considers upper- and lowercase
+ versions of letters to be equivalent when matching.
+
+ If REG_NOSUB is set, then when PREG is passed to regexec, that
+ routine will report only success or failure, and nothing about the
+ registers.
+
+ It returns 0 if it succeeds, nonzero if it doesn't. (See regex.h for
+ the return codes and their meanings.) */
+
+int
+regcomp (regex_t *_Restrict_ preg, const char *_Restrict_ pattern, int cflags)
+{
+ reg_errcode_t ret;
+ reg_syntax_t syntax = ((cflags & REG_EXTENDED) ? RE_SYNTAX_POSIX_EXTENDED
+ : RE_SYNTAX_POSIX_BASIC);
+
+ preg->buffer = NULL;
+ preg->allocated = 0;
+ preg->used = 0;
+
+ /* Try to allocate space for the fastmap. */
+ preg->fastmap = re_malloc (char, SBC_MAX);
+ if (BE (preg->fastmap == NULL, 0))
+ return REG_ESPACE;
+
+ syntax |= (cflags & REG_ICASE) ? RE_ICASE : 0;
+
+ /* If REG_NEWLINE is set, newlines are treated differently. */
+ if (cflags & REG_NEWLINE)
+ { /* REG_NEWLINE implies neither . nor [^...] match newline. */
+ syntax &= ~RE_DOT_NEWLINE;
+ syntax |= RE_HAT_LISTS_NOT_NEWLINE;
+ /* It also changes the matching behavior. */
+ preg->newline_anchor = 1;
+ }
+ else
+ preg->newline_anchor = 0;
+ preg->no_sub = !!(cflags & REG_NOSUB);
+ preg->translate = NULL;
+
+ ret = re_compile_internal (preg, pattern, strlen (pattern), syntax);
+
+ /* POSIX doesn't distinguish between an unmatched open-group and an
+ unmatched close-group: both are REG_EPAREN. */
+ if (ret == REG_ERPAREN)
+ ret = REG_EPAREN;
+
+ /* We have already checked preg->fastmap != NULL. */
+ if (BE (ret == REG_NOERROR, 1))
+ /* Compute the fastmap now, since regexec cannot modify the pattern
+ buffer. This function never fails in this implementation. */
+ (void) re_compile_fastmap (preg);
+ else
+ {
+ /* Some error occurred while compiling the expression. */
+ re_free (preg->fastmap);
+ preg->fastmap = NULL;
+ }
+
+ return (int) ret;
+}
+#ifdef _LIBC
+libc_hidden_def (__regcomp)
+weak_alias (__regcomp, regcomp)
+#endif
+
+/* Returns a message corresponding to an error code, ERRCODE, returned
+ from either regcomp or regexec. We don't use PREG here. */
+
+size_t
+regerror (int errcode, const regex_t *_Restrict_ preg, char *_Restrict_ errbuf,
+ size_t errbuf_size)
+{
+ const char *msg;
+ size_t msg_size;
+
+ if (BE (errcode < 0
+ || errcode >= (int) (sizeof (__re_error_msgid_idx)
+ / sizeof (__re_error_msgid_idx[0])), 0))
+ /* Only error codes returned by the rest of the code should be passed
+ to this routine. If we are given anything else, or if other regex
+ code generates an invalid error code, then the program has a bug.
+ Dump core so we can fix it. */
+ abort ();
+
+ msg = gettext (__re_error_msgid + __re_error_msgid_idx[errcode]);
+
+ msg_size = strlen (msg) + 1; /* Includes the null. */
+
+ if (BE (errbuf_size != 0, 1))
+ {
+ size_t cpy_size = msg_size;
+ if (BE (msg_size > errbuf_size, 0))
+ {
+ cpy_size = errbuf_size - 1;
+ errbuf[cpy_size] = '\0';
+ }
+ memcpy (errbuf, msg, cpy_size);
+ }
+
+ return msg_size;
+}
+#ifdef _LIBC
+weak_alias (__regerror, regerror)
+#endif
+
+
+#ifdef RE_ENABLE_I18N
+/* This static array is used for the map to single-byte characters when
+ UTF-8 is used. Otherwise we would allocate memory just to initialize
+ it the same all the time. UTF-8 is the preferred encoding so this is
+ a worthwhile optimization. */
+static const bitset_t utf8_sb_map =
+{
+ /* Set the first 128 bits. */
+# if defined __GNUC__ && !defined __STRICT_ANSI__
+ [0 ... 0x80 / BITSET_WORD_BITS - 1] = BITSET_WORD_MAX
+# else
+# if 4 * BITSET_WORD_BITS < ASCII_CHARS
+# error "bitset_word_t is narrower than 32 bits"
+# elif 3 * BITSET_WORD_BITS < ASCII_CHARS
+ BITSET_WORD_MAX, BITSET_WORD_MAX, BITSET_WORD_MAX,
+# elif 2 * BITSET_WORD_BITS < ASCII_CHARS
+ BITSET_WORD_MAX, BITSET_WORD_MAX,
+# elif 1 * BITSET_WORD_BITS < ASCII_CHARS
+ BITSET_WORD_MAX,
+# endif
+ (BITSET_WORD_MAX
+ >> (SBC_MAX % BITSET_WORD_BITS == 0
+ ? 0
+ : BITSET_WORD_BITS - SBC_MAX % BITSET_WORD_BITS))
+# endif
+};
+#endif
+
+
+static void
+free_dfa_content (re_dfa_t *dfa)
+{
+ Idx i, j;
+
+ if (dfa->nodes)
+ for (i = 0; i < dfa->nodes_len; ++i)
+ free_token (dfa->nodes + i);
+ re_free (dfa->nexts);
+ for (i = 0; i < dfa->nodes_len; ++i)
+ {
+ if (dfa->eclosures != NULL)
+ re_node_set_free (dfa->eclosures + i);
+ if (dfa->inveclosures != NULL)
+ re_node_set_free (dfa->inveclosures + i);
+ if (dfa->edests != NULL)
+ re_node_set_free (dfa->edests + i);
+ }
+ re_free (dfa->edests);
+ re_free (dfa->eclosures);
+ re_free (dfa->inveclosures);
+ re_free (dfa->nodes);
+
+ if (dfa->state_table)
+ for (i = 0; i <= dfa->state_hash_mask; ++i)
+ {
+ struct re_state_table_entry *entry = dfa->state_table + i;
+ for (j = 0; j < entry->num; ++j)
+ {
+ re_dfastate_t *state = entry->array[j];
+ free_state (state);
+ }
+ re_free (entry->array);
+ }
+ re_free (dfa->state_table);
+#ifdef RE_ENABLE_I18N
+ if (dfa->sb_char != utf8_sb_map)
+ re_free (dfa->sb_char);
+#endif
+ re_free (dfa->subexp_map);
+#ifdef DEBUG
+ re_free (dfa->re_str);
+#endif
+
+ re_free (dfa);
+}
+
+
+/* Free dynamically allocated space used by PREG. */
+
+void
+regfree (regex_t *preg)
+{
+ re_dfa_t *dfa = preg->buffer;
+ if (BE (dfa != NULL, 1))
+ {
+ lock_fini (dfa->lock);
+ free_dfa_content (dfa);
+ }
+ preg->buffer = NULL;
+ preg->allocated = 0;
+
+ re_free (preg->fastmap);
+ preg->fastmap = NULL;
+
+ re_free (preg->translate);
+ preg->translate = NULL;
+}
+#ifdef _LIBC
+libc_hidden_def (__regfree)
+weak_alias (__regfree, regfree)
+#endif
+
+/* Entry points compatible with 4.2 BSD regex library. We don't define
+ them unless specifically requested. */
+
+#if defined _REGEX_RE_COMP || defined _LIBC
+
+/* BSD has one and only one pattern buffer. */
+static struct re_pattern_buffer re_comp_buf;
+
+char *
+# ifdef _LIBC
+/* Make these definitions weak in libc, so POSIX programs can redefine
+ these names if they don't use our functions, and still use
+ regcomp/regexec above without link errors. */
+weak_function
+# endif
+re_comp (const char *s)
+{
+ reg_errcode_t ret;
+ char *fastmap;
+
+ if (!s)
+ {
+ if (!re_comp_buf.buffer)
+ return gettext ("No previous regular expression");
+ return 0;
+ }
+
+ if (re_comp_buf.buffer)
+ {
+ fastmap = re_comp_buf.fastmap;
+ re_comp_buf.fastmap = NULL;
+ __regfree (&re_comp_buf);
+ memset (&re_comp_buf, '\0', sizeof (re_comp_buf));
+ re_comp_buf.fastmap = fastmap;
+ }
+
+ if (re_comp_buf.fastmap == NULL)
+ {
+ re_comp_buf.fastmap = re_malloc (char, SBC_MAX);
+ if (re_comp_buf.fastmap == NULL)
+ return (char *) gettext (__re_error_msgid
+ + __re_error_msgid_idx[(int) REG_ESPACE]);
+ }
+
+ /* Since 're_exec' always passes NULL for the 'regs' argument, we
+ don't need to initialize the pattern buffer fields which affect it. */
+
+ /* Match anchors at newlines. */
+ re_comp_buf.newline_anchor = 1;
+
+ ret = re_compile_internal (&re_comp_buf, s, strlen (s), re_syntax_options);
+
+ if (!ret)
+ return NULL;
+
+ /* Yes, we're discarding 'const' here if !HAVE_LIBINTL. */
+ return (char *) gettext (__re_error_msgid + __re_error_msgid_idx[(int) ret]);
+}
+
+#ifdef _LIBC
+libc_freeres_fn (free_mem)
+{
+ __regfree (&re_comp_buf);
+}
+#endif
+
+#endif /* _REGEX_RE_COMP */
+
+/* Internal entry point.
+ Compile the regular expression PATTERN, whose length is LENGTH.
+ SYNTAX indicate regular expression's syntax. */
+
+static reg_errcode_t
+re_compile_internal (regex_t *preg, const char * pattern, size_t length,
+ reg_syntax_t syntax)
+{
+ reg_errcode_t err = REG_NOERROR;
+ re_dfa_t *dfa;
+ re_string_t regexp;
+
+ /* Initialize the pattern buffer. */
+ preg->fastmap_accurate = 0;
+ preg->syntax = syntax;
+ preg->not_bol = preg->not_eol = 0;
+ preg->used = 0;
+ preg->re_nsub = 0;
+ preg->can_be_null = 0;
+ preg->regs_allocated = REGS_UNALLOCATED;
+
+ /* Initialize the dfa. */
+ dfa = preg->buffer;
+ if (BE (preg->allocated < sizeof (re_dfa_t), 0))
+ {
+ /* If zero allocated, but buffer is non-null, try to realloc
+ enough space. This loses if buffer's address is bogus, but
+ that is the user's responsibility. If ->buffer is NULL this
+ is a simple allocation. */
+ dfa = re_realloc (preg->buffer, re_dfa_t, 1);
+ if (dfa == NULL)
+ return REG_ESPACE;
+ preg->allocated = sizeof (re_dfa_t);
+ preg->buffer = dfa;
+ }
+ preg->used = sizeof (re_dfa_t);
+
+ err = init_dfa (dfa, length);
+ if (BE (err == REG_NOERROR && lock_init (dfa->lock) != 0, 0))
+ err = REG_ESPACE;
+ if (BE (err != REG_NOERROR, 0))
+ {
+ free_dfa_content (dfa);
+ preg->buffer = NULL;
+ preg->allocated = 0;
+ return err;
+ }
+#ifdef DEBUG
+ /* Note: length+1 will not overflow since it is checked in init_dfa. */
+ dfa->re_str = re_malloc (char, length + 1);
+ strncpy (dfa->re_str, pattern, length + 1);
+#endif
+
+ err = re_string_construct (®exp, pattern, length, preg->translate,
+ (syntax & RE_ICASE) != 0, dfa);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_compile_internal_free_return:
+ free_workarea_compile (preg);
+ re_string_destruct (®exp);
+ lock_fini (dfa->lock);
+ free_dfa_content (dfa);
+ preg->buffer = NULL;
+ preg->allocated = 0;
+ return err;
+ }
+
+ /* Parse the regular expression, and build a structure tree. */
+ preg->re_nsub = 0;
+ dfa->str_tree = parse (®exp, preg, syntax, &err);
+ if (BE (dfa->str_tree == NULL, 0))
+ goto re_compile_internal_free_return;
+
+ /* Analyze the tree and create the nfa. */
+ err = analyze (preg);
+ if (BE (err != REG_NOERROR, 0))
+ goto re_compile_internal_free_return;
+
+#ifdef RE_ENABLE_I18N
+ /* If possible, do searching in single byte encoding to speed things up. */
+ if (dfa->is_utf8 && !(syntax & RE_ICASE) && preg->translate == NULL)
+ optimize_utf8 (dfa);
+#endif
+
+ /* Then create the initial state of the dfa. */
+ err = create_initial_state (dfa);
+
+ /* Release work areas. */
+ free_workarea_compile (preg);
+ re_string_destruct (®exp);
+
+ if (BE (err != REG_NOERROR, 0))
+ {
+ lock_fini (dfa->lock);
+ free_dfa_content (dfa);
+ preg->buffer = NULL;
+ preg->allocated = 0;
+ }
+
+ return err;
+}
+
+/* Initialize DFA. We use the length of the regular expression PAT_LEN
+ as the initial length of some arrays. */
+
+static reg_errcode_t
+init_dfa (re_dfa_t *dfa, size_t pat_len)
+{
+ __re_size_t table_size;
+#ifndef _LIBC
+ const char *codeset_name;
+#endif
+#ifdef RE_ENABLE_I18N
+ size_t max_i18n_object_size = MAX (sizeof (wchar_t), sizeof (wctype_t));
+#else
+ size_t max_i18n_object_size = 0;
+#endif
+ size_t max_object_size =
+ MAX (sizeof (struct re_state_table_entry),
+ MAX (sizeof (re_token_t),
+ MAX (sizeof (re_node_set),
+ MAX (sizeof (regmatch_t),
+ max_i18n_object_size))));
+
+ memset (dfa, '\0', sizeof (re_dfa_t));
+
+ /* Force allocation of str_tree_storage the first time. */
+ dfa->str_tree_storage_idx = BIN_TREE_STORAGE_SIZE;
+
+ /* Avoid overflows. The extra "/ 2" is for the table_size doubling
+ calculation below, and for similar doubling calculations
+ elsewhere. And it's <= rather than <, because some of the
+ doubling calculations add 1 afterwards. */
+ if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) / 2 <= pat_len, 0))
+ return REG_ESPACE;
+
+ dfa->nodes_alloc = pat_len + 1;
+ dfa->nodes = re_malloc (re_token_t, dfa->nodes_alloc);
+
+ /* table_size = 2 ^ ceil(log pat_len) */
+ for (table_size = 1; ; table_size <<= 1)
+ if (table_size > pat_len)
+ break;
+
+ dfa->state_table = calloc (sizeof (struct re_state_table_entry), table_size);
+ dfa->state_hash_mask = table_size - 1;
+
+ dfa->mb_cur_max = MB_CUR_MAX;
+#ifdef _LIBC
+ if (dfa->mb_cur_max == 6
+ && strcmp (_NL_CURRENT (LC_CTYPE, _NL_CTYPE_CODESET_NAME), "UTF-8") == 0)
+ dfa->is_utf8 = 1;
+ dfa->map_notascii = (_NL_CURRENT_WORD (LC_CTYPE, _NL_CTYPE_MAP_TO_NONASCII)
+ != 0);
+#else
+ codeset_name = nl_langinfo (CODESET);
+ if ((codeset_name[0] == 'U' || codeset_name[0] == 'u')
+ && (codeset_name[1] == 'T' || codeset_name[1] == 't')
+ && (codeset_name[2] == 'F' || codeset_name[2] == 'f')
+ && strcmp (codeset_name + 3 + (codeset_name[3] == '-'), "8") == 0)
+ dfa->is_utf8 = 1;
+
+ /* We check exhaustively in the loop below if this charset is a
+ superset of ASCII. */
+ dfa->map_notascii = 0;
+#endif
+
+#ifdef RE_ENABLE_I18N
+ if (dfa->mb_cur_max > 1)
+ {
+ if (dfa->is_utf8)
+ dfa->sb_char = (re_bitset_ptr_t) utf8_sb_map;
+ else
+ {
+ int i, j, ch;
+
+ dfa->sb_char = (re_bitset_ptr_t) calloc (sizeof (bitset_t), 1);
+ if (BE (dfa->sb_char == NULL, 0))
+ return REG_ESPACE;
+
+ /* Set the bits corresponding to single byte chars. */
+ for (i = 0, ch = 0; i < BITSET_WORDS; ++i)
+ for (j = 0; j < BITSET_WORD_BITS; ++j, ++ch)
+ {
+ wint_t wch = __btowc (ch);
+ if (wch != WEOF)
+ dfa->sb_char[i] |= (bitset_word_t) 1 << j;
+# ifndef _LIBC
+ if (isascii (ch) && wch != ch)
+ dfa->map_notascii = 1;
+# endif
+ }
+ }
+ }
+#endif
+
+ if (BE (dfa->nodes == NULL || dfa->state_table == NULL, 0))
+ return REG_ESPACE;
+ return REG_NOERROR;
+}
+
+/* Initialize WORD_CHAR table, which indicate which character is
+ "word". In this case "word" means that it is the word construction
+ character used by some operators like "\<", "\>", etc. */
+
+static void
+init_word_char (re_dfa_t *dfa)
+{
+ int i = 0;
+ int j;
+ int ch = 0;
+ dfa->word_ops_used = 1;
+ if (BE (dfa->map_notascii == 0, 1))
+ {
+ /* Avoid uint32_t and uint64_t as some non-GCC platforms lack
+ them, an issue when this code is used in Gnulib. */
+ bitset_word_t bits0 = 0x00000000;
+ bitset_word_t bits1 = 0x03ff0000;
+ bitset_word_t bits2 = 0x87fffffe;
+ bitset_word_t bits3 = 0x07fffffe;
+ if (BITSET_WORD_BITS == 64)
+ {
+ /* Pacify gcc -Woverflow on 32-bit platformns. */
+ dfa->word_char[0] = bits1 << 31 << 1 | bits0;
+ dfa->word_char[1] = bits3 << 31 << 1 | bits2;
+ i = 2;
+ }
+ else if (BITSET_WORD_BITS == 32)
+ {
+ dfa->word_char[0] = bits0;
+ dfa->word_char[1] = bits1;
+ dfa->word_char[2] = bits2;
+ dfa->word_char[3] = bits3;
+ i = 4;
+ }
+ else
+ goto general_case;
+ ch = 128;
+
+ if (BE (dfa->is_utf8, 1))
+ {
+ memset (&dfa->word_char[i], '\0', (SBC_MAX - ch) / 8);
+ return;
+ }
+ }
+
+ general_case:
+ for (; i < BITSET_WORDS; ++i)
+ for (j = 0; j < BITSET_WORD_BITS; ++j, ++ch)
+ if (isalnum (ch) || ch == '_')
+ dfa->word_char[i] |= (bitset_word_t) 1 << j;
+}
+
+/* Free the work area which are only used while compiling. */
+
+static void
+free_workarea_compile (regex_t *preg)
+{
+ re_dfa_t *dfa = preg->buffer;
+ bin_tree_storage_t *storage, *next;
+ for (storage = dfa->str_tree_storage; storage; storage = next)
+ {
+ next = storage->next;
+ re_free (storage);
+ }
+ dfa->str_tree_storage = NULL;
+ dfa->str_tree_storage_idx = BIN_TREE_STORAGE_SIZE;
+ dfa->str_tree = NULL;
+ re_free (dfa->org_indices);
+ dfa->org_indices = NULL;
+}
+
+/* Create initial states for all contexts. */
+
+static reg_errcode_t
+create_initial_state (re_dfa_t *dfa)
+{
+ Idx first, i;
+ reg_errcode_t err;
+ re_node_set init_nodes;
+
+ /* Initial states have the epsilon closure of the node which is
+ the first node of the regular expression. */
+ first = dfa->str_tree->first->node_idx;
+ dfa->init_node = first;
+ err = re_node_set_init_copy (&init_nodes, dfa->eclosures + first);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+
+ /* The back-references which are in initial states can epsilon transit,
+ since in this case all of the subexpressions can be null.
+ Then we add epsilon closures of the nodes which are the next nodes of
+ the back-references. */
+ if (dfa->nbackref > 0)
+ for (i = 0; i < init_nodes.nelem; ++i)
+ {
+ Idx node_idx = init_nodes.elems[i];
+ re_token_type_t type = dfa->nodes[node_idx].type;
+
+ Idx clexp_idx;
+ if (type != OP_BACK_REF)
+ continue;
+ for (clexp_idx = 0; clexp_idx < init_nodes.nelem; ++clexp_idx)
+ {
+ re_token_t *clexp_node;
+ clexp_node = dfa->nodes + init_nodes.elems[clexp_idx];
+ if (clexp_node->type == OP_CLOSE_SUBEXP
+ && clexp_node->opr.idx == dfa->nodes[node_idx].opr.idx)
+ break;
+ }
+ if (clexp_idx == init_nodes.nelem)
+ continue;
+
+ if (type == OP_BACK_REF)
+ {
+ Idx dest_idx = dfa->edests[node_idx].elems[0];
+ if (!re_node_set_contains (&init_nodes, dest_idx))
+ {
+ reg_errcode_t merge_err
+ = re_node_set_merge (&init_nodes, dfa->eclosures + dest_idx);
+ if (merge_err != REG_NOERROR)
+ return merge_err;
+ i = 0;
+ }
+ }
+ }
+
+ /* It must be the first time to invoke acquire_state. */
+ dfa->init_state = re_acquire_state_context (&err, dfa, &init_nodes, 0);
+ /* We don't check ERR here, since the initial state must not be NULL. */
+ if (BE (dfa->init_state == NULL, 0))
+ return err;
+ if (dfa->init_state->has_constraint)
+ {
+ dfa->init_state_word = re_acquire_state_context (&err, dfa, &init_nodes,
+ CONTEXT_WORD);
+ dfa->init_state_nl = re_acquire_state_context (&err, dfa, &init_nodes,
+ CONTEXT_NEWLINE);
+ dfa->init_state_begbuf = re_acquire_state_context (&err, dfa,
+ &init_nodes,
+ CONTEXT_NEWLINE
+ | CONTEXT_BEGBUF);
+ if (BE (dfa->init_state_word == NULL || dfa->init_state_nl == NULL
+ || dfa->init_state_begbuf == NULL, 0))
+ return err;
+ }
+ else
+ dfa->init_state_word = dfa->init_state_nl
+ = dfa->init_state_begbuf = dfa->init_state;
+
+ re_node_set_free (&init_nodes);
+ return REG_NOERROR;
+}
+
+#ifdef RE_ENABLE_I18N
+/* If it is possible to do searching in single byte encoding instead of UTF-8
+ to speed things up, set dfa->mb_cur_max to 1, clear is_utf8 and change
+ DFA nodes where needed. */
+
+static void
+optimize_utf8 (re_dfa_t *dfa)
+{
+ Idx node;
+ int i;
+ bool mb_chars = false;
+ bool has_period = false;
+
+ for (node = 0; node < dfa->nodes_len; ++node)
+ switch (dfa->nodes[node].type)
+ {
+ case CHARACTER:
+ if (dfa->nodes[node].opr.c >= ASCII_CHARS)
+ mb_chars = true;
+ break;
+ case ANCHOR:
+ switch (dfa->nodes[node].opr.ctx_type)
+ {
+ case LINE_FIRST:
+ case LINE_LAST:
+ case BUF_FIRST:
+ case BUF_LAST:
+ break;
+ default:
+ /* Word anchors etc. cannot be handled. It's okay to test
+ opr.ctx_type since constraints (for all DFA nodes) are
+ created by ORing one or more opr.ctx_type values. */
+ return;
+ }
+ break;
+ case OP_PERIOD:
+ has_period = true;
+ break;
+ case OP_BACK_REF:
+ case OP_ALT:
+ case END_OF_RE:
+ case OP_DUP_ASTERISK:
+ case OP_OPEN_SUBEXP:
+ case OP_CLOSE_SUBEXP:
+ break;
+ case COMPLEX_BRACKET:
+ return;
+ case SIMPLE_BRACKET:
+ /* Just double check. */
+ {
+ int rshift = (ASCII_CHARS % BITSET_WORD_BITS == 0
+ ? 0
+ : BITSET_WORD_BITS - ASCII_CHARS % BITSET_WORD_BITS);
+ for (i = ASCII_CHARS / BITSET_WORD_BITS; i < BITSET_WORDS; ++i)
+ {
+ if (dfa->nodes[node].opr.sbcset[i] >> rshift != 0)
+ return;
+ rshift = 0;
+ }
+ }
+ break;
+ default:
+ abort ();
+ }
+
+ if (mb_chars || has_period)
+ for (node = 0; node < dfa->nodes_len; ++node)
+ {
+ if (dfa->nodes[node].type == CHARACTER
+ && dfa->nodes[node].opr.c >= ASCII_CHARS)
+ dfa->nodes[node].mb_partial = 0;
+ else if (dfa->nodes[node].type == OP_PERIOD)
+ dfa->nodes[node].type = OP_UTF8_PERIOD;
+ }
+
+ /* The search can be in single byte locale. */
+ dfa->mb_cur_max = 1;
+ dfa->is_utf8 = 0;
+ dfa->has_mb_node = dfa->nbackref > 0 || has_period;
+}
+#endif
+
+/* Analyze the structure tree, and calculate "first", "next", "edest",
+ "eclosure", and "inveclosure". */
+
+static reg_errcode_t
+analyze (regex_t *preg)
+{
+ re_dfa_t *dfa = preg->buffer;
+ reg_errcode_t ret;
+
+ /* Allocate arrays. */
+ dfa->nexts = re_malloc (Idx, dfa->nodes_alloc);
+ dfa->org_indices = re_malloc (Idx, dfa->nodes_alloc);
+ dfa->edests = re_malloc (re_node_set, dfa->nodes_alloc);
+ dfa->eclosures = re_malloc (re_node_set, dfa->nodes_alloc);
+ if (BE (dfa->nexts == NULL || dfa->org_indices == NULL || dfa->edests == NULL
+ || dfa->eclosures == NULL, 0))
+ return REG_ESPACE;
+
+ dfa->subexp_map = re_malloc (Idx, preg->re_nsub);
+ if (dfa->subexp_map != NULL)
+ {
+ Idx i;
+ for (i = 0; i < preg->re_nsub; i++)
+ dfa->subexp_map[i] = i;
+ preorder (dfa->str_tree, optimize_subexps, dfa);
+ for (i = 0; i < preg->re_nsub; i++)
+ if (dfa->subexp_map[i] != i)
+ break;
+ if (i == preg->re_nsub)
+ {
+ re_free (dfa->subexp_map);
+ dfa->subexp_map = NULL;
+ }
+ }
+
+ ret = postorder (dfa->str_tree, lower_subexps, preg);
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+ ret = postorder (dfa->str_tree, calc_first, dfa);
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+ preorder (dfa->str_tree, calc_next, dfa);
+ ret = preorder (dfa->str_tree, link_nfa_nodes, dfa);
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+ ret = calc_eclosure (dfa);
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+
+ /* We only need this during the prune_impossible_nodes pass in regexec.c;
+ skip it if p_i_n will not run, as calc_inveclosure can be quadratic. */
+ if ((!preg->no_sub && preg->re_nsub > 0 && dfa->has_plural_match)
+ || dfa->nbackref)
+ {
+ dfa->inveclosures = re_malloc (re_node_set, dfa->nodes_len);
+ if (BE (dfa->inveclosures == NULL, 0))
+ return REG_ESPACE;
+ ret = calc_inveclosure (dfa);
+ }
+
+ return ret;
+}
+
+/* Our parse trees are very unbalanced, so we cannot use a stack to
+ implement parse tree visits. Instead, we use parent pointers and
+ some hairy code in these two functions. */
+static reg_errcode_t
+postorder (bin_tree_t *root, reg_errcode_t (fn (void *, bin_tree_t *)),
+ void *extra)
+{
+ bin_tree_t *node, *prev;
+
+ for (node = root; ; )
+ {
+ /* Descend down the tree, preferably to the left (or to the right
+ if that's the only child). */
+ while (node->left || node->right)
+ if (node->left)
+ node = node->left;
+ else
+ node = node->right;
+
+ do
+ {
+ reg_errcode_t err = fn (extra, node);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ if (node->parent == NULL)
+ return REG_NOERROR;
+ prev = node;
+ node = node->parent;
+ }
+ /* Go up while we have a node that is reached from the right. */
+ while (node->right == prev || node->right == NULL);
+ node = node->right;
+ }
+}
+
+static reg_errcode_t
+preorder (bin_tree_t *root, reg_errcode_t (fn (void *, bin_tree_t *)),
+ void *extra)
+{
+ bin_tree_t *node;
+
+ for (node = root; ; )
+ {
+ reg_errcode_t err = fn (extra, node);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+
+ /* Go to the left node, or up and to the right. */
+ if (node->left)
+ node = node->left;
+ else
+ {
+ bin_tree_t *prev = NULL;
+ while (node->right == prev || node->right == NULL)
+ {
+ prev = node;
+ node = node->parent;
+ if (!node)
+ return REG_NOERROR;
+ }
+ node = node->right;
+ }
+ }
+}
+
+/* Optimization pass: if a SUBEXP is entirely contained, strip it and tell
+ re_search_internal to map the inner one's opr.idx to this one's. Adjust
+ backreferences as well. Requires a preorder visit. */
+static reg_errcode_t
+optimize_subexps (void *extra, bin_tree_t *node)
+{
+ re_dfa_t *dfa = (re_dfa_t *) extra;
+
+ if (node->token.type == OP_BACK_REF && dfa->subexp_map)
+ {
+ int idx = node->token.opr.idx;
+ node->token.opr.idx = dfa->subexp_map[idx];
+ dfa->used_bkref_map |= 1 << node->token.opr.idx;
+ }
+
+ else if (node->token.type == SUBEXP
+ && node->left && node->left->token.type == SUBEXP)
+ {
+ Idx other_idx = node->left->token.opr.idx;
+
+ node->left = node->left->left;
+ if (node->left)
+ node->left->parent = node;
+
+ dfa->subexp_map[other_idx] = dfa->subexp_map[node->token.opr.idx];
+ if (other_idx < BITSET_WORD_BITS)
+ dfa->used_bkref_map &= ~((bitset_word_t) 1 << other_idx);
+ }
+
+ return REG_NOERROR;
+}
+
+/* Lowering pass: Turn each SUBEXP node into the appropriate concatenation
+ of OP_OPEN_SUBEXP, the body of the SUBEXP (if any) and OP_CLOSE_SUBEXP. */
+static reg_errcode_t
+lower_subexps (void *extra, bin_tree_t *node)
+{
+ regex_t *preg = (regex_t *) extra;
+ reg_errcode_t err = REG_NOERROR;
+
+ if (node->left && node->left->token.type == SUBEXP)
+ {
+ node->left = lower_subexp (&err, preg, node->left);
+ if (node->left)
+ node->left->parent = node;
+ }
+ if (node->right && node->right->token.type == SUBEXP)
+ {
+ node->right = lower_subexp (&err, preg, node->right);
+ if (node->right)
+ node->right->parent = node;
+ }
+
+ return err;
+}
+
+static bin_tree_t *
+lower_subexp (reg_errcode_t *err, regex_t *preg, bin_tree_t *node)
+{
+ re_dfa_t *dfa = preg->buffer;
+ bin_tree_t *body = node->left;
+ bin_tree_t *op, *cls, *tree1, *tree;
+
+ if (preg->no_sub
+ /* We do not optimize empty subexpressions, because otherwise we may
+ have bad CONCAT nodes with NULL children. This is obviously not
+ very common, so we do not lose much. An example that triggers
+ this case is the sed "script" /\(\)/x. */
+ && node->left != NULL
+ && (node->token.opr.idx >= BITSET_WORD_BITS
+ || !(dfa->used_bkref_map
+ & ((bitset_word_t) 1 << node->token.opr.idx))))
+ return node->left;
+
+ /* Convert the SUBEXP node to the concatenation of an
+ OP_OPEN_SUBEXP, the contents, and an OP_CLOSE_SUBEXP. */
+ op = create_tree (dfa, NULL, NULL, OP_OPEN_SUBEXP);
+ cls = create_tree (dfa, NULL, NULL, OP_CLOSE_SUBEXP);
+ tree1 = body ? create_tree (dfa, body, cls, CONCAT) : cls;
+ tree = create_tree (dfa, op, tree1, CONCAT);
+ if (BE (tree == NULL || tree1 == NULL || op == NULL || cls == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+
+ op->token.opr.idx = cls->token.opr.idx = node->token.opr.idx;
+ op->token.opt_subexp = cls->token.opt_subexp = node->token.opt_subexp;
+ return tree;
+}
+
+/* Pass 1 in building the NFA: compute FIRST and create unlinked automaton
+ nodes. Requires a postorder visit. */
+static reg_errcode_t
+calc_first (void *extra, bin_tree_t *node)
+{
+ re_dfa_t *dfa = (re_dfa_t *) extra;
+ if (node->token.type == CONCAT)
+ {
+ node->first = node->left->first;
+ node->node_idx = node->left->node_idx;
+ }
+ else
+ {
+ node->first = node;
+ node->node_idx = re_dfa_add_node (dfa, node->token);
+ if (BE (node->node_idx == -1, 0))
+ return REG_ESPACE;
+ if (node->token.type == ANCHOR)
+ dfa->nodes[node->node_idx].constraint = node->token.opr.ctx_type;
+ }
+ return REG_NOERROR;
+}
+
+/* Pass 2: compute NEXT on the tree. Preorder visit. */
+static reg_errcode_t
+calc_next (void *extra, bin_tree_t *node)
+{
+ switch (node->token.type)
+ {
+ case OP_DUP_ASTERISK:
+ node->left->next = node;
+ break;
+ case CONCAT:
+ node->left->next = node->right->first;
+ node->right->next = node->next;
+ break;
+ default:
+ if (node->left)
+ node->left->next = node->next;
+ if (node->right)
+ node->right->next = node->next;
+ break;
+ }
+ return REG_NOERROR;
+}
+
+/* Pass 3: link all DFA nodes to their NEXT node (any order will do). */
+static reg_errcode_t
+link_nfa_nodes (void *extra, bin_tree_t *node)
+{
+ re_dfa_t *dfa = (re_dfa_t *) extra;
+ Idx idx = node->node_idx;
+ reg_errcode_t err = REG_NOERROR;
+
+ switch (node->token.type)
+ {
+ case CONCAT:
+ break;
+
+ case END_OF_RE:
+ assert (node->next == NULL);
+ break;
+
+ case OP_DUP_ASTERISK:
+ case OP_ALT:
+ {
+ Idx left, right;
+ dfa->has_plural_match = 1;
+ if (node->left != NULL)
+ left = node->left->first->node_idx;
+ else
+ left = node->next->node_idx;
+ if (node->right != NULL)
+ right = node->right->first->node_idx;
+ else
+ right = node->next->node_idx;
+ assert (left > -1);
+ assert (right > -1);
+ err = re_node_set_init_2 (dfa->edests + idx, left, right);
+ }
+ break;
+
+ case ANCHOR:
+ case OP_OPEN_SUBEXP:
+ case OP_CLOSE_SUBEXP:
+ err = re_node_set_init_1 (dfa->edests + idx, node->next->node_idx);
+ break;
+
+ case OP_BACK_REF:
+ dfa->nexts[idx] = node->next->node_idx;
+ if (node->token.type == OP_BACK_REF)
+ err = re_node_set_init_1 (dfa->edests + idx, dfa->nexts[idx]);
+ break;
+
+ default:
+ assert (!IS_EPSILON_NODE (node->token.type));
+ dfa->nexts[idx] = node->next->node_idx;
+ break;
+ }
+
+ return err;
+}
+
+/* Duplicate the epsilon closure of the node ROOT_NODE.
+ Note that duplicated nodes have constraint INIT_CONSTRAINT in addition
+ to their own constraint. */
+
+static reg_errcode_t
+duplicate_node_closure (re_dfa_t *dfa, Idx top_org_node, Idx top_clone_node,
+ Idx root_node, unsigned int init_constraint)
+{
+ Idx org_node, clone_node;
+ bool ok;
+ unsigned int constraint = init_constraint;
+ for (org_node = top_org_node, clone_node = top_clone_node;;)
+ {
+ Idx org_dest, clone_dest;
+ if (dfa->nodes[org_node].type == OP_BACK_REF)
+ {
+ /* If the back reference epsilon-transit, its destination must
+ also have the constraint. Then duplicate the epsilon closure
+ of the destination of the back reference, and store it in
+ edests of the back reference. */
+ org_dest = dfa->nexts[org_node];
+ re_node_set_empty (dfa->edests + clone_node);
+ clone_dest = duplicate_node (dfa, org_dest, constraint);
+ if (BE (clone_dest == -1, 0))
+ return REG_ESPACE;
+ dfa->nexts[clone_node] = dfa->nexts[org_node];
+ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ }
+ else if (dfa->edests[org_node].nelem == 0)
+ {
+ /* In case of the node can't epsilon-transit, don't duplicate the
+ destination and store the original destination as the
+ destination of the node. */
+ dfa->nexts[clone_node] = dfa->nexts[org_node];
+ break;
+ }
+ else if (dfa->edests[org_node].nelem == 1)
+ {
+ /* In case of the node can epsilon-transit, and it has only one
+ destination. */
+ org_dest = dfa->edests[org_node].elems[0];
+ re_node_set_empty (dfa->edests + clone_node);
+ /* If the node is root_node itself, it means the epsilon closure
+ has a loop. Then tie it to the destination of the root_node. */
+ if (org_node == root_node && clone_node != org_node)
+ {
+ ok = re_node_set_insert (dfa->edests + clone_node, org_dest);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ break;
+ }
+ /* In case the node has another constraint, append it. */
+ constraint |= dfa->nodes[org_node].constraint;
+ clone_dest = duplicate_node (dfa, org_dest, constraint);
+ if (BE (clone_dest == -1, 0))
+ return REG_ESPACE;
+ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ }
+ else /* dfa->edests[org_node].nelem == 2 */
+ {
+ /* In case of the node can epsilon-transit, and it has two
+ destinations. In the bin_tree_t and DFA, that's '|' and '*'. */
+ org_dest = dfa->edests[org_node].elems[0];
+ re_node_set_empty (dfa->edests + clone_node);
+ /* Search for a duplicated node which satisfies the constraint. */
+ clone_dest = search_duplicated_node (dfa, org_dest, constraint);
+ if (clone_dest == -1)
+ {
+ /* There is no such duplicated node, create a new one. */
+ reg_errcode_t err;
+ clone_dest = duplicate_node (dfa, org_dest, constraint);
+ if (BE (clone_dest == -1, 0))
+ return REG_ESPACE;
+ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ err = duplicate_node_closure (dfa, org_dest, clone_dest,
+ root_node, constraint);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ else
+ {
+ /* There is a duplicated node which satisfies the constraint,
+ use it to avoid infinite loop. */
+ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ }
+
+ org_dest = dfa->edests[org_node].elems[1];
+ clone_dest = duplicate_node (dfa, org_dest, constraint);
+ if (BE (clone_dest == -1, 0))
+ return REG_ESPACE;
+ ok = re_node_set_insert (dfa->edests + clone_node, clone_dest);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ }
+ org_node = org_dest;
+ clone_node = clone_dest;
+ }
+ return REG_NOERROR;
+}
+
+/* Search for a node which is duplicated from the node ORG_NODE, and
+ satisfies the constraint CONSTRAINT. */
+
+static Idx
+search_duplicated_node (const re_dfa_t *dfa, Idx org_node,
+ unsigned int constraint)
+{
+ Idx idx;
+ for (idx = dfa->nodes_len - 1; dfa->nodes[idx].duplicated && idx > 0; --idx)
+ {
+ if (org_node == dfa->org_indices[idx]
+ && constraint == dfa->nodes[idx].constraint)
+ return idx; /* Found. */
+ }
+ return -1; /* Not found. */
+}
+
+/* Duplicate the node whose index is ORG_IDX and set the constraint CONSTRAINT.
+ Return the index of the new node, or -1 if insufficient storage is
+ available. */
+
+static Idx
+duplicate_node (re_dfa_t *dfa, Idx org_idx, unsigned int constraint)
+{
+ Idx dup_idx = re_dfa_add_node (dfa, dfa->nodes[org_idx]);
+ if (BE (dup_idx != -1, 1))
+ {
+ dfa->nodes[dup_idx].constraint = constraint;
+ dfa->nodes[dup_idx].constraint |= dfa->nodes[org_idx].constraint;
+ dfa->nodes[dup_idx].duplicated = 1;
+
+ /* Store the index of the original node. */
+ dfa->org_indices[dup_idx] = org_idx;
+ }
+ return dup_idx;
+}
+
+static reg_errcode_t
+calc_inveclosure (re_dfa_t *dfa)
+{
+ Idx src, idx;
+ bool ok;
+ for (idx = 0; idx < dfa->nodes_len; ++idx)
+ re_node_set_init_empty (dfa->inveclosures + idx);
+
+ for (src = 0; src < dfa->nodes_len; ++src)
+ {
+ Idx *elems = dfa->eclosures[src].elems;
+ for (idx = 0; idx < dfa->eclosures[src].nelem; ++idx)
+ {
+ ok = re_node_set_insert_last (dfa->inveclosures + elems[idx], src);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ }
+ }
+
+ return REG_NOERROR;
+}
+
+/* Calculate "eclosure" for all the node in DFA. */
+
+static reg_errcode_t
+calc_eclosure (re_dfa_t *dfa)
+{
+ Idx node_idx;
+ bool incomplete;
+#ifdef DEBUG
+ assert (dfa->nodes_len > 0);
+#endif
+ incomplete = false;
+ /* For each nodes, calculate epsilon closure. */
+ for (node_idx = 0; ; ++node_idx)
+ {
+ reg_errcode_t err;
+ re_node_set eclosure_elem;
+ if (node_idx == dfa->nodes_len)
+ {
+ if (!incomplete)
+ break;
+ incomplete = false;
+ node_idx = 0;
+ }
+
+#ifdef DEBUG
+ assert (dfa->eclosures[node_idx].nelem != -1);
+#endif
+
+ /* If we have already calculated, skip it. */
+ if (dfa->eclosures[node_idx].nelem != 0)
+ continue;
+ /* Calculate epsilon closure of 'node_idx'. */
+ err = calc_eclosure_iter (&eclosure_elem, dfa, node_idx, true);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+
+ if (dfa->eclosures[node_idx].nelem == 0)
+ {
+ incomplete = true;
+ re_node_set_free (&eclosure_elem);
+ }
+ }
+ return REG_NOERROR;
+}
+
+/* Calculate epsilon closure of NODE. */
+
+static reg_errcode_t
+calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, Idx node, bool root)
+{
+ reg_errcode_t err;
+ Idx i;
+ re_node_set eclosure;
+ bool ok;
+ bool incomplete = false;
+ err = re_node_set_alloc (&eclosure, dfa->edests[node].nelem + 1);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+
+ /* This indicates that we are calculating this node now.
+ We reference this value to avoid infinite loop. */
+ dfa->eclosures[node].nelem = -1;
+
+ /* If the current node has constraints, duplicate all nodes
+ since they must inherit the constraints. */
+ if (dfa->nodes[node].constraint
+ && dfa->edests[node].nelem
+ && !dfa->nodes[dfa->edests[node].elems[0]].duplicated)
+ {
+ err = duplicate_node_closure (dfa, node, node, node,
+ dfa->nodes[node].constraint);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+
+ /* Expand each epsilon destination nodes. */
+ if (IS_EPSILON_NODE(dfa->nodes[node].type))
+ for (i = 0; i < dfa->edests[node].nelem; ++i)
+ {
+ re_node_set eclosure_elem;
+ Idx edest = dfa->edests[node].elems[i];
+ /* If calculating the epsilon closure of 'edest' is in progress,
+ return intermediate result. */
+ if (dfa->eclosures[edest].nelem == -1)
+ {
+ incomplete = true;
+ continue;
+ }
+ /* If we haven't calculated the epsilon closure of 'edest' yet,
+ calculate now. Otherwise use calculated epsilon closure. */
+ if (dfa->eclosures[edest].nelem == 0)
+ {
+ err = calc_eclosure_iter (&eclosure_elem, dfa, edest, false);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ else
+ eclosure_elem = dfa->eclosures[edest];
+ /* Merge the epsilon closure of 'edest'. */
+ err = re_node_set_merge (&eclosure, &eclosure_elem);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ /* If the epsilon closure of 'edest' is incomplete,
+ the epsilon closure of this node is also incomplete. */
+ if (dfa->eclosures[edest].nelem == 0)
+ {
+ incomplete = true;
+ re_node_set_free (&eclosure_elem);
+ }
+ }
+
+ /* An epsilon closure includes itself. */
+ ok = re_node_set_insert (&eclosure, node);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ if (incomplete && !root)
+ dfa->eclosures[node].nelem = 0;
+ else
+ dfa->eclosures[node] = eclosure;
+ *new_set = eclosure;
+ return REG_NOERROR;
+}
+
+/* Functions for token which are used in the parser. */
+
+/* Fetch a token from INPUT.
+ We must not use this function inside bracket expressions. */
+
+static void
+fetch_token (re_token_t *result, re_string_t *input, reg_syntax_t syntax)
+{
+ re_string_skip_bytes (input, peek_token (result, input, syntax));
+}
+
+/* Peek a token from INPUT, and return the length of the token.
+ We must not use this function inside bracket expressions. */
+
+static int
+peek_token (re_token_t *token, re_string_t *input, reg_syntax_t syntax)
+{
+ unsigned char c;
+
+ if (re_string_eoi (input))
+ {
+ token->type = END_OF_RE;
+ return 0;
+ }
+
+ c = re_string_peek_byte (input, 0);
+ token->opr.c = c;
+
+ token->word_char = 0;
+#ifdef RE_ENABLE_I18N
+ token->mb_partial = 0;
+ if (input->mb_cur_max > 1 &&
+ !re_string_first_byte (input, re_string_cur_idx (input)))
+ {
+ token->type = CHARACTER;
+ token->mb_partial = 1;
+ return 1;
+ }
+#endif
+ if (c == '\\')
+ {
+ unsigned char c2;
+ if (re_string_cur_idx (input) + 1 >= re_string_length (input))
+ {
+ token->type = BACK_SLASH;
+ return 1;
+ }
+
+ c2 = re_string_peek_byte_case (input, 1);
+ token->opr.c = c2;
+ token->type = CHARACTER;
+#ifdef RE_ENABLE_I18N
+ if (input->mb_cur_max > 1)
+ {
+ wint_t wc = re_string_wchar_at (input,
+ re_string_cur_idx (input) + 1);
+ token->word_char = IS_WIDE_WORD_CHAR (wc) != 0;
+ }
+ else
+#endif
+ token->word_char = IS_WORD_CHAR (c2) != 0;
+
+ switch (c2)
+ {
+ case '|':
+ if (!(syntax & RE_LIMITED_OPS) && !(syntax & RE_NO_BK_VBAR))
+ token->type = OP_ALT;
+ break;
+ case '1': case '2': case '3': case '4': case '5':
+ case '6': case '7': case '8': case '9':
+ if (!(syntax & RE_NO_BK_REFS))
+ {
+ token->type = OP_BACK_REF;
+ token->opr.idx = c2 - '1';
+ }
+ break;
+ case '<':
+ if (!(syntax & RE_NO_GNU_OPS))
+ {
+ token->type = ANCHOR;
+ token->opr.ctx_type = WORD_FIRST;
+ }
+ break;
+ case '>':
+ if (!(syntax & RE_NO_GNU_OPS))
+ {
+ token->type = ANCHOR;
+ token->opr.ctx_type = WORD_LAST;
+ }
+ break;
+ case 'b':
+ if (!(syntax & RE_NO_GNU_OPS))
+ {
+ token->type = ANCHOR;
+ token->opr.ctx_type = WORD_DELIM;
+ }
+ break;
+ case 'B':
+ if (!(syntax & RE_NO_GNU_OPS))
+ {
+ token->type = ANCHOR;
+ token->opr.ctx_type = NOT_WORD_DELIM;
+ }
+ break;
+ case 'w':
+ if (!(syntax & RE_NO_GNU_OPS))
+ token->type = OP_WORD;
+ break;
+ case 'W':
+ if (!(syntax & RE_NO_GNU_OPS))
+ token->type = OP_NOTWORD;
+ break;
+ case 's':
+ if (!(syntax & RE_NO_GNU_OPS))
+ token->type = OP_SPACE;
+ break;
+ case 'S':
+ if (!(syntax & RE_NO_GNU_OPS))
+ token->type = OP_NOTSPACE;
+ break;
+ case '`':
+ if (!(syntax & RE_NO_GNU_OPS))
+ {
+ token->type = ANCHOR;
+ token->opr.ctx_type = BUF_FIRST;
+ }
+ break;
+ case '\'':
+ if (!(syntax & RE_NO_GNU_OPS))
+ {
+ token->type = ANCHOR;
+ token->opr.ctx_type = BUF_LAST;
+ }
+ break;
+ case '(':
+ if (!(syntax & RE_NO_BK_PARENS))
+ token->type = OP_OPEN_SUBEXP;
+ break;
+ case ')':
+ if (!(syntax & RE_NO_BK_PARENS))
+ token->type = OP_CLOSE_SUBEXP;
+ break;
+ case '+':
+ if (!(syntax & RE_LIMITED_OPS) && (syntax & RE_BK_PLUS_QM))
+ token->type = OP_DUP_PLUS;
+ break;
+ case '?':
+ if (!(syntax & RE_LIMITED_OPS) && (syntax & RE_BK_PLUS_QM))
+ token->type = OP_DUP_QUESTION;
+ break;
+ case '{':
+ if ((syntax & RE_INTERVALS) && (!(syntax & RE_NO_BK_BRACES)))
+ token->type = OP_OPEN_DUP_NUM;
+ break;
+ case '}':
+ if ((syntax & RE_INTERVALS) && (!(syntax & RE_NO_BK_BRACES)))
+ token->type = OP_CLOSE_DUP_NUM;
+ break;
+ default:
+ break;
+ }
+ return 2;
+ }
+
+ token->type = CHARACTER;
+#ifdef RE_ENABLE_I18N
+ if (input->mb_cur_max > 1)
+ {
+ wint_t wc = re_string_wchar_at (input, re_string_cur_idx (input));
+ token->word_char = IS_WIDE_WORD_CHAR (wc) != 0;
+ }
+ else
+#endif
+ token->word_char = IS_WORD_CHAR (token->opr.c);
+
+ switch (c)
+ {
+ case '\n':
+ if (syntax & RE_NEWLINE_ALT)
+ token->type = OP_ALT;
+ break;
+ case '|':
+ if (!(syntax & RE_LIMITED_OPS) && (syntax & RE_NO_BK_VBAR))
+ token->type = OP_ALT;
+ break;
+ case '*':
+ token->type = OP_DUP_ASTERISK;
+ break;
+ case '+':
+ if (!(syntax & RE_LIMITED_OPS) && !(syntax & RE_BK_PLUS_QM))
+ token->type = OP_DUP_PLUS;
+ break;
+ case '?':
+ if (!(syntax & RE_LIMITED_OPS) && !(syntax & RE_BK_PLUS_QM))
+ token->type = OP_DUP_QUESTION;
+ break;
+ case '{':
+ if ((syntax & RE_INTERVALS) && (syntax & RE_NO_BK_BRACES))
+ token->type = OP_OPEN_DUP_NUM;
+ break;
+ case '}':
+ if ((syntax & RE_INTERVALS) && (syntax & RE_NO_BK_BRACES))
+ token->type = OP_CLOSE_DUP_NUM;
+ break;
+ case '(':
+ if (syntax & RE_NO_BK_PARENS)
+ token->type = OP_OPEN_SUBEXP;
+ break;
+ case ')':
+ if (syntax & RE_NO_BK_PARENS)
+ token->type = OP_CLOSE_SUBEXP;
+ break;
+ case '[':
+ token->type = OP_OPEN_BRACKET;
+ break;
+ case '.':
+ token->type = OP_PERIOD;
+ break;
+ case '^':
+ if (!(syntax & (RE_CONTEXT_INDEP_ANCHORS | RE_CARET_ANCHORS_HERE)) &&
+ re_string_cur_idx (input) != 0)
+ {
+ char prev = re_string_peek_byte (input, -1);
+ if (!(syntax & RE_NEWLINE_ALT) || prev != '\n')
+ break;
+ }
+ token->type = ANCHOR;
+ token->opr.ctx_type = LINE_FIRST;
+ break;
+ case '$':
+ if (!(syntax & RE_CONTEXT_INDEP_ANCHORS) &&
+ re_string_cur_idx (input) + 1 != re_string_length (input))
+ {
+ re_token_t next;
+ re_string_skip_bytes (input, 1);
+ peek_token (&next, input, syntax);
+ re_string_skip_bytes (input, -1);
+ if (next.type != OP_ALT && next.type != OP_CLOSE_SUBEXP)
+ break;
+ }
+ token->type = ANCHOR;
+ token->opr.ctx_type = LINE_LAST;
+ break;
+ default:
+ break;
+ }
+ return 1;
+}
+
+/* Peek a token from INPUT, and return the length of the token.
+ We must not use this function out of bracket expressions. */
+
+static int
+peek_token_bracket (re_token_t *token, re_string_t *input, reg_syntax_t syntax)
+{
+ unsigned char c;
+ if (re_string_eoi (input))
+ {
+ token->type = END_OF_RE;
+ return 0;
+ }
+ c = re_string_peek_byte (input, 0);
+ token->opr.c = c;
+
+#ifdef RE_ENABLE_I18N
+ if (input->mb_cur_max > 1 &&
+ !re_string_first_byte (input, re_string_cur_idx (input)))
+ {
+ token->type = CHARACTER;
+ return 1;
+ }
+#endif /* RE_ENABLE_I18N */
+
+ if (c == '\\' && (syntax & RE_BACKSLASH_ESCAPE_IN_LISTS)
+ && re_string_cur_idx (input) + 1 < re_string_length (input))
+ {
+ /* In this case, '\' escape a character. */
+ unsigned char c2;
+ re_string_skip_bytes (input, 1);
+ c2 = re_string_peek_byte (input, 0);
+ token->opr.c = c2;
+ token->type = CHARACTER;
+ return 1;
+ }
+ if (c == '[') /* '[' is a special char in a bracket exps. */
+ {
+ unsigned char c2;
+ int token_len;
+ if (re_string_cur_idx (input) + 1 < re_string_length (input))
+ c2 = re_string_peek_byte (input, 1);
+ else
+ c2 = 0;
+ token->opr.c = c2;
+ token_len = 2;
+ switch (c2)
+ {
+ case '.':
+ token->type = OP_OPEN_COLL_ELEM;
+ break;
+
+ case '=':
+ token->type = OP_OPEN_EQUIV_CLASS;
+ break;
+
+ case ':':
+ if (syntax & RE_CHAR_CLASSES)
+ {
+ token->type = OP_OPEN_CHAR_CLASS;
+ break;
+ }
+ FALLTHROUGH;
+ default:
+ token->type = CHARACTER;
+ token->opr.c = c;
+ token_len = 1;
+ break;
+ }
+ return token_len;
+ }
+ switch (c)
+ {
+ case '-':
+ token->type = OP_CHARSET_RANGE;
+ break;
+ case ']':
+ token->type = OP_CLOSE_BRACKET;
+ break;
+ case '^':
+ token->type = OP_NON_MATCH_LIST;
+ break;
+ default:
+ token->type = CHARACTER;
+ }
+ return 1;
+}
+
+/* Functions for parser. */
+
+/* Entry point of the parser.
+ Parse the regular expression REGEXP and return the structure tree.
+ If an error occurs, ERR is set by error code, and return NULL.
+ This function build the following tree, from regular expression <reg_exp>:
+ CAT
+ / \
+ / \
+ <reg_exp> EOR
+
+ CAT means concatenation.
+ EOR means end of regular expression. */
+
+static bin_tree_t *
+parse (re_string_t *regexp, regex_t *preg, reg_syntax_t syntax,
+ reg_errcode_t *err)
+{
+ re_dfa_t *dfa = preg->buffer;
+ bin_tree_t *tree, *eor, *root;
+ re_token_t current_token;
+ dfa->syntax = syntax;
+ fetch_token (¤t_token, regexp, syntax | RE_CARET_ANCHORS_HERE);
+ tree = parse_reg_exp (regexp, preg, ¤t_token, syntax, 0, err);
+ if (BE (*err != REG_NOERROR && tree == NULL, 0))
+ return NULL;
+ eor = create_tree (dfa, NULL, NULL, END_OF_RE);
+ if (tree != NULL)
+ root = create_tree (dfa, tree, eor, CONCAT);
+ else
+ root = eor;
+ if (BE (eor == NULL || root == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ return root;
+}
+
+/* This function build the following tree, from regular expression
+ <branch1>|<branch2>:
+ ALT
+ / \
+ / \
+ <branch1> <branch2>
+
+ ALT means alternative, which represents the operator '|'. */
+
+static bin_tree_t *
+parse_reg_exp (re_string_t *regexp, regex_t *preg, re_token_t *token,
+ reg_syntax_t syntax, Idx nest, reg_errcode_t *err)
+{
+ re_dfa_t *dfa = preg->buffer;
+ bin_tree_t *tree, *branch = NULL;
+ bitset_word_t initial_bkref_map = dfa->completed_bkref_map;
+ tree = parse_branch (regexp, preg, token, syntax, nest, err);
+ if (BE (*err != REG_NOERROR && tree == NULL, 0))
+ return NULL;
+
+ while (token->type == OP_ALT)
+ {
+ fetch_token (token, regexp, syntax | RE_CARET_ANCHORS_HERE);
+ if (token->type != OP_ALT && token->type != END_OF_RE
+ && (nest == 0 || token->type != OP_CLOSE_SUBEXP))
+ {
+ bitset_word_t accumulated_bkref_map = dfa->completed_bkref_map;
+ dfa->completed_bkref_map = initial_bkref_map;
+ branch = parse_branch (regexp, preg, token, syntax, nest, err);
+ if (BE (*err != REG_NOERROR && branch == NULL, 0))
+ {
+ if (tree != NULL)
+ postorder (tree, free_tree, NULL);
+ return NULL;
+ }
+ dfa->completed_bkref_map |= accumulated_bkref_map;
+ }
+ else
+ branch = NULL;
+ tree = create_tree (dfa, tree, branch, OP_ALT);
+ if (BE (tree == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ }
+ return tree;
+}
+
+/* This function build the following tree, from regular expression
+ <exp1><exp2>:
+ CAT
+ / \
+ / \
+ <exp1> <exp2>
+
+ CAT means concatenation. */
+
+static bin_tree_t *
+parse_branch (re_string_t *regexp, regex_t *preg, re_token_t *token,
+ reg_syntax_t syntax, Idx nest, reg_errcode_t *err)
+{
+ bin_tree_t *tree, *expr;
+ re_dfa_t *dfa = preg->buffer;
+ tree = parse_expression (regexp, preg, token, syntax, nest, err);
+ if (BE (*err != REG_NOERROR && tree == NULL, 0))
+ return NULL;
+
+ while (token->type != OP_ALT && token->type != END_OF_RE
+ && (nest == 0 || token->type != OP_CLOSE_SUBEXP))
+ {
+ expr = parse_expression (regexp, preg, token, syntax, nest, err);
+ if (BE (*err != REG_NOERROR && expr == NULL, 0))
+ {
+ if (tree != NULL)
+ postorder (tree, free_tree, NULL);
+ return NULL;
+ }
+ if (tree != NULL && expr != NULL)
+ {
+ bin_tree_t *newtree = create_tree (dfa, tree, expr, CONCAT);
+ if (newtree == NULL)
+ {
+ postorder (expr, free_tree, NULL);
+ postorder (tree, free_tree, NULL);
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ tree = newtree;
+ }
+ else if (tree == NULL)
+ tree = expr;
+ /* Otherwise expr == NULL, we don't need to create new tree. */
+ }
+ return tree;
+}
+
+/* This function build the following tree, from regular expression a*:
+ *
+ |
+ a
+*/
+
+static bin_tree_t *
+parse_expression (re_string_t *regexp, regex_t *preg, re_token_t *token,
+ reg_syntax_t syntax, Idx nest, reg_errcode_t *err)
+{
+ re_dfa_t *dfa = preg->buffer;
+ bin_tree_t *tree;
+ switch (token->type)
+ {
+ case CHARACTER:
+ tree = create_token_tree (dfa, NULL, NULL, token);
+ if (BE (tree == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+#ifdef RE_ENABLE_I18N
+ if (dfa->mb_cur_max > 1)
+ {
+ while (!re_string_eoi (regexp)
+ && !re_string_first_byte (regexp, re_string_cur_idx (regexp)))
+ {
+ bin_tree_t *mbc_remain;
+ fetch_token (token, regexp, syntax);
+ mbc_remain = create_token_tree (dfa, NULL, NULL, token);
+ tree = create_tree (dfa, tree, mbc_remain, CONCAT);
+ if (BE (mbc_remain == NULL || tree == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ }
+ }
+#endif
+ break;
+
+ case OP_OPEN_SUBEXP:
+ tree = parse_sub_exp (regexp, preg, token, syntax, nest + 1, err);
+ if (BE (*err != REG_NOERROR && tree == NULL, 0))
+ return NULL;
+ break;
+
+ case OP_OPEN_BRACKET:
+ tree = parse_bracket_exp (regexp, dfa, token, syntax, err);
+ if (BE (*err != REG_NOERROR && tree == NULL, 0))
+ return NULL;
+ break;
+
+ case OP_BACK_REF:
+ if (!BE (dfa->completed_bkref_map & (1 << token->opr.idx), 1))
+ {
+ *err = REG_ESUBREG;
+ return NULL;
+ }
+ dfa->used_bkref_map |= 1 << token->opr.idx;
+ tree = create_token_tree (dfa, NULL, NULL, token);
+ if (BE (tree == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ ++dfa->nbackref;
+ dfa->has_mb_node = 1;
+ break;
+
+ case OP_OPEN_DUP_NUM:
+ if (syntax & RE_CONTEXT_INVALID_DUP)
+ {
+ *err = REG_BADRPT;
+ return NULL;
+ }
+ FALLTHROUGH;
+ case OP_DUP_ASTERISK:
+ case OP_DUP_PLUS:
+ case OP_DUP_QUESTION:
+ if (syntax & RE_CONTEXT_INVALID_OPS)
+ {
+ *err = REG_BADRPT;
+ return NULL;
+ }
+ else if (syntax & RE_CONTEXT_INDEP_OPS)
+ {
+ fetch_token (token, regexp, syntax);
+ return parse_expression (regexp, preg, token, syntax, nest, err);
+ }
+ FALLTHROUGH;
+ case OP_CLOSE_SUBEXP:
+ if ((token->type == OP_CLOSE_SUBEXP) &&
+ !(syntax & RE_UNMATCHED_RIGHT_PAREN_ORD))
+ {
+ *err = REG_ERPAREN;
+ return NULL;
+ }
+ FALLTHROUGH;
+ case OP_CLOSE_DUP_NUM:
+ /* We treat it as a normal character. */
+
+ /* Then we can these characters as normal characters. */
+ token->type = CHARACTER;
+ /* mb_partial and word_char bits should be initialized already
+ by peek_token. */
+ tree = create_token_tree (dfa, NULL, NULL, token);
+ if (BE (tree == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ break;
+
+ case ANCHOR:
+ if ((token->opr.ctx_type
+ & (WORD_DELIM | NOT_WORD_DELIM | WORD_FIRST | WORD_LAST))
+ && dfa->word_ops_used == 0)
+ init_word_char (dfa);
+ if (token->opr.ctx_type == WORD_DELIM
+ || token->opr.ctx_type == NOT_WORD_DELIM)
+ {
+ bin_tree_t *tree_first, *tree_last;
+ if (token->opr.ctx_type == WORD_DELIM)
+ {
+ token->opr.ctx_type = WORD_FIRST;
+ tree_first = create_token_tree (dfa, NULL, NULL, token);
+ token->opr.ctx_type = WORD_LAST;
+ }
+ else
+ {
+ token->opr.ctx_type = INSIDE_WORD;
+ tree_first = create_token_tree (dfa, NULL, NULL, token);
+ token->opr.ctx_type = INSIDE_NOTWORD;
+ }
+ tree_last = create_token_tree (dfa, NULL, NULL, token);
+ tree = create_tree (dfa, tree_first, tree_last, OP_ALT);
+ if (BE (tree_first == NULL || tree_last == NULL || tree == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ }
+ else
+ {
+ tree = create_token_tree (dfa, NULL, NULL, token);
+ if (BE (tree == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ }
+ /* We must return here, since ANCHORs can't be followed
+ by repetition operators.
+ eg. RE"^*" is invalid or "<ANCHOR(^)><CHAR(*)>",
+ it must not be "<ANCHOR(^)><REPEAT(*)>". */
+ fetch_token (token, regexp, syntax);
+ return tree;
+
+ case OP_PERIOD:
+ tree = create_token_tree (dfa, NULL, NULL, token);
+ if (BE (tree == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ if (dfa->mb_cur_max > 1)
+ dfa->has_mb_node = 1;
+ break;
+
+ case OP_WORD:
+ case OP_NOTWORD:
+ tree = build_charclass_op (dfa, regexp->trans,
+ "alnum",
+ "_",
+ token->type == OP_NOTWORD, err);
+ if (BE (*err != REG_NOERROR && tree == NULL, 0))
+ return NULL;
+ break;
+
+ case OP_SPACE:
+ case OP_NOTSPACE:
+ tree = build_charclass_op (dfa, regexp->trans,
+ "space",
+ "",
+ token->type == OP_NOTSPACE, err);
+ if (BE (*err != REG_NOERROR && tree == NULL, 0))
+ return NULL;
+ break;
+
+ case OP_ALT:
+ case END_OF_RE:
+ return NULL;
+
+ case BACK_SLASH:
+ *err = REG_EESCAPE;
+ return NULL;
+
+ default:
+ /* Must not happen? */
+#ifdef DEBUG
+ assert (0);
+#endif
+ return NULL;
+ }
+ fetch_token (token, regexp, syntax);
+
+ while (token->type == OP_DUP_ASTERISK || token->type == OP_DUP_PLUS
+ || token->type == OP_DUP_QUESTION || token->type == OP_OPEN_DUP_NUM)
+ {
+ bin_tree_t *dup_tree = parse_dup_op (tree, regexp, dfa, token,
+ syntax, err);
+ if (BE (*err != REG_NOERROR && dup_tree == NULL, 0))
+ {
+ if (tree != NULL)
+ postorder (tree, free_tree, NULL);
+ return NULL;
+ }
+ tree = dup_tree;
+ /* In BRE consecutive duplications are not allowed. */
+ if ((syntax & RE_CONTEXT_INVALID_DUP)
+ && (token->type == OP_DUP_ASTERISK
+ || token->type == OP_OPEN_DUP_NUM))
+ {
+ if (tree != NULL)
+ postorder (tree, free_tree, NULL);
+ *err = REG_BADRPT;
+ return NULL;
+ }
+ }
+
+ return tree;
+}
+
+/* This function build the following tree, from regular expression
+ (<reg_exp>):
+ SUBEXP
+ |
+ <reg_exp>
+*/
+
+static bin_tree_t *
+parse_sub_exp (re_string_t *regexp, regex_t *preg, re_token_t *token,
+ reg_syntax_t syntax, Idx nest, reg_errcode_t *err)
+{
+ re_dfa_t *dfa = preg->buffer;
+ bin_tree_t *tree;
+ size_t cur_nsub;
+ cur_nsub = preg->re_nsub++;
+
+ fetch_token (token, regexp, syntax | RE_CARET_ANCHORS_HERE);
+
+ /* The subexpression may be a null string. */
+ if (token->type == OP_CLOSE_SUBEXP)
+ tree = NULL;
+ else
+ {
+ tree = parse_reg_exp (regexp, preg, token, syntax, nest, err);
+ if (BE (*err == REG_NOERROR && token->type != OP_CLOSE_SUBEXP, 0))
+ {
+ if (tree != NULL)
+ postorder (tree, free_tree, NULL);
+ *err = REG_EPAREN;
+ }
+ if (BE (*err != REG_NOERROR, 0))
+ return NULL;
+ }
+
+ if (cur_nsub <= '9' - '1')
+ dfa->completed_bkref_map |= 1 << cur_nsub;
+
+ tree = create_tree (dfa, tree, NULL, SUBEXP);
+ if (BE (tree == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ tree->token.opr.idx = cur_nsub;
+ return tree;
+}
+
+/* This function parse repetition operators like "*", "+", "{1,3}" etc. */
+
+static bin_tree_t *
+parse_dup_op (bin_tree_t *elem, re_string_t *regexp, re_dfa_t *dfa,
+ re_token_t *token, reg_syntax_t syntax, reg_errcode_t *err)
+{
+ bin_tree_t *tree = NULL, *old_tree = NULL;
+ Idx i, start, end, start_idx = re_string_cur_idx (regexp);
+ re_token_t start_token = *token;
+
+ if (token->type == OP_OPEN_DUP_NUM)
+ {
+ end = 0;
+ start = fetch_number (regexp, token, syntax);
+ if (start == -1)
+ {
+ if (token->type == CHARACTER && token->opr.c == ',')
+ start = 0; /* We treat "{,m}" as "{0,m}". */
+ else
+ {
+ *err = REG_BADBR; /* <re>{} is invalid. */
+ return NULL;
+ }
+ }
+ if (BE (start != -2, 1))
+ {
+ /* We treat "{n}" as "{n,n}". */
+ end = ((token->type == OP_CLOSE_DUP_NUM) ? start
+ : ((token->type == CHARACTER && token->opr.c == ',')
+ ? fetch_number (regexp, token, syntax) : -2));
+ }
+ if (BE (start == -2 || end == -2, 0))
+ {
+ /* Invalid sequence. */
+ if (BE (!(syntax & RE_INVALID_INTERVAL_ORD), 0))
+ {
+ if (token->type == END_OF_RE)
+ *err = REG_EBRACE;
+ else
+ *err = REG_BADBR;
+
+ return NULL;
+ }
+
+ /* If the syntax bit is set, rollback. */
+ re_string_set_index (regexp, start_idx);
+ *token = start_token;
+ token->type = CHARACTER;
+ /* mb_partial and word_char bits should be already initialized by
+ peek_token. */
+ return elem;
+ }
+
+ if (BE ((end != -1 && start > end)
+ || token->type != OP_CLOSE_DUP_NUM, 0))
+ {
+ /* First number greater than second. */
+ *err = REG_BADBR;
+ return NULL;
+ }
+
+ if (BE (RE_DUP_MAX < (end == -1 ? start : end), 0))
+ {
+ *err = REG_ESIZE;
+ return NULL;
+ }
+ }
+ else
+ {
+ start = (token->type == OP_DUP_PLUS) ? 1 : 0;
+ end = (token->type == OP_DUP_QUESTION) ? 1 : -1;
+ }
+
+ fetch_token (token, regexp, syntax);
+
+ if (BE (elem == NULL, 0))
+ return NULL;
+ if (BE (start == 0 && end == 0, 0))
+ {
+ postorder (elem, free_tree, NULL);
+ return NULL;
+ }
+
+ /* Extract "<re>{n,m}" to "<re><re>...<re><re>{0,<m-n>}". */
+ if (BE (start > 0, 0))
+ {
+ tree = elem;
+ for (i = 2; i <= start; ++i)
+ {
+ elem = duplicate_tree (elem, dfa);
+ tree = create_tree (dfa, tree, elem, CONCAT);
+ if (BE (elem == NULL || tree == NULL, 0))
+ goto parse_dup_op_espace;
+ }
+
+ if (start == end)
+ return tree;
+
+ /* Duplicate ELEM before it is marked optional. */
+ elem = duplicate_tree (elem, dfa);
+ if (BE (elem == NULL, 0))
+ goto parse_dup_op_espace;
+ old_tree = tree;
+ }
+ else
+ old_tree = NULL;
+
+ if (elem->token.type == SUBEXP)
+ {
+ uintptr_t subidx = elem->token.opr.idx;
+ postorder (elem, mark_opt_subexp, (void *) subidx);
+ }
+
+ tree = create_tree (dfa, elem, NULL,
+ (end == -1 ? OP_DUP_ASTERISK : OP_ALT));
+ if (BE (tree == NULL, 0))
+ goto parse_dup_op_espace;
+
+ /* This loop is actually executed only when end != -1,
+ to rewrite <re>{0,n} as (<re>(<re>...<re>?)?)?... We have
+ already created the start+1-th copy. */
+ if (TYPE_SIGNED (Idx) || end != -1)
+ for (i = start + 2; i <= end; ++i)
+ {
+ elem = duplicate_tree (elem, dfa);
+ tree = create_tree (dfa, tree, elem, CONCAT);
+ if (BE (elem == NULL || tree == NULL, 0))
+ goto parse_dup_op_espace;
+
+ tree = create_tree (dfa, tree, NULL, OP_ALT);
+ if (BE (tree == NULL, 0))
+ goto parse_dup_op_espace;
+ }
+
+ if (old_tree)
+ tree = create_tree (dfa, old_tree, tree, CONCAT);
+
+ return tree;
+
+ parse_dup_op_espace:
+ *err = REG_ESPACE;
+ return NULL;
+}
+
+/* Size of the names for collating symbol/equivalence_class/character_class.
+ I'm not sure, but maybe enough. */
+#define BRACKET_NAME_BUF_SIZE 32
+
+#ifndef _LIBC
+
+# ifdef RE_ENABLE_I18N
+/* Convert the byte B to the corresponding wide character. In a
+ unibyte locale, treat B as itself if it is an encoding error.
+ In a multibyte locale, return WEOF if B is an encoding error. */
+static wint_t
+parse_byte (unsigned char b, re_charset_t *mbcset)
+{
+ wint_t wc = __btowc (b);
+ return wc == WEOF && !mbcset ? b : wc;
+}
+#endif
+
+ /* Local function for parse_bracket_exp only used in case of NOT _LIBC.
+ Build the range expression which starts from START_ELEM, and ends
+ at END_ELEM. The result are written to MBCSET and SBCSET.
+ RANGE_ALLOC is the allocated size of mbcset->range_starts, and
+ mbcset->range_ends, is a pointer argument since we may
+ update it. */
+
+static reg_errcode_t
+# ifdef RE_ENABLE_I18N
+build_range_exp (const reg_syntax_t syntax,
+ bitset_t sbcset,
+ re_charset_t *mbcset,
+ Idx *range_alloc,
+ const bracket_elem_t *start_elem,
+ const bracket_elem_t *end_elem)
+# else /* not RE_ENABLE_I18N */
+build_range_exp (const reg_syntax_t syntax,
+ bitset_t sbcset,
+ const bracket_elem_t *start_elem,
+ const bracket_elem_t *end_elem)
+# endif /* not RE_ENABLE_I18N */
+{
+ unsigned int start_ch, end_ch;
+ /* Equivalence Classes and Character Classes can't be a range start/end. */
+ if (BE (start_elem->type == EQUIV_CLASS || start_elem->type == CHAR_CLASS
+ || end_elem->type == EQUIV_CLASS || end_elem->type == CHAR_CLASS,
+ 0))
+ return REG_ERANGE;
+
+ /* We can handle no multi character collating elements without libc
+ support. */
+ if (BE ((start_elem->type == COLL_SYM
+ && strlen ((char *) start_elem->opr.name) > 1)
+ || (end_elem->type == COLL_SYM
+ && strlen ((char *) end_elem->opr.name) > 1), 0))
+ return REG_ECOLLATE;
+
+# ifdef RE_ENABLE_I18N
+ {
+ wchar_t wc;
+ wint_t start_wc;
+ wint_t end_wc;
+
+ start_ch = ((start_elem->type == SB_CHAR) ? start_elem->opr.ch
+ : ((start_elem->type == COLL_SYM) ? start_elem->opr.name[0]
+ : 0));
+ end_ch = ((end_elem->type == SB_CHAR) ? end_elem->opr.ch
+ : ((end_elem->type == COLL_SYM) ? end_elem->opr.name[0]
+ : 0));
+ start_wc = ((start_elem->type == SB_CHAR || start_elem->type == COLL_SYM)
+ ? parse_byte (start_ch, mbcset) : start_elem->opr.wch);
+ end_wc = ((end_elem->type == SB_CHAR || end_elem->type == COLL_SYM)
+ ? parse_byte (end_ch, mbcset) : end_elem->opr.wch);
+ if (start_wc == WEOF || end_wc == WEOF)
+ return REG_ECOLLATE;
+ else if (BE ((syntax & RE_NO_EMPTY_RANGES) && start_wc > end_wc, 0))
+ return REG_ERANGE;
+
+ /* Got valid collation sequence values, add them as a new entry.
+ However, for !_LIBC we have no collation elements: if the
+ character set is single byte, the single byte character set
+ that we build below suffices. parse_bracket_exp passes
+ no MBCSET if dfa->mb_cur_max == 1. */
+ if (mbcset)
+ {
+ /* Check the space of the arrays. */
+ if (BE (*range_alloc == mbcset->nranges, 0))
+ {
+ /* There is not enough space, need realloc. */
+ wchar_t *new_array_start, *new_array_end;
+ Idx new_nranges;
+
+ /* +1 in case of mbcset->nranges is 0. */
+ new_nranges = 2 * mbcset->nranges + 1;
+ /* Use realloc since mbcset->range_starts and mbcset->range_ends
+ are NULL if *range_alloc == 0. */
+ new_array_start = re_realloc (mbcset->range_starts, wchar_t,
+ new_nranges);
+ new_array_end = re_realloc (mbcset->range_ends, wchar_t,
+ new_nranges);
+
+ if (BE (new_array_start == NULL || new_array_end == NULL, 0))
+ {
+ re_free (new_array_start);
+ re_free (new_array_end);
+ return REG_ESPACE;
+ }
+
+ mbcset->range_starts = new_array_start;
+ mbcset->range_ends = new_array_end;
+ *range_alloc = new_nranges;
+ }
+
+ mbcset->range_starts[mbcset->nranges] = start_wc;
+ mbcset->range_ends[mbcset->nranges++] = end_wc;
+ }
+
+ /* Build the table for single byte characters. */
+ for (wc = 0; wc < SBC_MAX; ++wc)
+ {
+ if (start_wc <= wc && wc <= end_wc)
+ bitset_set (sbcset, wc);
+ }
+ }
+# else /* not RE_ENABLE_I18N */
+ {
+ unsigned int ch;
+ start_ch = ((start_elem->type == SB_CHAR ) ? start_elem->opr.ch
+ : ((start_elem->type == COLL_SYM) ? start_elem->opr.name[0]
+ : 0));
+ end_ch = ((end_elem->type == SB_CHAR ) ? end_elem->opr.ch
+ : ((end_elem->type == COLL_SYM) ? end_elem->opr.name[0]
+ : 0));
+ if (start_ch > end_ch)
+ return REG_ERANGE;
+ /* Build the table for single byte characters. */
+ for (ch = 0; ch < SBC_MAX; ++ch)
+ if (start_ch <= ch && ch <= end_ch)
+ bitset_set (sbcset, ch);
+ }
+# endif /* not RE_ENABLE_I18N */
+ return REG_NOERROR;
+}
+#endif /* not _LIBC */
+
+#ifndef _LIBC
+/* Helper function for parse_bracket_exp only used in case of NOT _LIBC..
+ Build the collating element which is represented by NAME.
+ The result are written to MBCSET and SBCSET.
+ COLL_SYM_ALLOC is the allocated size of mbcset->coll_sym, is a
+ pointer argument since we may update it. */
+
+static reg_errcode_t
+# ifdef RE_ENABLE_I18N
+build_collating_symbol (bitset_t sbcset, re_charset_t *mbcset,
+ Idx *coll_sym_alloc, const unsigned char *name)
+# else /* not RE_ENABLE_I18N */
+build_collating_symbol (bitset_t sbcset, const unsigned char *name)
+# endif /* not RE_ENABLE_I18N */
+{
+ size_t name_len = strlen ((const char *) name);
+ if (BE (name_len != 1, 0))
+ return REG_ECOLLATE;
+ else
+ {
+ bitset_set (sbcset, name[0]);
+ return REG_NOERROR;
+ }
+}
+#endif /* not _LIBC */
+
+/* This function parse bracket expression like "[abc]", "[a-c]",
+ "[[.a-a.]]" etc. */
+
+static bin_tree_t *
+parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token,
+ reg_syntax_t syntax, reg_errcode_t *err)
+{
+#ifdef _LIBC
+ const unsigned char *collseqmb;
+ const char *collseqwc;
+ uint32_t nrules;
+ int32_t table_size;
+ const int32_t *symb_table;
+ const unsigned char *extra;
+
+ /* Local function for parse_bracket_exp used in _LIBC environment.
+ Seek the collating symbol entry corresponding to NAME.
+ Return the index of the symbol in the SYMB_TABLE,
+ or -1 if not found. */
+
+ auto inline int32_t
+ __attribute__ ((always_inline))
+ seek_collating_symbol_entry (const unsigned char *name, size_t name_len)
+ {
+ int32_t elem;
+
+ for (elem = 0; elem < table_size; elem++)
+ if (symb_table[2 * elem] != 0)
+ {
+ int32_t idx = symb_table[2 * elem + 1];
+ /* Skip the name of collating element name. */
+ idx += 1 + extra[idx];
+ if (/* Compare the length of the name. */
+ name_len == extra[idx]
+ /* Compare the name. */
+ && memcmp (name, &extra[idx + 1], name_len) == 0)
+ /* Yep, this is the entry. */
+ return elem;
+ }
+ return -1;
+ }
+
+ /* Local function for parse_bracket_exp used in _LIBC environment.
+ Look up the collation sequence value of BR_ELEM.
+ Return the value if succeeded, UINT_MAX otherwise. */
+
+ auto inline unsigned int
+ __attribute__ ((always_inline))
+ lookup_collation_sequence_value (bracket_elem_t *br_elem)
+ {
+ if (br_elem->type == SB_CHAR)
+ {
+ /*
+ if (MB_CUR_MAX == 1)
+ */
+ if (nrules == 0)
+ return collseqmb[br_elem->opr.ch];
+ else
+ {
+ wint_t wc = __btowc (br_elem->opr.ch);
+ return __collseq_table_lookup (collseqwc, wc);
+ }
+ }
+ else if (br_elem->type == MB_CHAR)
+ {
+ if (nrules != 0)
+ return __collseq_table_lookup (collseqwc, br_elem->opr.wch);
+ }
+ else if (br_elem->type == COLL_SYM)
+ {
+ size_t sym_name_len = strlen ((char *) br_elem->opr.name);
+ if (nrules != 0)
+ {
+ int32_t elem, idx;
+ elem = seek_collating_symbol_entry (br_elem->opr.name,
+ sym_name_len);
+ if (elem != -1)
+ {
+ /* We found the entry. */
+ idx = symb_table[2 * elem + 1];
+ /* Skip the name of collating element name. */
+ idx += 1 + extra[idx];
+ /* Skip the byte sequence of the collating element. */
+ idx += 1 + extra[idx];
+ /* Adjust for the alignment. */
+ idx = (idx + 3) & ~3;
+ /* Skip the multibyte collation sequence value. */
+ idx += sizeof (unsigned int);
+ /* Skip the wide char sequence of the collating element. */
+ idx += sizeof (unsigned int) *
+ (1 + *(unsigned int *) (extra + idx));
+ /* Return the collation sequence value. */
+ return *(unsigned int *) (extra + idx);
+ }
+ else if (sym_name_len == 1)
+ {
+ /* No valid character. Match it as a single byte
+ character. */
+ return collseqmb[br_elem->opr.name[0]];
+ }
+ }
+ else if (sym_name_len == 1)
+ return collseqmb[br_elem->opr.name[0]];
+ }
+ return UINT_MAX;
+ }
+
+ /* Local function for parse_bracket_exp used in _LIBC environment.
+ Build the range expression which starts from START_ELEM, and ends
+ at END_ELEM. The result are written to MBCSET and SBCSET.
+ RANGE_ALLOC is the allocated size of mbcset->range_starts, and
+ mbcset->range_ends, is a pointer argument since we may
+ update it. */
+
+ auto inline reg_errcode_t
+ __attribute__ ((always_inline))
+ build_range_exp (bitset_t sbcset, re_charset_t *mbcset, int *range_alloc,
+ bracket_elem_t *start_elem, bracket_elem_t *end_elem)
+ {
+ unsigned int ch;
+ uint32_t start_collseq;
+ uint32_t end_collseq;
+
+ /* Equivalence Classes and Character Classes can't be a range
+ start/end. */
+ if (BE (start_elem->type == EQUIV_CLASS || start_elem->type == CHAR_CLASS
+ || end_elem->type == EQUIV_CLASS || end_elem->type == CHAR_CLASS,
+ 0))
+ return REG_ERANGE;
+
+ /* FIXME: Implement rational ranges here, too. */
+ start_collseq = lookup_collation_sequence_value (start_elem);
+ end_collseq = lookup_collation_sequence_value (end_elem);
+ /* Check start/end collation sequence values. */
+ if (BE (start_collseq == UINT_MAX || end_collseq == UINT_MAX, 0))
+ return REG_ECOLLATE;
+ if (BE ((syntax & RE_NO_EMPTY_RANGES) && start_collseq > end_collseq, 0))
+ return REG_ERANGE;
+
+ /* Got valid collation sequence values, add them as a new entry.
+ However, if we have no collation elements, and the character set
+ is single byte, the single byte character set that we
+ build below suffices. */
+ if (nrules > 0 || dfa->mb_cur_max > 1)
+ {
+ /* Check the space of the arrays. */
+ if (BE (*range_alloc == mbcset->nranges, 0))
+ {
+ /* There is not enough space, need realloc. */
+ uint32_t *new_array_start;
+ uint32_t *new_array_end;
+ Idx new_nranges;
+
+ /* +1 in case of mbcset->nranges is 0. */
+ new_nranges = 2 * mbcset->nranges + 1;
+ new_array_start = re_realloc (mbcset->range_starts, uint32_t,
+ new_nranges);
+ new_array_end = re_realloc (mbcset->range_ends, uint32_t,
+ new_nranges);
+
+ if (BE (new_array_start == NULL || new_array_end == NULL, 0))
+ return REG_ESPACE;
+
+ mbcset->range_starts = new_array_start;
+ mbcset->range_ends = new_array_end;
+ *range_alloc = new_nranges;
+ }
+
+ mbcset->range_starts[mbcset->nranges] = start_collseq;
+ mbcset->range_ends[mbcset->nranges++] = end_collseq;
+ }
+
+ /* Build the table for single byte characters. */
+ for (ch = 0; ch < SBC_MAX; ch++)
+ {
+ uint32_t ch_collseq;
+ /*
+ if (MB_CUR_MAX == 1)
+ */
+ if (nrules == 0)
+ ch_collseq = collseqmb[ch];
+ else
+ ch_collseq = __collseq_table_lookup (collseqwc, __btowc (ch));
+ if (start_collseq <= ch_collseq && ch_collseq <= end_collseq)
+ bitset_set (sbcset, ch);
+ }
+ return REG_NOERROR;
+ }
+
+ /* Local function for parse_bracket_exp used in _LIBC environment.
+ Build the collating element which is represented by NAME.
+ The result are written to MBCSET and SBCSET.
+ COLL_SYM_ALLOC is the allocated size of mbcset->coll_sym, is a
+ pointer argument since we may update it. */
+
+ auto inline reg_errcode_t
+ __attribute__ ((always_inline))
+ build_collating_symbol (bitset_t sbcset, re_charset_t *mbcset,
+ Idx *coll_sym_alloc, const unsigned char *name)
+ {
+ int32_t elem, idx;
+ size_t name_len = strlen ((const char *) name);
+ if (nrules != 0)
+ {
+ elem = seek_collating_symbol_entry (name, name_len);
+ if (elem != -1)
+ {
+ /* We found the entry. */
+ idx = symb_table[2 * elem + 1];
+ /* Skip the name of collating element name. */
+ idx += 1 + extra[idx];
+ }
+ else if (name_len == 1)
+ {
+ /* No valid character, treat it as a normal
+ character. */
+ bitset_set (sbcset, name[0]);
+ return REG_NOERROR;
+ }
+ else
+ return REG_ECOLLATE;
+
+ /* Got valid collation sequence, add it as a new entry. */
+ /* Check the space of the arrays. */
+ if (BE (*coll_sym_alloc == mbcset->ncoll_syms, 0))
+ {
+ /* Not enough, realloc it. */
+ /* +1 in case of mbcset->ncoll_syms is 0. */
+ Idx new_coll_sym_alloc = 2 * mbcset->ncoll_syms + 1;
+ /* Use realloc since mbcset->coll_syms is NULL
+ if *alloc == 0. */
+ int32_t *new_coll_syms = re_realloc (mbcset->coll_syms, int32_t,
+ new_coll_sym_alloc);
+ if (BE (new_coll_syms == NULL, 0))
+ return REG_ESPACE;
+ mbcset->coll_syms = new_coll_syms;
+ *coll_sym_alloc = new_coll_sym_alloc;
+ }
+ mbcset->coll_syms[mbcset->ncoll_syms++] = idx;
+ return REG_NOERROR;
+ }
+ else
+ {
+ if (BE (name_len != 1, 0))
+ return REG_ECOLLATE;
+ else
+ {
+ bitset_set (sbcset, name[0]);
+ return REG_NOERROR;
+ }
+ }
+ }
+#endif
+
+ re_token_t br_token;
+ re_bitset_ptr_t sbcset;
+#ifdef RE_ENABLE_I18N
+ re_charset_t *mbcset;
+ Idx coll_sym_alloc = 0, range_alloc = 0, mbchar_alloc = 0;
+ Idx equiv_class_alloc = 0, char_class_alloc = 0;
+#endif /* not RE_ENABLE_I18N */
+ bool non_match = false;
+ bin_tree_t *work_tree;
+ int token_len;
+ bool first_round = true;
+#ifdef _LIBC
+ collseqmb = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQMB);
+ nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
+ if (nrules)
+ {
+ /*
+ if (MB_CUR_MAX > 1)
+ */
+ collseqwc = _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQWC);
+ table_size = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_SYMB_HASH_SIZEMB);
+ symb_table = (const int32_t *) _NL_CURRENT (LC_COLLATE,
+ _NL_COLLATE_SYMB_TABLEMB);
+ extra = (const unsigned char *) _NL_CURRENT (LC_COLLATE,
+ _NL_COLLATE_SYMB_EXTRAMB);
+ }
+#endif
+ sbcset = (re_bitset_ptr_t) calloc (sizeof (bitset_t), 1);
+#ifdef RE_ENABLE_I18N
+ mbcset = (re_charset_t *) calloc (sizeof (re_charset_t), 1);
+#endif /* RE_ENABLE_I18N */
+#ifdef RE_ENABLE_I18N
+ if (BE (sbcset == NULL || mbcset == NULL, 0))
+#else
+ if (BE (sbcset == NULL, 0))
+#endif /* RE_ENABLE_I18N */
+ {
+ re_free (sbcset);
+#ifdef RE_ENABLE_I18N
+ re_free (mbcset);
+#endif
+ *err = REG_ESPACE;
+ return NULL;
+ }
+
+ token_len = peek_token_bracket (token, regexp, syntax);
+ if (BE (token->type == END_OF_RE, 0))
+ {
+ *err = REG_BADPAT;
+ goto parse_bracket_exp_free_return;
+ }
+ if (token->type == OP_NON_MATCH_LIST)
+ {
+#ifdef RE_ENABLE_I18N
+ mbcset->non_match = 1;
+#endif /* not RE_ENABLE_I18N */
+ non_match = true;
+ if (syntax & RE_HAT_LISTS_NOT_NEWLINE)
+ bitset_set (sbcset, '\n');
+ re_string_skip_bytes (regexp, token_len); /* Skip a token. */
+ token_len = peek_token_bracket (token, regexp, syntax);
+ if (BE (token->type == END_OF_RE, 0))
+ {
+ *err = REG_BADPAT;
+ goto parse_bracket_exp_free_return;
+ }
+ }
+
+ /* We treat the first ']' as a normal character. */
+ if (token->type == OP_CLOSE_BRACKET)
+ token->type = CHARACTER;
+
+ while (1)
+ {
+ bracket_elem_t start_elem, end_elem;
+ unsigned char start_name_buf[BRACKET_NAME_BUF_SIZE];
+ unsigned char end_name_buf[BRACKET_NAME_BUF_SIZE];
+ reg_errcode_t ret;
+ int token_len2 = 0;
+ bool is_range_exp = false;
+ re_token_t token2;
+
+ start_elem.opr.name = start_name_buf;
+ start_elem.type = COLL_SYM;
+ ret = parse_bracket_element (&start_elem, regexp, token, token_len, dfa,
+ syntax, first_round);
+ if (BE (ret != REG_NOERROR, 0))
+ {
+ *err = ret;
+ goto parse_bracket_exp_free_return;
+ }
+ first_round = false;
+
+ /* Get information about the next token. We need it in any case. */
+ token_len = peek_token_bracket (token, regexp, syntax);
+
+ /* Do not check for ranges if we know they are not allowed. */
+ if (start_elem.type != CHAR_CLASS && start_elem.type != EQUIV_CLASS)
+ {
+ if (BE (token->type == END_OF_RE, 0))
+ {
+ *err = REG_EBRACK;
+ goto parse_bracket_exp_free_return;
+ }
+ if (token->type == OP_CHARSET_RANGE)
+ {
+ re_string_skip_bytes (regexp, token_len); /* Skip '-'. */
+ token_len2 = peek_token_bracket (&token2, regexp, syntax);
+ if (BE (token2.type == END_OF_RE, 0))
+ {
+ *err = REG_EBRACK;
+ goto parse_bracket_exp_free_return;
+ }
+ if (token2.type == OP_CLOSE_BRACKET)
+ {
+ /* We treat the last '-' as a normal character. */
+ re_string_skip_bytes (regexp, -token_len);
+ token->type = CHARACTER;
+ }
+ else
+ is_range_exp = true;
+ }
+ }
+
+ if (is_range_exp == true)
+ {
+ end_elem.opr.name = end_name_buf;
+ end_elem.type = COLL_SYM;
+ ret = parse_bracket_element (&end_elem, regexp, &token2, token_len2,
+ dfa, syntax, true);
+ if (BE (ret != REG_NOERROR, 0))
+ {
+ *err = ret;
+ goto parse_bracket_exp_free_return;
+ }
+
+ token_len = peek_token_bracket (token, regexp, syntax);
+
+#ifdef _LIBC
+ *err = build_range_exp (sbcset, mbcset, &range_alloc,
+ &start_elem, &end_elem);
+#else
+# ifdef RE_ENABLE_I18N
+ *err = build_range_exp (syntax, sbcset,
+ dfa->mb_cur_max > 1 ? mbcset : NULL,
+ &range_alloc, &start_elem, &end_elem);
+# else
+ *err = build_range_exp (syntax, sbcset, &start_elem, &end_elem);
+# endif
+#endif /* RE_ENABLE_I18N */
+ if (BE (*err != REG_NOERROR, 0))
+ goto parse_bracket_exp_free_return;
+ }
+ else
+ {
+ switch (start_elem.type)
+ {
+ case SB_CHAR:
+ bitset_set (sbcset, start_elem.opr.ch);
+ break;
+#ifdef RE_ENABLE_I18N
+ case MB_CHAR:
+ /* Check whether the array has enough space. */
+ if (BE (mbchar_alloc == mbcset->nmbchars, 0))
+ {
+ wchar_t *new_mbchars;
+ /* Not enough, realloc it. */
+ /* +1 in case of mbcset->nmbchars is 0. */
+ mbchar_alloc = 2 * mbcset->nmbchars + 1;
+ /* Use realloc since array is NULL if *alloc == 0. */
+ new_mbchars = re_realloc (mbcset->mbchars, wchar_t,
+ mbchar_alloc);
+ if (BE (new_mbchars == NULL, 0))
+ goto parse_bracket_exp_espace;
+ mbcset->mbchars = new_mbchars;
+ }
+ mbcset->mbchars[mbcset->nmbchars++] = start_elem.opr.wch;
+ break;
+#endif /* RE_ENABLE_I18N */
+ case EQUIV_CLASS:
+ *err = build_equiv_class (sbcset,
+#ifdef RE_ENABLE_I18N
+ mbcset, &equiv_class_alloc,
+#endif /* RE_ENABLE_I18N */
+ start_elem.opr.name);
+ if (BE (*err != REG_NOERROR, 0))
+ goto parse_bracket_exp_free_return;
+ break;
+ case COLL_SYM:
+ *err = build_collating_symbol (sbcset,
+#ifdef RE_ENABLE_I18N
+ mbcset, &coll_sym_alloc,
+#endif /* RE_ENABLE_I18N */
+ start_elem.opr.name);
+ if (BE (*err != REG_NOERROR, 0))
+ goto parse_bracket_exp_free_return;
+ break;
+ case CHAR_CLASS:
+ *err = build_charclass (regexp->trans, sbcset,
+#ifdef RE_ENABLE_I18N
+ mbcset, &char_class_alloc,
+#endif /* RE_ENABLE_I18N */
+ (const char *) start_elem.opr.name,
+ syntax);
+ if (BE (*err != REG_NOERROR, 0))
+ goto parse_bracket_exp_free_return;
+ break;
+ default:
+ assert (0);
+ break;
+ }
+ }
+ if (BE (token->type == END_OF_RE, 0))
+ {
+ *err = REG_EBRACK;
+ goto parse_bracket_exp_free_return;
+ }
+ if (token->type == OP_CLOSE_BRACKET)
+ break;
+ }
+
+ re_string_skip_bytes (regexp, token_len); /* Skip a token. */
+
+ /* If it is non-matching list. */
+ if (non_match)
+ bitset_not (sbcset);
+
+#ifdef RE_ENABLE_I18N
+ /* Ensure only single byte characters are set. */
+ if (dfa->mb_cur_max > 1)
+ bitset_mask (sbcset, dfa->sb_char);
+
+ if (mbcset->nmbchars || mbcset->ncoll_syms || mbcset->nequiv_classes
+ || mbcset->nranges || (dfa->mb_cur_max > 1 && (mbcset->nchar_classes
+ || mbcset->non_match)))
+ {
+ bin_tree_t *mbc_tree;
+ int sbc_idx;
+ /* Build a tree for complex bracket. */
+ dfa->has_mb_node = 1;
+ br_token.type = COMPLEX_BRACKET;
+ br_token.opr.mbcset = mbcset;
+ mbc_tree = create_token_tree (dfa, NULL, NULL, &br_token);
+ if (BE (mbc_tree == NULL, 0))
+ goto parse_bracket_exp_espace;
+ for (sbc_idx = 0; sbc_idx < BITSET_WORDS; ++sbc_idx)
+ if (sbcset[sbc_idx])
+ break;
+ /* If there are no bits set in sbcset, there is no point
+ of having both SIMPLE_BRACKET and COMPLEX_BRACKET. */
+ if (sbc_idx < BITSET_WORDS)
+ {
+ /* Build a tree for simple bracket. */
+ br_token.type = SIMPLE_BRACKET;
+ br_token.opr.sbcset = sbcset;
+ work_tree = create_token_tree (dfa, NULL, NULL, &br_token);
+ if (BE (work_tree == NULL, 0))
+ goto parse_bracket_exp_espace;
+
+ /* Then join them by ALT node. */
+ work_tree = create_tree (dfa, work_tree, mbc_tree, OP_ALT);
+ if (BE (work_tree == NULL, 0))
+ goto parse_bracket_exp_espace;
+ }
+ else
+ {
+ re_free (sbcset);
+ work_tree = mbc_tree;
+ }
+ }
+ else
+#endif /* not RE_ENABLE_I18N */
+ {
+#ifdef RE_ENABLE_I18N
+ free_charset (mbcset);
+#endif
+ /* Build a tree for simple bracket. */
+ br_token.type = SIMPLE_BRACKET;
+ br_token.opr.sbcset = sbcset;
+ work_tree = create_token_tree (dfa, NULL, NULL, &br_token);
+ if (BE (work_tree == NULL, 0))
+ goto parse_bracket_exp_espace;
+ }
+ return work_tree;
+
+ parse_bracket_exp_espace:
+ *err = REG_ESPACE;
+ parse_bracket_exp_free_return:
+ re_free (sbcset);
+#ifdef RE_ENABLE_I18N
+ free_charset (mbcset);
+#endif /* RE_ENABLE_I18N */
+ return NULL;
+}
+
+/* Parse an element in the bracket expression. */
+
+static reg_errcode_t
+parse_bracket_element (bracket_elem_t *elem, re_string_t *regexp,
+ re_token_t *token, int token_len, re_dfa_t *dfa,
+ reg_syntax_t syntax, bool accept_hyphen)
+{
+#ifdef RE_ENABLE_I18N
+ int cur_char_size;
+ cur_char_size = re_string_char_size_at (regexp, re_string_cur_idx (regexp));
+ if (cur_char_size > 1)
+ {
+ elem->type = MB_CHAR;
+ elem->opr.wch = re_string_wchar_at (regexp, re_string_cur_idx (regexp));
+ re_string_skip_bytes (regexp, cur_char_size);
+ return REG_NOERROR;
+ }
+#endif /* RE_ENABLE_I18N */
+ re_string_skip_bytes (regexp, token_len); /* Skip a token. */
+ if (token->type == OP_OPEN_COLL_ELEM || token->type == OP_OPEN_CHAR_CLASS
+ || token->type == OP_OPEN_EQUIV_CLASS)
+ return parse_bracket_symbol (elem, regexp, token);
+ if (BE (token->type == OP_CHARSET_RANGE, 0) && !accept_hyphen)
+ {
+ /* A '-' must only appear as anything but a range indicator before
+ the closing bracket. Everything else is an error. */
+ re_token_t token2;
+ (void) peek_token_bracket (&token2, regexp, syntax);
+ if (token2.type != OP_CLOSE_BRACKET)
+ /* The actual error value is not standardized since this whole
+ case is undefined. But ERANGE makes good sense. */
+ return REG_ERANGE;
+ }
+ elem->type = SB_CHAR;
+ elem->opr.ch = token->opr.c;
+ return REG_NOERROR;
+}
+
+/* Parse a bracket symbol in the bracket expression. Bracket symbols are
+ such as [:<character_class>:], [.<collating_element>.], and
+ [=<equivalent_class>=]. */
+
+static reg_errcode_t
+parse_bracket_symbol (bracket_elem_t *elem, re_string_t *regexp,
+ re_token_t *token)
+{
+ unsigned char ch, delim = token->opr.c;
+ int i = 0;
+ if (re_string_eoi(regexp))
+ return REG_EBRACK;
+ for (;; ++i)
+ {
+ if (i >= BRACKET_NAME_BUF_SIZE)
+ return REG_EBRACK;
+ if (token->type == OP_OPEN_CHAR_CLASS)
+ ch = re_string_fetch_byte_case (regexp);
+ else
+ ch = re_string_fetch_byte (regexp);
+ if (re_string_eoi(regexp))
+ return REG_EBRACK;
+ if (ch == delim && re_string_peek_byte (regexp, 0) == ']')
+ break;
+ elem->opr.name[i] = ch;
+ }
+ re_string_skip_bytes (regexp, 1);
+ elem->opr.name[i] = '\0';
+ switch (token->type)
+ {
+ case OP_OPEN_COLL_ELEM:
+ elem->type = COLL_SYM;
+ break;
+ case OP_OPEN_EQUIV_CLASS:
+ elem->type = EQUIV_CLASS;
+ break;
+ case OP_OPEN_CHAR_CLASS:
+ elem->type = CHAR_CLASS;
+ break;
+ default:
+ break;
+ }
+ return REG_NOERROR;
+}
+
+ /* Helper function for parse_bracket_exp.
+ Build the equivalence class which is represented by NAME.
+ The result are written to MBCSET and SBCSET.
+ EQUIV_CLASS_ALLOC is the allocated size of mbcset->equiv_classes,
+ is a pointer argument since we may update it. */
+
+static reg_errcode_t
+#ifdef RE_ENABLE_I18N
+build_equiv_class (bitset_t sbcset, re_charset_t *mbcset,
+ Idx *equiv_class_alloc, const unsigned char *name)
+#else /* not RE_ENABLE_I18N */
+build_equiv_class (bitset_t sbcset, const unsigned char *name)
+#endif /* not RE_ENABLE_I18N */
+{
+#ifdef _LIBC
+ uint32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
+ if (nrules != 0)
+ {
+ const int32_t *table, *indirect;
+ const unsigned char *weights, *extra, *cp;
+ unsigned char char_buf[2];
+ int32_t idx1, idx2;
+ unsigned int ch;
+ size_t len;
+ /* Calculate the index for equivalence class. */
+ cp = name;
+ table = (const int32_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB);
+ weights = (const unsigned char *) _NL_CURRENT (LC_COLLATE,
+ _NL_COLLATE_WEIGHTMB);
+ extra = (const unsigned char *) _NL_CURRENT (LC_COLLATE,
+ _NL_COLLATE_EXTRAMB);
+ indirect = (const int32_t *) _NL_CURRENT (LC_COLLATE,
+ _NL_COLLATE_INDIRECTMB);
+ idx1 = findidx (table, indirect, extra, &cp, -1);
+ if (BE (idx1 == 0 || *cp != '\0', 0))
+ /* This isn't a valid character. */
+ return REG_ECOLLATE;
+
+ /* Build single byte matching table for this equivalence class. */
+ len = weights[idx1 & 0xffffff];
+ for (ch = 0; ch < SBC_MAX; ++ch)
+ {
+ char_buf[0] = ch;
+ cp = char_buf;
+ idx2 = findidx (table, indirect, extra, &cp, 1);
+/*
+ idx2 = table[ch];
+*/
+ if (idx2 == 0)
+ /* This isn't a valid character. */
+ continue;
+ /* Compare only if the length matches and the collation rule
+ index is the same. */
+ if (len == weights[idx2 & 0xffffff] && (idx1 >> 24) == (idx2 >> 24))
+ {
+ int cnt = 0;
+
+ while (cnt <= len &&
+ weights[(idx1 & 0xffffff) + 1 + cnt]
+ == weights[(idx2 & 0xffffff) + 1 + cnt])
+ ++cnt;
+
+ if (cnt > len)
+ bitset_set (sbcset, ch);
+ }
+ }
+ /* Check whether the array has enough space. */
+ if (BE (*equiv_class_alloc == mbcset->nequiv_classes, 0))
+ {
+ /* Not enough, realloc it. */
+ /* +1 in case of mbcset->nequiv_classes is 0. */
+ Idx new_equiv_class_alloc = 2 * mbcset->nequiv_classes + 1;
+ /* Use realloc since the array is NULL if *alloc == 0. */
+ int32_t *new_equiv_classes = re_realloc (mbcset->equiv_classes,
+ int32_t,
+ new_equiv_class_alloc);
+ if (BE (new_equiv_classes == NULL, 0))
+ return REG_ESPACE;
+ mbcset->equiv_classes = new_equiv_classes;
+ *equiv_class_alloc = new_equiv_class_alloc;
+ }
+ mbcset->equiv_classes[mbcset->nequiv_classes++] = idx1;
+ }
+ else
+#endif /* _LIBC */
+ {
+ if (BE (strlen ((const char *) name) != 1, 0))
+ return REG_ECOLLATE;
+ bitset_set (sbcset, *name);
+ }
+ return REG_NOERROR;
+}
+
+ /* Helper function for parse_bracket_exp.
+ Build the character class which is represented by NAME.
+ The result are written to MBCSET and SBCSET.
+ CHAR_CLASS_ALLOC is the allocated size of mbcset->char_classes,
+ is a pointer argument since we may update it. */
+
+static reg_errcode_t
+#ifdef RE_ENABLE_I18N
+build_charclass (RE_TRANSLATE_TYPE trans, bitset_t sbcset,
+ re_charset_t *mbcset, Idx *char_class_alloc,
+ const char *class_name, reg_syntax_t syntax)
+#else /* not RE_ENABLE_I18N */
+build_charclass (RE_TRANSLATE_TYPE trans, bitset_t sbcset,
+ const char *class_name, reg_syntax_t syntax)
+#endif /* not RE_ENABLE_I18N */
+{
+ int i;
+ const char *name = class_name;
+
+ /* In case of REG_ICASE "upper" and "lower" match the both of
+ upper and lower cases. */
+ if ((syntax & RE_ICASE)
+ && (strcmp (name, "upper") == 0 || strcmp (name, "lower") == 0))
+ name = "alpha";
+
+#ifdef RE_ENABLE_I18N
+ /* Check the space of the arrays. */
+ if (BE (*char_class_alloc == mbcset->nchar_classes, 0))
+ {
+ /* Not enough, realloc it. */
+ /* +1 in case of mbcset->nchar_classes is 0. */
+ Idx new_char_class_alloc = 2 * mbcset->nchar_classes + 1;
+ /* Use realloc since array is NULL if *alloc == 0. */
+ wctype_t *new_char_classes = re_realloc (mbcset->char_classes, wctype_t,
+ new_char_class_alloc);
+ if (BE (new_char_classes == NULL, 0))
+ return REG_ESPACE;
+ mbcset->char_classes = new_char_classes;
+ *char_class_alloc = new_char_class_alloc;
+ }
+ mbcset->char_classes[mbcset->nchar_classes++] = __wctype (name);
+#endif /* RE_ENABLE_I18N */
+
+#define BUILD_CHARCLASS_LOOP(ctype_func) \
+ do { \
+ if (BE (trans != NULL, 0)) \
+ { \
+ for (i = 0; i < SBC_MAX; ++i) \
+ if (ctype_func (i)) \
+ bitset_set (sbcset, trans[i]); \
+ } \
+ else \
+ { \
+ for (i = 0; i < SBC_MAX; ++i) \
+ if (ctype_func (i)) \
+ bitset_set (sbcset, i); \
+ } \
+ } while (0)
+
+ if (strcmp (name, "alnum") == 0)
+ BUILD_CHARCLASS_LOOP (isalnum);
+ else if (strcmp (name, "cntrl") == 0)
+ BUILD_CHARCLASS_LOOP (iscntrl);
+ else if (strcmp (name, "lower") == 0)
+ BUILD_CHARCLASS_LOOP (islower);
+ else if (strcmp (name, "space") == 0)
+ BUILD_CHARCLASS_LOOP (isspace);
+ else if (strcmp (name, "alpha") == 0)
+ BUILD_CHARCLASS_LOOP (isalpha);
+ else if (strcmp (name, "digit") == 0)
+ BUILD_CHARCLASS_LOOP (isdigit);
+ else if (strcmp (name, "print") == 0)
+ BUILD_CHARCLASS_LOOP (isprint);
+ else if (strcmp (name, "upper") == 0)
+ BUILD_CHARCLASS_LOOP (isupper);
+ else if (strcmp (name, "blank") == 0)
+ BUILD_CHARCLASS_LOOP (isblank);
+ else if (strcmp (name, "graph") == 0)
+ BUILD_CHARCLASS_LOOP (isgraph);
+ else if (strcmp (name, "punct") == 0)
+ BUILD_CHARCLASS_LOOP (ispunct);
+ else if (strcmp (name, "xdigit") == 0)
+ BUILD_CHARCLASS_LOOP (isxdigit);
+ else
+ return REG_ECTYPE;
+
+ return REG_NOERROR;
+}
+
+static bin_tree_t *
+build_charclass_op (re_dfa_t *dfa, RE_TRANSLATE_TYPE trans,
+ const char *class_name,
+ const char *extra, bool non_match,
+ reg_errcode_t *err)
+{
+ re_bitset_ptr_t sbcset;
+#ifdef RE_ENABLE_I18N
+ re_charset_t *mbcset;
+ Idx alloc = 0;
+#endif /* not RE_ENABLE_I18N */
+ reg_errcode_t ret;
+ re_token_t br_token;
+ bin_tree_t *tree;
+
+ sbcset = (re_bitset_ptr_t) calloc (sizeof (bitset_t), 1);
+ if (BE (sbcset == NULL, 0))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+#ifdef RE_ENABLE_I18N
+ mbcset = (re_charset_t *) calloc (sizeof (re_charset_t), 1);
+ if (BE (mbcset == NULL, 0))
+ {
+ re_free (sbcset);
+ *err = REG_ESPACE;
+ return NULL;
+ }
+ mbcset->non_match = non_match;
+#endif /* RE_ENABLE_I18N */
+
+ /* We don't care the syntax in this case. */
+ ret = build_charclass (trans, sbcset,
+#ifdef RE_ENABLE_I18N
+ mbcset, &alloc,
+#endif /* RE_ENABLE_I18N */
+ class_name, 0);
+
+ if (BE (ret != REG_NOERROR, 0))
+ {
+ re_free (sbcset);
+#ifdef RE_ENABLE_I18N
+ free_charset (mbcset);
+#endif /* RE_ENABLE_I18N */
+ *err = ret;
+ return NULL;
+ }
+ /* \w match '_' also. */
+ for (; *extra; extra++)
+ bitset_set (sbcset, *extra);
+
+ /* If it is non-matching list. */
+ if (non_match)
+ bitset_not (sbcset);
+
+#ifdef RE_ENABLE_I18N
+ /* Ensure only single byte characters are set. */
+ if (dfa->mb_cur_max > 1)
+ bitset_mask (sbcset, dfa->sb_char);
+#endif
+
+ /* Build a tree for simple bracket. */
+#if defined GCC_LINT || defined lint
+ memset (&br_token, 0, sizeof br_token);
+#endif
+ br_token.type = SIMPLE_BRACKET;
+ br_token.opr.sbcset = sbcset;
+ tree = create_token_tree (dfa, NULL, NULL, &br_token);
+ if (BE (tree == NULL, 0))
+ goto build_word_op_espace;
+
+#ifdef RE_ENABLE_I18N
+ if (dfa->mb_cur_max > 1)
+ {
+ bin_tree_t *mbc_tree;
+ /* Build a tree for complex bracket. */
+ br_token.type = COMPLEX_BRACKET;
+ br_token.opr.mbcset = mbcset;
+ dfa->has_mb_node = 1;
+ mbc_tree = create_token_tree (dfa, NULL, NULL, &br_token);
+ if (BE (mbc_tree == NULL, 0))
+ goto build_word_op_espace;
+ /* Then join them by ALT node. */
+ tree = create_tree (dfa, tree, mbc_tree, OP_ALT);
+ if (BE (mbc_tree != NULL, 1))
+ return tree;
+ }
+ else
+ {
+ free_charset (mbcset);
+ return tree;
+ }
+#else /* not RE_ENABLE_I18N */
+ return tree;
+#endif /* not RE_ENABLE_I18N */
+
+ build_word_op_espace:
+ re_free (sbcset);
+#ifdef RE_ENABLE_I18N
+ free_charset (mbcset);
+#endif /* RE_ENABLE_I18N */
+ *err = REG_ESPACE;
+ return NULL;
+}
+
+/* This is intended for the expressions like "a{1,3}".
+ Fetch a number from 'input', and return the number.
+ Return -1 if the number field is empty like "{,1}".
+ Return RE_DUP_MAX + 1 if the number field is too large.
+ Return -2 if an error occurred. */
+
+static Idx
+fetch_number (re_string_t *input, re_token_t *token, reg_syntax_t syntax)
+{
+ Idx num = -1;
+ unsigned char c;
+ while (1)
+ {
+ fetch_token (token, input, syntax);
+ c = token->opr.c;
+ if (BE (token->type == END_OF_RE, 0))
+ return -2;
+ if (token->type == OP_CLOSE_DUP_NUM || c == ',')
+ break;
+ num = ((token->type != CHARACTER || c < '0' || '9' < c || num == -2)
+ ? -2
+ : num == -1
+ ? c - '0'
+ : MIN (RE_DUP_MAX + 1, num * 10 + c - '0'));
+ }
+ return num;
+}
+
+#ifdef RE_ENABLE_I18N
+static void
+free_charset (re_charset_t *cset)
+{
+ re_free (cset->mbchars);
+# ifdef _LIBC
+ re_free (cset->coll_syms);
+ re_free (cset->equiv_classes);
+ re_free (cset->range_starts);
+ re_free (cset->range_ends);
+# endif
+ re_free (cset->char_classes);
+ re_free (cset);
+}
+#endif /* RE_ENABLE_I18N */
+
+/* Functions for binary tree operation. */
+
+/* Create a tree node. */
+
+static bin_tree_t *
+create_tree (re_dfa_t *dfa, bin_tree_t *left, bin_tree_t *right,
+ re_token_type_t type)
+{
+ re_token_t t;
+#if defined GCC_LINT || defined lint
+ memset (&t, 0, sizeof t);
+#endif
+ t.type = type;
+ return create_token_tree (dfa, left, right, &t);
+}
+
+static bin_tree_t *
+create_token_tree (re_dfa_t *dfa, bin_tree_t *left, bin_tree_t *right,
+ const re_token_t *token)
+{
+ bin_tree_t *tree;
+ if (BE (dfa->str_tree_storage_idx == BIN_TREE_STORAGE_SIZE, 0))
+ {
+ bin_tree_storage_t *storage = re_malloc (bin_tree_storage_t, 1);
+
+ if (storage == NULL)
+ return NULL;
+ storage->next = dfa->str_tree_storage;
+ dfa->str_tree_storage = storage;
+ dfa->str_tree_storage_idx = 0;
+ }
+ tree = &dfa->str_tree_storage->data[dfa->str_tree_storage_idx++];
+
+ tree->parent = NULL;
+ tree->left = left;
+ tree->right = right;
+ tree->token = *token;
+ tree->token.duplicated = 0;
+ tree->token.opt_subexp = 0;
+ tree->first = NULL;
+ tree->next = NULL;
+ tree->node_idx = -1;
+
+ if (left != NULL)
+ left->parent = tree;
+ if (right != NULL)
+ right->parent = tree;
+ return tree;
+}
+
+/* Mark the tree SRC as an optional subexpression.
+ To be called from preorder or postorder. */
+
+static reg_errcode_t
+mark_opt_subexp (void *extra, bin_tree_t *node)
+{
+ Idx idx = (uintptr_t) extra;
+ if (node->token.type == SUBEXP && node->token.opr.idx == idx)
+ node->token.opt_subexp = 1;
+
+ return REG_NOERROR;
+}
+
+/* Free the allocated memory inside NODE. */
+
+static void
+free_token (re_token_t *node)
+{
+#ifdef RE_ENABLE_I18N
+ if (node->type == COMPLEX_BRACKET && node->duplicated == 0)
+ free_charset (node->opr.mbcset);
+ else
+#endif /* RE_ENABLE_I18N */
+ if (node->type == SIMPLE_BRACKET && node->duplicated == 0)
+ re_free (node->opr.sbcset);
+}
+
+/* Worker function for tree walking. Free the allocated memory inside NODE
+ and its children. */
+
+static reg_errcode_t
+free_tree (void *extra, bin_tree_t *node)
+{
+ free_token (&node->token);
+ return REG_NOERROR;
+}
+
+
+/* Duplicate the node SRC, and return new node. This is a preorder
+ visit similar to the one implemented by the generic visitor, but
+ we need more infrastructure to maintain two parallel trees --- so,
+ it's easier to duplicate. */
+
+static bin_tree_t *
+duplicate_tree (const bin_tree_t *root, re_dfa_t *dfa)
+{
+ const bin_tree_t *node;
+ bin_tree_t *dup_root;
+ bin_tree_t **p_new = &dup_root, *dup_node = root->parent;
+
+ for (node = root; ; )
+ {
+ /* Create a new tree and link it back to the current parent. */
+ *p_new = create_token_tree (dfa, NULL, NULL, &node->token);
+ if (*p_new == NULL)
+ return NULL;
+ (*p_new)->parent = dup_node;
+ (*p_new)->token.duplicated = 1;
+ dup_node = *p_new;
+
+ /* Go to the left node, or up and to the right. */
+ if (node->left)
+ {
+ node = node->left;
+ p_new = &dup_node->left;
+ }
+ else
+ {
+ const bin_tree_t *prev = NULL;
+ while (node->right == prev || node->right == NULL)
+ {
+ prev = node;
+ node = node->parent;
+ dup_node = dup_node->parent;
+ if (!node)
+ return dup_root;
+ }
+ node = node->right;
+ p_new = &dup_node->right;
+ }
+ }
+}
diff --git a/lib/regex.c b/lib/regex.c
new file mode 100644
index 0000000..499e1f0
--- /dev/null
+++ b/lib/regex.c
@@ -0,0 +1,81 @@
+/* Extended regular expression matching and search library.
+ Copyright (C) 2002-2018 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+ Contributed by Isamu Hasegawa <address@hidden>.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU General Public
+ License as published by the Free Software Foundation; either
+ version 3 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ General Public License for more details.
+
+ You should have received a copy of the GNU General Public
+ License along with the GNU C Library; if not, see
+ <https://www.gnu.org/licenses/>. */
+
+#ifndef _LIBC
+# include <config.h>
+
+# if (__GNUC__ == 4 && 6 <= __GNUC_MINOR__) || 4 < __GNUC__
+# pragma GCC diagnostic ignored "-Wsuggest-attribute=pure"
+# endif
+# if (__GNUC__ == 4 && 3 <= __GNUC_MINOR__) || 4 < __GNUC__
+# pragma GCC diagnostic ignored "-Wold-style-definition"
+# pragma GCC diagnostic ignored "-Wtype-limits"
+# endif
+#endif
+
+/* Make sure no one compiles this code with a C++ compiler. */
+#if defined __cplusplus && defined _LIBC
+# error "This is C code, use a C compiler"
+#endif
+
+#ifdef _LIBC
+/* We have to keep the namespace clean. */
+# define regfree(preg) __regfree (preg)
+# define regexec(pr, st, nm, pm, ef) __regexec (pr, st, nm, pm, ef)
+# define regcomp(preg, pattern, cflags) __regcomp (preg, pattern, cflags)
+# define regerror(errcode, preg, errbuf, errbuf_size) \
+ __regerror(errcode, preg, errbuf, errbuf_size)
+# define re_set_registers(bu, re, nu, st, en) \
+ __re_set_registers (bu, re, nu, st, en)
+# define re_match_2(bufp, string1, size1, string2, size2, pos, regs, stop) \
+ __re_match_2 (bufp, string1, size1, string2, size2, pos, regs, stop)
+# define re_match(bufp, string, size, pos, regs) \
+ __re_match (bufp, string, size, pos, regs)
+# define re_search(bufp, string, size, startpos, range, regs) \
+ __re_search (bufp, string, size, startpos, range, regs)
+# define re_compile_pattern(pattern, length, bufp) \
+ __re_compile_pattern (pattern, length, bufp)
+# define re_set_syntax(syntax) __re_set_syntax (syntax)
+# define re_search_2(bufp, st1, s1, st2, s2, startpos, range, regs, stop) \
+ __re_search_2 (bufp, st1, s1, st2, s2, startpos, range, regs, stop)
+# define re_compile_fastmap(bufp) __re_compile_fastmap (bufp)
+
+# include "../locale/localeinfo.h"
+#endif
+
+/* On some systems, limits.h sets RE_DUP_MAX to a lower value than
+ GNU regex allows. Include it before <regex.h>, which correctly
+ #undefs RE_DUP_MAX and sets it to the right value. */
+#include <limits.h>
+
+#include <regex.h>
+#include "regex_internal.h"
+
+#include "regex_internal.c"
+#include "regcomp.c"
+#include "regexec.c"
+
+/* Binary backward compatibility. */
+#if _LIBC
+# include <shlib-compat.h>
+# if SHLIB_COMPAT (libc, GLIBC_2_0, GLIBC_2_3)
+link_warning (re_max_failures, "the 're_max_failures' variable is obsolete and
will go away.")
+int re_max_failures = 2000;
+# endif
+#endif
diff --git a/lib/regex.h b/lib/regex.h
new file mode 100644
index 0000000..f2ac950
--- /dev/null
+++ b/lib/regex.h
@@ -0,0 +1,658 @@
+/* Definitions for data structures and routines for the regular
+ expression library.
+ Copyright (C) 1985, 1989-2018 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU General Public
+ License as published by the Free Software Foundation; either
+ version 3 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ General Public License for more details.
+
+ You should have received a copy of the GNU General Public
+ License along with the GNU C Library; if not, see
+ <https://www.gnu.org/licenses/>. */
+
+#ifndef _REGEX_H
+#define _REGEX_H 1
+
+#include <sys/types.h>
+
+/* Allow the use in C++ code. */
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* Define __USE_GNU to declare GNU extensions that violate the
+ POSIX name space rules. */
+#ifdef _GNU_SOURCE
+# define __USE_GNU 1
+#endif
+
+#ifdef _REGEX_LARGE_OFFSETS
+
+/* Use types and values that are wide enough to represent signed and
+ unsigned byte offsets in memory. This currently works only when
+ the regex code is used outside of the GNU C library; it is not yet
+ supported within glibc itself, and glibc users should not define
+ _REGEX_LARGE_OFFSETS. */
+
+/* The type of object sizes. */
+typedef size_t __re_size_t;
+
+/* The type of object sizes, in places where the traditional code
+ uses unsigned long int. */
+typedef size_t __re_long_size_t;
+
+#else
+
+/* The traditional GNU regex implementation mishandles strings longer
+ than INT_MAX. */
+typedef unsigned int __re_size_t;
+typedef unsigned long int __re_long_size_t;
+
+#endif
+
+/* The following two types have to be signed and unsigned integer type
+ wide enough to hold a value of a pointer. For most ANSI compilers
+ ptrdiff_t and size_t should be likely OK. Still size of these two
+ types is 2 for Microsoft C. Ugh... */
+typedef long int s_reg_t;
+typedef unsigned long int active_reg_t;
+
+/* The following bits are used to determine the regexp syntax we
+ recognize. The set/not-set meanings are chosen so that Emacs syntax
+ remains the value 0. The bits are given in alphabetical order, and
+ the definitions shifted by one from the previous bit; thus, when we
+ add or remove a bit, only one other definition need change. */
+typedef unsigned long int reg_syntax_t;
+
+#ifdef __USE_GNU
+/* If this bit is not set, then \ inside a bracket expression is literal.
+ If set, then such a \ quotes the following character. */
+# define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
+
+/* If this bit is not set, then + and ? are operators, and \+ and \? are
+ literals.
+ If set, then \+ and \? are operators and + and ? are literals. */
+# define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
+
+/* If this bit is set, then character classes are supported. They are:
+ [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
+ [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
+ If not set, then character classes are not supported. */
+# define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
+
+/* If this bit is set, then ^ and $ are always anchors (outside bracket
+ expressions, of course).
+ If this bit is not set, then it depends:
+ ^ is an anchor if it is at the beginning of a regular
+ expression or after an open-group or an alternation operator;
+ $ is an anchor if it is at the end of a regular expression, or
+ before a close-group or an alternation operator.
+
+ This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
+ POSIX draft 11.2 says that * etc. in leading positions is undefined.
+ We already implemented a previous draft which made those constructs
+ invalid, though, so we haven't changed the code back. */
+# define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
+
+/* If this bit is set, then special characters are always special
+ regardless of where they are in the pattern.
+ If this bit is not set, then special characters are special only in
+ some contexts; otherwise they are ordinary. Specifically,
+ * + ? and intervals are only special when not after the beginning,
+ open-group, or alternation operator. */
+# define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
+
+/* If this bit is set, then *, +, ?, and { cannot be first in an re or
+ immediately after an alternation or begin-group operator. */
+# define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
+
+/* If this bit is set, then . matches newline.
+ If not set, then it doesn't. */
+# define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
+
+/* If this bit is set, then . doesn't match NUL.
+ If not set, then it does. */
+# define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
+
+/* If this bit is set, nonmatching lists [^...] do not match newline.
+ If not set, they do. */
+# define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
+
+/* If this bit is set, either \{...\} or {...} defines an
+ interval, depending on RE_NO_BK_BRACES.
+ If not set, \{, \}, {, and } are literals. */
+# define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
+
+/* If this bit is set, +, ? and | aren't recognized as operators.
+ If not set, they are. */
+# define RE_LIMITED_OPS (RE_INTERVALS << 1)
+
+/* If this bit is set, newline is an alternation operator.
+ If not set, newline is literal. */
+# define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
+
+/* If this bit is set, then '{...}' defines an interval, and \{ and \}
+ are literals.
+ If not set, then '\{...\}' defines an interval. */
+# define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
+
+/* If this bit is set, (...) defines a group, and \( and \) are literals.
+ If not set, \(...\) defines a group, and ( and ) are literals. */
+# define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
+
+/* If this bit is set, then \<digit> matches <digit>.
+ If not set, then \<digit> is a back-reference. */
+# define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
+
+/* If this bit is set, then | is an alternation operator, and \| is literal.
+ If not set, then \| is an alternation operator, and | is literal. */
+# define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
+
+/* If this bit is set, then an ending range point collating higher
+ than the starting range point, as in [z-a], is invalid.
+ If not set, then when ending range point collates higher than the
+ starting range point, the range is ignored. */
+# define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
+
+/* If this bit is set, then an unmatched ) is ordinary.
+ If not set, then an unmatched ) is invalid. */
+# define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
+
+/* If this bit is set, succeed as soon as we match the whole pattern,
+ without further backtracking. */
+# define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
+
+/* If this bit is set, do not process the GNU regex operators.
+ If not set, then the GNU regex operators are recognized. */
+# define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
+
+/* If this bit is set, turn on internal regex debugging.
+ If not set, and debugging was on, turn it off.
+ This only works if regex.c is compiled -DDEBUG.
+ We define this bit always, so that all that's needed to turn on
+ debugging is to recompile regex.c; the calling code can always have
+ this bit set, and it won't affect anything in the normal case. */
+# define RE_DEBUG (RE_NO_GNU_OPS << 1)
+
+/* If this bit is set, a syntactically invalid interval is treated as
+ a string of ordinary characters. For example, the ERE 'a{1' is
+ treated as 'a\{1'. */
+# define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1)
+
+/* If this bit is set, then ignore case when matching.
+ If not set, then case is significant. */
+# define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1)
+
+/* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only
+ for ^, because it is difficult to scan the regex backwards to find
+ whether ^ should be special. */
+# define RE_CARET_ANCHORS_HERE (RE_ICASE << 1)
+
+/* If this bit is set, then \{ cannot be first in a regex or
+ immediately after an alternation, open-group or \} operator. */
+# define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1)
+
+/* If this bit is set, then no_sub will be set to 1 during
+ re_compile_pattern. */
+# define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1)
+#endif
+
+/* This global variable defines the particular regexp syntax to use (for
+ some interfaces). When a regexp is compiled, the syntax used is
+ stored in the pattern buffer, so changing this does not affect
+ already-compiled regexps. */
+extern reg_syntax_t re_syntax_options;
+
+#ifdef __USE_GNU
+/* Define combinations of the above bits for the standard possibilities.
+ (The [[[ comments delimit what gets put into the Texinfo file, so
+ don't delete them!) */
+/* [[[begin syntaxes]]] */
+# define RE_SYNTAX_EMACS 0
+
+# define RE_SYNTAX_AWK \
+ (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
+ | RE_NO_BK_PARENS | RE_NO_BK_REFS \
+ | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
+ | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_CHAR_CLASSES \
+ | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
+
+# define RE_SYNTAX_GNU_AWK \
+ ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
+ | RE_INVALID_INTERVAL_ORD) \
+ & ~(RE_DOT_NOT_NULL | RE_CONTEXT_INDEP_OPS \
+ | RE_CONTEXT_INVALID_OPS ))
+
+# define RE_SYNTAX_POSIX_AWK \
+ (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
+ | RE_INTERVALS | RE_NO_GNU_OPS \
+ | RE_INVALID_INTERVAL_ORD)
+
+# define RE_SYNTAX_GREP
\
+ ((RE_SYNTAX_POSIX_BASIC | RE_NEWLINE_ALT) \
+ & ~(RE_CONTEXT_INVALID_DUP | RE_DOT_NOT_NULL))
+
+# define RE_SYNTAX_EGREP \
+ ((RE_SYNTAX_POSIX_EXTENDED | RE_INVALID_INTERVAL_ORD | RE_NEWLINE_ALT) \
+ & ~(RE_CONTEXT_INVALID_OPS | RE_DOT_NOT_NULL))
+
+/* POSIX grep -E behavior is no longer incompatible with GNU. */
+# define RE_SYNTAX_POSIX_EGREP \
+ RE_SYNTAX_EGREP
+
+/* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
+# define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
+
+# define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
+
+/* Syntax bits common to both basic and extended POSIX regex syntax. */
+# define _RE_SYNTAX_POSIX_COMMON \
+ (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
+ | RE_INTERVALS | RE_NO_EMPTY_RANGES)
+
+# define RE_SYNTAX_POSIX_BASIC \
+ (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP)
+
+/* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
+ RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
+ isn't minimal, since other operators, such as \`, aren't disabled. */
+# define RE_SYNTAX_POSIX_MINIMAL_BASIC \
+ (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
+
+# define RE_SYNTAX_POSIX_EXTENDED \
+ (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
+ | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
+ | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
+
+/* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
+ removed and RE_NO_BK_REFS is added. */
+# define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
+ (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
+ | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
+ | RE_NO_BK_PARENS | RE_NO_BK_REFS \
+ | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
+/* [[[end syntaxes]]] */
+
+/* Maximum number of duplicates an interval can allow. POSIX-conforming
+ systems might define this in <limits.h>, but we want our
+ value, so remove any previous define. */
+# ifdef _REGEX_INCLUDE_LIMITS_H
+# include <limits.h>
+# endif
+# ifdef RE_DUP_MAX
+# undef RE_DUP_MAX
+# endif
+
+/* RE_DUP_MAX is 2**15 - 1 because an earlier implementation stored
+ the counter as a 2-byte signed integer. This is no longer true, so
+ RE_DUP_MAX could be increased to (INT_MAX / 10 - 1), or to
+ ((SIZE_MAX - 9) / 10) if _REGEX_LARGE_OFFSETS is defined.
+ However, there would be a huge performance problem if someone
+ actually used a pattern like a\{214748363\}, so RE_DUP_MAX retains
+ its historical value. */
+# define RE_DUP_MAX (0x7fff)
+#endif
+
+
+/* POSIX 'cflags' bits (i.e., information for 'regcomp'). */
+
+/* If this bit is set, then use extended regular expression syntax.
+ If not set, then use basic regular expression syntax. */
+#define REG_EXTENDED 1
+
+/* If this bit is set, then ignore case when matching.
+ If not set, then case is significant. */
+#define REG_ICASE (1 << 1)
+
+/* If this bit is set, then anchors do not match at newline
+ characters in the string.
+ If not set, then anchors do match at newlines. */
+#define REG_NEWLINE (1 << 2)
+
+/* If this bit is set, then report only success or fail in regexec.
+ If not set, then returns differ between not matching and errors. */
+#define REG_NOSUB (1 << 3)
+
+
+/* POSIX 'eflags' bits (i.e., information for regexec). */
+
+/* If this bit is set, then the beginning-of-line operator doesn't match
+ the beginning of the string (presumably because it's not the
+ beginning of a line).
+ If not set, then the beginning-of-line operator does match the
+ beginning of the string. */
+#define REG_NOTBOL 1
+
+/* Like REG_NOTBOL, except for the end-of-line. */
+#define REG_NOTEOL (1 << 1)
+
+/* Use PMATCH[0] to delimit the start and end of the search in the
+ buffer. */
+#define REG_STARTEND (1 << 2)
+
+
+/* If any error codes are removed, changed, or added, update the
+ '__re_error_msgid' table in regcomp.c. */
+
+typedef enum
+{
+ _REG_ENOSYS = -1, /* This will never happen for this implementation. */
+ _REG_NOERROR = 0, /* Success. */
+ _REG_NOMATCH, /* Didn't find a match (for regexec). */
+
+ /* POSIX regcomp return error codes. (In the order listed in the
+ standard.) */
+ _REG_BADPAT, /* Invalid pattern. */
+ _REG_ECOLLATE, /* Invalid collating element. */
+ _REG_ECTYPE, /* Invalid character class name. */
+ _REG_EESCAPE, /* Trailing backslash. */
+ _REG_ESUBREG, /* Invalid back reference. */
+ _REG_EBRACK, /* Unmatched left bracket. */
+ _REG_EPAREN, /* Parenthesis imbalance. */
+ _REG_EBRACE, /* Unmatched \{. */
+ _REG_BADBR, /* Invalid contents of \{\}. */
+ _REG_ERANGE, /* Invalid range end. */
+ _REG_ESPACE, /* Ran out of memory. */
+ _REG_BADRPT, /* No preceding re for repetition op. */
+
+ /* Error codes we've added. */
+ _REG_EEND, /* Premature end. */
+ _REG_ESIZE, /* Too large (e.g., repeat count too large). */
+ _REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
+} reg_errcode_t;
+
+#if defined _XOPEN_SOURCE || defined __USE_XOPEN2K
+# define REG_ENOSYS _REG_ENOSYS
+#endif
+#define REG_NOERROR _REG_NOERROR
+#define REG_NOMATCH _REG_NOMATCH
+#define REG_BADPAT _REG_BADPAT
+#define REG_ECOLLATE _REG_ECOLLATE
+#define REG_ECTYPE _REG_ECTYPE
+#define REG_EESCAPE _REG_EESCAPE
+#define REG_ESUBREG _REG_ESUBREG
+#define REG_EBRACK _REG_EBRACK
+#define REG_EPAREN _REG_EPAREN
+#define REG_EBRACE _REG_EBRACE
+#define REG_BADBR _REG_BADBR
+#define REG_ERANGE _REG_ERANGE
+#define REG_ESPACE _REG_ESPACE
+#define REG_BADRPT _REG_BADRPT
+#define REG_EEND _REG_EEND
+#define REG_ESIZE _REG_ESIZE
+#define REG_ERPAREN _REG_ERPAREN
+
+/* This data structure represents a compiled pattern. Before calling
+ the pattern compiler, the fields 'buffer', 'allocated', 'fastmap',
+ and 'translate' can be set. After the pattern has been compiled,
+ the fields 're_nsub', 'not_bol' and 'not_eol' are available. All
+ other fields are private to the regex routines. */
+
+#ifndef RE_TRANSLATE_TYPE
+# define __RE_TRANSLATE_TYPE unsigned char *
+# ifdef __USE_GNU
+# define RE_TRANSLATE_TYPE __RE_TRANSLATE_TYPE
+# endif
+#endif
+
+#ifdef __USE_GNU
+# define __REPB_PREFIX(name) name
+#else
+# define __REPB_PREFIX(name) __##name
+#endif
+
+struct re_pattern_buffer
+{
+ /* Space that holds the compiled pattern. The type
+ 'struct re_dfa_t' is private and is not declared here. */
+ struct re_dfa_t *__REPB_PREFIX(buffer);
+
+ /* Number of bytes to which 'buffer' points. */
+ __re_long_size_t __REPB_PREFIX(allocated);
+
+ /* Number of bytes actually used in 'buffer'. */
+ __re_long_size_t __REPB_PREFIX(used);
+
+ /* Syntax setting with which the pattern was compiled. */
+ reg_syntax_t __REPB_PREFIX(syntax);
+
+ /* Pointer to a fastmap, if any, otherwise zero. re_search uses the
+ fastmap, if there is one, to skip over impossible starting points
+ for matches. */
+ char *__REPB_PREFIX(fastmap);
+
+ /* Either a translate table to apply to all characters before
+ comparing them, or zero for no translation. The translation is
+ applied to a pattern when it is compiled and to a string when it
+ is matched. */
+ __RE_TRANSLATE_TYPE __REPB_PREFIX(translate);
+
+ /* Number of subexpressions found by the compiler. */
+ size_t re_nsub;
+
+ /* Zero if this pattern cannot match the empty string, one else.
+ Well, in truth it's used only in 're_search_2', to see whether or
+ not we should use the fastmap, so we don't set this absolutely
+ perfectly; see 're_compile_fastmap' (the "duplicate" case). */
+ unsigned __REPB_PREFIX(can_be_null) : 1;
+
+ /* If REGS_UNALLOCATED, allocate space in the 'regs' structure
+ for 'max (RE_NREGS, re_nsub + 1)' groups.
+ If REGS_REALLOCATE, reallocate space if necessary.
+ If REGS_FIXED, use what's there. */
+#ifdef __USE_GNU
+# define REGS_UNALLOCATED 0
+# define REGS_REALLOCATE 1
+# define REGS_FIXED 2
+#endif
+ unsigned __REPB_PREFIX(regs_allocated) : 2;
+
+ /* Set to zero when 're_compile_pattern' compiles a pattern; set to
+ one by 're_compile_fastmap' if it updates the fastmap. */
+ unsigned __REPB_PREFIX(fastmap_accurate) : 1;
+
+ /* If set, 're_match_2' does not return information about
+ subexpressions. */
+ unsigned __REPB_PREFIX(no_sub) : 1;
+
+ /* If set, a beginning-of-line anchor doesn't match at the beginning
+ of the string. */
+ unsigned __REPB_PREFIX(not_bol) : 1;
+
+ /* Similarly for an end-of-line anchor. */
+ unsigned __REPB_PREFIX(not_eol) : 1;
+
+ /* If true, an anchor at a newline matches. */
+ unsigned __REPB_PREFIX(newline_anchor) : 1;
+};
+
+typedef struct re_pattern_buffer regex_t;
+
+/* Type for byte offsets within the string. POSIX mandates this. */
+#ifdef _REGEX_LARGE_OFFSETS
+/* POSIX 1003.1-2008 requires that regoff_t be at least as wide as
+ ptrdiff_t and ssize_t. We don't know of any hosts where ptrdiff_t
+ is wider than ssize_t, so ssize_t is safe. ptrdiff_t is not
+ visible here, so use ssize_t. */
+typedef ssize_t regoff_t;
+#else
+/* The traditional GNU regex implementation mishandles strings longer
+ than INT_MAX. */
+typedef int regoff_t;
+#endif
+
+
+#ifdef __USE_GNU
+/* This is the structure we store register match data in. See
+ regex.texinfo for a full description of what registers match. */
+struct re_registers
+{
+ __re_size_t num_regs;
+ regoff_t *start;
+ regoff_t *end;
+};
+
+
+/* If 'regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
+ 're_match_2' returns information about at least this many registers
+ the first time a 'regs' structure is passed. */
+# ifndef RE_NREGS
+# define RE_NREGS 30
+# endif
+#endif
+
+
+/* POSIX specification for registers. Aside from the different names than
+ 're_registers', POSIX uses an array of structures, instead of a
+ structure of arrays. */
+typedef struct
+{
+ regoff_t rm_so; /* Byte offset from string's start to substring's start. */
+ regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
+} regmatch_t;
+
+/* Declarations for routines. */
+
+#ifdef __USE_GNU
+/* Sets the current default syntax to SYNTAX, and return the old syntax.
+ You can also simply assign to the 're_syntax_options' variable. */
+extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
+
+/* Compile the regular expression PATTERN, with length LENGTH
+ and syntax given by the global 're_syntax_options', into the buffer
+ BUFFER. Return NULL if successful, and an error string if not.
+
+ To free the allocated storage, you must call 'regfree' on BUFFER.
+ Note that the translate table must either have been initialized by
+ 'regcomp', with a malloc'ed value, or set to NULL before calling
+ 'regfree'. */
+extern const char *re_compile_pattern (const char *__pattern, size_t __length,
+ struct re_pattern_buffer *__buffer);
+
+
+/* Compile a fastmap for the compiled pattern in BUFFER; used to
+ accelerate searches. Return 0 if successful and -2 if was an
+ internal error. */
+extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
+
+
+/* Search in the string STRING (with length LENGTH) for the pattern
+ compiled into BUFFER. Start searching at position START, for RANGE
+ characters. Return the starting position of the match, -1 for no
+ match, or -2 for an internal error. Also return register
+ information in REGS (if REGS and BUFFER->no_sub are nonzero). */
+extern regoff_t re_search (struct re_pattern_buffer *__buffer,
+ const char *__String, regoff_t __length,
+ regoff_t __start, regoff_t __range,
+ struct re_registers *__regs);
+
+
+/* Like 're_search', but search in the concatenation of STRING1 and
+ STRING2. Also, stop searching at index START + STOP. */
+extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
+ const char *__string1, regoff_t __length1,
+ const char *__string2, regoff_t __length2,
+ regoff_t __start, regoff_t __range,
+ struct re_registers *__regs,
+ regoff_t __stop);
+
+
+/* Like 're_search', but return how many characters in STRING the regexp
+ in BUFFER matched, starting at position START. */
+extern regoff_t re_match (struct re_pattern_buffer *__buffer,
+ const char *__String, regoff_t __length,
+ regoff_t __start, struct re_registers *__regs);
+
+
+/* Relates to 're_match' as 're_search_2' relates to 're_search'. */
+extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
+ const char *__string1, regoff_t __length1,
+ const char *__string2, regoff_t __length2,
+ regoff_t __start, struct re_registers *__regs,
+ regoff_t __stop);
+
+
+/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
+ ENDS. Subsequent matches using BUFFER and REGS will use this memory
+ for recording register information. STARTS and ENDS must be
+ allocated with malloc, and must each be at least 'NUM_REGS * sizeof
+ (regoff_t)' bytes long.
+
+ If NUM_REGS == 0, then subsequent matches should allocate their own
+ register data.
+
+ Unless this function is called, the first search or match using
+ BUFFER will allocate its own register data, without
+ freeing the old data. */
+extern void re_set_registers (struct re_pattern_buffer *__buffer,
+ struct re_registers *__regs,
+ __re_size_t __num_regs,
+ regoff_t *__starts, regoff_t *__ends);
+#endif /* Use GNU */
+
+#if defined _REGEX_RE_COMP || (defined _LIBC && defined __USE_MISC)
+# ifndef _CRAY
+/* 4.2 bsd compatibility. */
+extern char *re_comp (const char *);
+extern int re_exec (const char *);
+# endif
+#endif
+
+/* For plain 'restrict', use glibc's __restrict if defined.
+ Otherwise, GCC 2.95 and later have "__restrict"; C99 compilers have
+ "restrict", and "configure" may have defined "restrict".
+ Other compilers use __restrict, __restrict__, and _Restrict, and
+ 'configure' might #define 'restrict' to those words, so pick a
+ different name. */
+#ifndef _Restrict_
+# if defined __restrict || 2 < __GNUC__ + (95 <= __GNUC_MINOR__)
+# define _Restrict_ __restrict
+# elif 199901L <= __STDC_VERSION__ || defined restrict
+# define _Restrict_ restrict
+# else
+# define _Restrict_
+# endif
+#endif
+/* For [restrict], use glibc's __restrict_arr if available.
+ Otherwise, GCC 3.1 (not in C++ mode) and C99 support [restrict]. */
+#ifndef _Restrict_arr_
+# ifdef __restrict_arr
+# define _Restrict_arr_ __restrict_arr
+# elif ((199901L <= __STDC_VERSION__ || 3 < __GNUC__ + (1 <= __GNUC_MINOR__)) \
+ && !defined __GNUG__)
+# define _Restrict_arr_ _Restrict_
+# else
+# define _Restrict_arr_
+# endif
+#endif
+
+/* POSIX compatibility. */
+extern int regcomp (regex_t *_Restrict_ __preg,
+ const char *_Restrict_ __pattern,
+ int __cflags);
+
+extern int regexec (const regex_t *_Restrict_ __preg,
+ const char *_Restrict_ __String, size_t __nmatch,
+ regmatch_t __pmatch[_Restrict_arr_],
+ int __eflags);
+
+extern size_t regerror (int __errcode, const regex_t *_Restrict_ __preg,
+ char *_Restrict_ __errbuf, size_t __errbuf_size);
+
+extern void regfree (regex_t *__preg);
+
+
+#ifdef __cplusplus
+}
+#endif /* C++ */
+
+#endif /* regex.h */
diff --git a/lib/regex_internal.c b/lib/regex_internal.c
new file mode 100644
index 0000000..3237356
--- /dev/null
+++ b/lib/regex_internal.c
@@ -0,0 +1,1740 @@
+/* Extended regular expression matching and search library.
+ Copyright (C) 2002-2018 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+ Contributed by Isamu Hasegawa <address@hidden>.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU General Public
+ License as published by the Free Software Foundation; either
+ version 3 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ General Public License for more details.
+
+ You should have received a copy of the GNU General Public
+ License along with the GNU C Library; if not, see
+ <https://www.gnu.org/licenses/>. */
+
+static void re_string_construct_common (const char *str, Idx len,
+ re_string_t *pstr,
+ RE_TRANSLATE_TYPE trans, bool icase,
+ const re_dfa_t *dfa);
+static re_dfastate_t *create_ci_newstate (const re_dfa_t *dfa,
+ const re_node_set *nodes,
+ re_hashval_t hash);
+static re_dfastate_t *create_cd_newstate (const re_dfa_t *dfa,
+ const re_node_set *nodes,
+ unsigned int context,
+ re_hashval_t hash);
+static reg_errcode_t re_string_realloc_buffers (re_string_t *pstr,
+ Idx new_buf_len);
+#ifdef RE_ENABLE_I18N
+static void build_wcs_buffer (re_string_t *pstr);
+static reg_errcode_t build_wcs_upper_buffer (re_string_t *pstr);
+#endif /* RE_ENABLE_I18N */
+static void build_upper_buffer (re_string_t *pstr);
+static void re_string_translate_buffer (re_string_t *pstr);
+static unsigned int re_string_context_at (const re_string_t *input, Idx idx,
+ int eflags) __attribute__ ((pure));
+
+/* Functions for string operation. */
+
+/* This function allocate the buffers. It is necessary to call
+ re_string_reconstruct before using the object. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_string_allocate (re_string_t *pstr, const char *str, Idx len, Idx init_len,
+ RE_TRANSLATE_TYPE trans, bool icase, const re_dfa_t *dfa)
+{
+ reg_errcode_t ret;
+ Idx init_buf_len;
+
+ /* Ensure at least one character fits into the buffers. */
+ if (init_len < dfa->mb_cur_max)
+ init_len = dfa->mb_cur_max;
+ init_buf_len = (len + 1 < init_len) ? len + 1: init_len;
+ re_string_construct_common (str, len, pstr, trans, icase, dfa);
+
+ ret = re_string_realloc_buffers (pstr, init_buf_len);
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+
+ pstr->word_char = dfa->word_char;
+ pstr->word_ops_used = dfa->word_ops_used;
+ pstr->mbs = pstr->mbs_allocated ? pstr->mbs : (unsigned char *) str;
+ pstr->valid_len = (pstr->mbs_allocated || dfa->mb_cur_max > 1) ? 0 : len;
+ pstr->valid_raw_len = pstr->valid_len;
+ return REG_NOERROR;
+}
+
+/* This function allocate the buffers, and initialize them. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_string_construct (re_string_t *pstr, const char *str, Idx len,
+ RE_TRANSLATE_TYPE trans, bool icase, const re_dfa_t *dfa)
+{
+ reg_errcode_t ret;
+ memset (pstr, '\0', sizeof (re_string_t));
+ re_string_construct_common (str, len, pstr, trans, icase, dfa);
+
+ if (len > 0)
+ {
+ ret = re_string_realloc_buffers (pstr, len + 1);
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+ }
+ pstr->mbs = pstr->mbs_allocated ? pstr->mbs : (unsigned char *) str;
+
+ if (icase)
+ {
+#ifdef RE_ENABLE_I18N
+ if (dfa->mb_cur_max > 1)
+ {
+ while (1)
+ {
+ ret = build_wcs_upper_buffer (pstr);
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+ if (pstr->valid_raw_len >= len)
+ break;
+ if (pstr->bufs_len > pstr->valid_len + dfa->mb_cur_max)
+ break;
+ ret = re_string_realloc_buffers (pstr, pstr->bufs_len * 2);
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+ }
+ }
+ else
+#endif /* RE_ENABLE_I18N */
+ build_upper_buffer (pstr);
+ }
+ else
+ {
+#ifdef RE_ENABLE_I18N
+ if (dfa->mb_cur_max > 1)
+ build_wcs_buffer (pstr);
+ else
+#endif /* RE_ENABLE_I18N */
+ {
+ if (trans != NULL)
+ re_string_translate_buffer (pstr);
+ else
+ {
+ pstr->valid_len = pstr->bufs_len;
+ pstr->valid_raw_len = pstr->bufs_len;
+ }
+ }
+ }
+
+ return REG_NOERROR;
+}
+
+/* Helper functions for re_string_allocate, and re_string_construct. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_string_realloc_buffers (re_string_t *pstr, Idx new_buf_len)
+{
+#ifdef RE_ENABLE_I18N
+ if (pstr->mb_cur_max > 1)
+ {
+ wint_t *new_wcs;
+
+ /* Avoid overflow in realloc. */
+ const size_t max_object_size = MAX (sizeof (wint_t), sizeof (Idx));
+ if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) < new_buf_len, 0))
+ return REG_ESPACE;
+
+ new_wcs = re_realloc (pstr->wcs, wint_t, new_buf_len);
+ if (BE (new_wcs == NULL, 0))
+ return REG_ESPACE;
+ pstr->wcs = new_wcs;
+ if (pstr->offsets != NULL)
+ {
+ Idx *new_offsets = re_realloc (pstr->offsets, Idx, new_buf_len);
+ if (BE (new_offsets == NULL, 0))
+ return REG_ESPACE;
+ pstr->offsets = new_offsets;
+ }
+ }
+#endif /* RE_ENABLE_I18N */
+ if (pstr->mbs_allocated)
+ {
+ unsigned char *new_mbs = re_realloc (pstr->mbs, unsigned char,
+ new_buf_len);
+ if (BE (new_mbs == NULL, 0))
+ return REG_ESPACE;
+ pstr->mbs = new_mbs;
+ }
+ pstr->bufs_len = new_buf_len;
+ return REG_NOERROR;
+}
+
+
+static void
+re_string_construct_common (const char *str, Idx len, re_string_t *pstr,
+ RE_TRANSLATE_TYPE trans, bool icase,
+ const re_dfa_t *dfa)
+{
+ pstr->raw_mbs = (const unsigned char *) str;
+ pstr->len = len;
+ pstr->raw_len = len;
+ pstr->trans = trans;
+ pstr->icase = icase;
+ pstr->mbs_allocated = (trans != NULL || icase);
+ pstr->mb_cur_max = dfa->mb_cur_max;
+ pstr->is_utf8 = dfa->is_utf8;
+ pstr->map_notascii = dfa->map_notascii;
+ pstr->stop = pstr->len;
+ pstr->raw_stop = pstr->stop;
+}
+
+#ifdef RE_ENABLE_I18N
+
+/* Build wide character buffer PSTR->WCS.
+ If the byte sequence of the string are:
+ <mb1>(0), <mb1>(1), <mb2>(0), <mb2>(1), <sb3>
+ Then wide character buffer will be:
+ <wc1> , WEOF , <wc2> , WEOF , <wc3>
+ We use WEOF for padding, they indicate that the position isn't
+ a first byte of a multibyte character.
+
+ Note that this function assumes PSTR->VALID_LEN elements are already
+ built and starts from PSTR->VALID_LEN. */
+
+static void
+build_wcs_buffer (re_string_t *pstr)
+{
+#ifdef _LIBC
+ unsigned char buf[MB_LEN_MAX];
+ assert (MB_LEN_MAX >= pstr->mb_cur_max);
+#else
+ unsigned char buf[64];
+#endif
+ mbstate_t prev_st;
+ Idx byte_idx, end_idx, remain_len;
+ size_t mbclen;
+
+ /* Build the buffers from pstr->valid_len to either pstr->len or
+ pstr->bufs_len. */
+ end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len;
+ for (byte_idx = pstr->valid_len; byte_idx < end_idx;)
+ {
+ wchar_t wc;
+ const char *p;
+
+ remain_len = end_idx - byte_idx;
+ prev_st = pstr->cur_state;
+ /* Apply the translation if we need. */
+ if (BE (pstr->trans != NULL, 0))
+ {
+ int i, ch;
+
+ for (i = 0; i < pstr->mb_cur_max && i < remain_len; ++i)
+ {
+ ch = pstr->raw_mbs [pstr->raw_mbs_idx + byte_idx + i];
+ buf[i] = pstr->mbs[byte_idx + i] = pstr->trans[ch];
+ }
+ p = (const char *) buf;
+ }
+ else
+ p = (const char *) pstr->raw_mbs + pstr->raw_mbs_idx + byte_idx;
+ mbclen = __mbrtowc (&wc, p, remain_len, &pstr->cur_state);
+ if (BE (mbclen == (size_t) -1 || mbclen == 0
+ || (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len), 0))
+ {
+ /* We treat these cases as a singlebyte character. */
+ mbclen = 1;
+ wc = (wchar_t) pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx];
+ if (BE (pstr->trans != NULL, 0))
+ wc = pstr->trans[wc];
+ pstr->cur_state = prev_st;
+ }
+ else if (BE (mbclen == (size_t) -2, 0))
+ {
+ /* The buffer doesn't have enough space, finish to build. */
+ pstr->cur_state = prev_st;
+ break;
+ }
+
+ /* Write wide character and padding. */
+ pstr->wcs[byte_idx++] = wc;
+ /* Write paddings. */
+ for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;)
+ pstr->wcs[byte_idx++] = WEOF;
+ }
+ pstr->valid_len = byte_idx;
+ pstr->valid_raw_len = byte_idx;
+}
+
+/* Build wide character buffer PSTR->WCS like build_wcs_buffer,
+ but for REG_ICASE. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+build_wcs_upper_buffer (re_string_t *pstr)
+{
+ mbstate_t prev_st;
+ Idx src_idx, byte_idx, end_idx, remain_len;
+ size_t mbclen;
+#ifdef _LIBC
+ char buf[MB_LEN_MAX];
+ assert (MB_LEN_MAX >= pstr->mb_cur_max);
+#else
+ char buf[64];
+#endif
+
+ byte_idx = pstr->valid_len;
+ end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len;
+
+ /* The following optimization assumes that ASCII characters can be
+ mapped to wide characters with a simple cast. */
+ if (! pstr->map_notascii && pstr->trans == NULL && !pstr->offsets_needed)
+ {
+ while (byte_idx < end_idx)
+ {
+ wchar_t wc;
+
+ if (isascii (pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx])
+ && mbsinit (&pstr->cur_state))
+ {
+ /* In case of a singlebyte character. */
+ pstr->mbs[byte_idx]
+ = toupper (pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx]);
+ /* The next step uses the assumption that wchar_t is encoded
+ ASCII-safe: all ASCII values can be converted like this. */
+ pstr->wcs[byte_idx] = (wchar_t) pstr->mbs[byte_idx];
+ ++byte_idx;
+ continue;
+ }
+
+ remain_len = end_idx - byte_idx;
+ prev_st = pstr->cur_state;
+ mbclen = __mbrtowc (&wc,
+ ((const char *) pstr->raw_mbs + pstr->raw_mbs_idx
+ + byte_idx), remain_len, &pstr->cur_state);
+ if (BE (mbclen < (size_t) -2, 1))
+ {
+ wchar_t wcu = __towupper (wc);
+ if (wcu != wc)
+ {
+ size_t mbcdlen;
+
+ mbcdlen = __wcrtomb (buf, wcu, &prev_st);
+ if (BE (mbclen == mbcdlen, 1))
+ memcpy (pstr->mbs + byte_idx, buf, mbclen);
+ else
+ {
+ src_idx = byte_idx;
+ goto offsets_needed;
+ }
+ }
+ else
+ memcpy (pstr->mbs + byte_idx,
+ pstr->raw_mbs + pstr->raw_mbs_idx + byte_idx, mbclen);
+ pstr->wcs[byte_idx++] = wcu;
+ /* Write paddings. */
+ for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;)
+ pstr->wcs[byte_idx++] = WEOF;
+ }
+ else if (mbclen == (size_t) -1 || mbclen == 0
+ || (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len))
+ {
+ /* It is an invalid character, an incomplete character
+ at the end of the string, or '\0'. Just use the byte. */
+ int ch = pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx];
+ pstr->mbs[byte_idx] = ch;
+ /* And also cast it to wide char. */
+ pstr->wcs[byte_idx++] = (wchar_t) ch;
+ if (BE (mbclen == (size_t) -1, 0))
+ pstr->cur_state = prev_st;
+ }
+ else
+ {
+ /* The buffer doesn't have enough space, finish to build. */
+ pstr->cur_state = prev_st;
+ break;
+ }
+ }
+ pstr->valid_len = byte_idx;
+ pstr->valid_raw_len = byte_idx;
+ return REG_NOERROR;
+ }
+ else
+ for (src_idx = pstr->valid_raw_len; byte_idx < end_idx;)
+ {
+ wchar_t wc;
+ const char *p;
+ offsets_needed:
+ remain_len = end_idx - byte_idx;
+ prev_st = pstr->cur_state;
+ if (BE (pstr->trans != NULL, 0))
+ {
+ int i, ch;
+
+ for (i = 0; i < pstr->mb_cur_max && i < remain_len; ++i)
+ {
+ ch = pstr->raw_mbs [pstr->raw_mbs_idx + src_idx + i];
+ buf[i] = pstr->trans[ch];
+ }
+ p = (const char *) buf;
+ }
+ else
+ p = (const char *) pstr->raw_mbs + pstr->raw_mbs_idx + src_idx;
+ mbclen = __mbrtowc (&wc, p, remain_len, &pstr->cur_state);
+ if (BE (mbclen < (size_t) -2, 1))
+ {
+ wchar_t wcu = __towupper (wc);
+ if (wcu != wc)
+ {
+ size_t mbcdlen;
+
+ mbcdlen = __wcrtomb ((char *) buf, wcu, &prev_st);
+ if (BE (mbclen == mbcdlen, 1))
+ memcpy (pstr->mbs + byte_idx, buf, mbclen);
+ else if (mbcdlen != (size_t) -1)
+ {
+ size_t i;
+
+ if (byte_idx + mbcdlen > pstr->bufs_len)
+ {
+ pstr->cur_state = prev_st;
+ break;
+ }
+
+ if (pstr->offsets == NULL)
+ {
+ pstr->offsets = re_malloc (Idx, pstr->bufs_len);
+
+ if (pstr->offsets == NULL)
+ return REG_ESPACE;
+ }
+ if (!pstr->offsets_needed)
+ {
+ for (i = 0; i < (size_t) byte_idx; ++i)
+ pstr->offsets[i] = i;
+ pstr->offsets_needed = 1;
+ }
+
+ memcpy (pstr->mbs + byte_idx, buf, mbcdlen);
+ pstr->wcs[byte_idx] = wcu;
+ pstr->offsets[byte_idx] = src_idx;
+ for (i = 1; i < mbcdlen; ++i)
+ {
+ pstr->offsets[byte_idx + i]
+ = src_idx + (i < mbclen ? i : mbclen - 1);
+ pstr->wcs[byte_idx + i] = WEOF;
+ }
+ pstr->len += mbcdlen - mbclen;
+ if (pstr->raw_stop > src_idx)
+ pstr->stop += mbcdlen - mbclen;
+ end_idx = (pstr->bufs_len > pstr->len)
+ ? pstr->len : pstr->bufs_len;
+ byte_idx += mbcdlen;
+ src_idx += mbclen;
+ continue;
+ }
+ else
+ memcpy (pstr->mbs + byte_idx, p, mbclen);
+ }
+ else
+ memcpy (pstr->mbs + byte_idx, p, mbclen);
+
+ if (BE (pstr->offsets_needed != 0, 0))
+ {
+ size_t i;
+ for (i = 0; i < mbclen; ++i)
+ pstr->offsets[byte_idx + i] = src_idx + i;
+ }
+ src_idx += mbclen;
+
+ pstr->wcs[byte_idx++] = wcu;
+ /* Write paddings. */
+ for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;)
+ pstr->wcs[byte_idx++] = WEOF;
+ }
+ else if (mbclen == (size_t) -1 || mbclen == 0
+ || (mbclen == (size_t) -2 && pstr->bufs_len >= pstr->len))
+ {
+ /* It is an invalid character or '\0'. Just use the byte. */
+ int ch = pstr->raw_mbs[pstr->raw_mbs_idx + src_idx];
+
+ if (BE (pstr->trans != NULL, 0))
+ ch = pstr->trans [ch];
+ pstr->mbs[byte_idx] = ch;
+
+ if (BE (pstr->offsets_needed != 0, 0))
+ pstr->offsets[byte_idx] = src_idx;
+ ++src_idx;
+
+ /* And also cast it to wide char. */
+ pstr->wcs[byte_idx++] = (wchar_t) ch;
+ if (BE (mbclen == (size_t) -1, 0))
+ pstr->cur_state = prev_st;
+ }
+ else
+ {
+ /* The buffer doesn't have enough space, finish to build. */
+ pstr->cur_state = prev_st;
+ break;
+ }
+ }
+ pstr->valid_len = byte_idx;
+ pstr->valid_raw_len = src_idx;
+ return REG_NOERROR;
+}
+
+/* Skip characters until the index becomes greater than NEW_RAW_IDX.
+ Return the index. */
+
+static Idx
+re_string_skip_chars (re_string_t *pstr, Idx new_raw_idx, wint_t *last_wc)
+{
+ mbstate_t prev_st;
+ Idx rawbuf_idx;
+ size_t mbclen;
+ wint_t wc = WEOF;
+
+ /* Skip the characters which are not necessary to check. */
+ for (rawbuf_idx = pstr->raw_mbs_idx + pstr->valid_raw_len;
+ rawbuf_idx < new_raw_idx;)
+ {
+ wchar_t wc2;
+ Idx remain_len = pstr->raw_len - rawbuf_idx;
+ prev_st = pstr->cur_state;
+ mbclen = __mbrtowc (&wc2, (const char *) pstr->raw_mbs + rawbuf_idx,
+ remain_len, &pstr->cur_state);
+ if (BE (mbclen == (size_t) -2 || mbclen == (size_t) -1 || mbclen == 0,
0))
+ {
+ /* We treat these cases as a single byte character. */
+ if (mbclen == 0 || remain_len == 0)
+ wc = L'\0';
+ else
+ wc = *(unsigned char *) (pstr->raw_mbs + rawbuf_idx);
+ mbclen = 1;
+ pstr->cur_state = prev_st;
+ }
+ else
+ wc = wc2;
+ /* Then proceed the next character. */
+ rawbuf_idx += mbclen;
+ }
+ *last_wc = wc;
+ return rawbuf_idx;
+}
+#endif /* RE_ENABLE_I18N */
+
+/* Build the buffer PSTR->MBS, and apply the translation if we need.
+ This function is used in case of REG_ICASE. */
+
+static void
+build_upper_buffer (re_string_t *pstr)
+{
+ Idx char_idx, end_idx;
+ end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len;
+
+ for (char_idx = pstr->valid_len; char_idx < end_idx; ++char_idx)
+ {
+ int ch = pstr->raw_mbs[pstr->raw_mbs_idx + char_idx];
+ if (BE (pstr->trans != NULL, 0))
+ ch = pstr->trans[ch];
+ pstr->mbs[char_idx] = toupper (ch);
+ }
+ pstr->valid_len = char_idx;
+ pstr->valid_raw_len = char_idx;
+}
+
+/* Apply TRANS to the buffer in PSTR. */
+
+static void
+re_string_translate_buffer (re_string_t *pstr)
+{
+ Idx buf_idx, end_idx;
+ end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len;
+
+ for (buf_idx = pstr->valid_len; buf_idx < end_idx; ++buf_idx)
+ {
+ int ch = pstr->raw_mbs[pstr->raw_mbs_idx + buf_idx];
+ pstr->mbs[buf_idx] = pstr->trans[ch];
+ }
+
+ pstr->valid_len = buf_idx;
+ pstr->valid_raw_len = buf_idx;
+}
+
+/* This function re-construct the buffers.
+ Concretely, convert to wide character in case of pstr->mb_cur_max > 1,
+ convert to upper case in case of REG_ICASE, apply translation. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_string_reconstruct (re_string_t *pstr, Idx idx, int eflags)
+{
+ Idx offset;
+
+ if (BE (pstr->raw_mbs_idx <= idx, 0))
+ offset = idx - pstr->raw_mbs_idx;
+ else
+ {
+ /* Reset buffer. */
+#ifdef RE_ENABLE_I18N
+ if (pstr->mb_cur_max > 1)
+ memset (&pstr->cur_state, '\0', sizeof (mbstate_t));
+#endif /* RE_ENABLE_I18N */
+ pstr->len = pstr->raw_len;
+ pstr->stop = pstr->raw_stop;
+ pstr->valid_len = 0;
+ pstr->raw_mbs_idx = 0;
+ pstr->valid_raw_len = 0;
+ pstr->offsets_needed = 0;
+ pstr->tip_context = ((eflags & REG_NOTBOL) ? CONTEXT_BEGBUF
+ : CONTEXT_NEWLINE | CONTEXT_BEGBUF);
+ if (!pstr->mbs_allocated)
+ pstr->mbs = (unsigned char *) pstr->raw_mbs;
+ offset = idx;
+ }
+
+ if (BE (offset != 0, 1))
+ {
+ /* Should the already checked characters be kept? */
+ if (BE (offset < pstr->valid_raw_len, 1))
+ {
+ /* Yes, move them to the front of the buffer. */
+#ifdef RE_ENABLE_I18N
+ if (BE (pstr->offsets_needed, 0))
+ {
+ Idx low = 0, high = pstr->valid_len, mid;
+ do
+ {
+ mid = (high + low) / 2;
+ if (pstr->offsets[mid] > offset)
+ high = mid;
+ else if (pstr->offsets[mid] < offset)
+ low = mid + 1;
+ else
+ break;
+ }
+ while (low < high);
+ if (pstr->offsets[mid] < offset)
+ ++mid;
+ pstr->tip_context = re_string_context_at (pstr, mid - 1,
+ eflags);
+ /* This can be quite complicated, so handle specially
+ only the common and easy case where the character with
+ different length representation of lower and upper
+ case is present at or after offset. */
+ if (pstr->valid_len > offset
+ && mid == offset && pstr->offsets[mid] == offset)
+ {
+ memmove (pstr->wcs, pstr->wcs + offset,
+ (pstr->valid_len - offset) * sizeof (wint_t));
+ memmove (pstr->mbs, pstr->mbs + offset, pstr->valid_len -
offset);
+ pstr->valid_len -= offset;
+ pstr->valid_raw_len -= offset;
+ for (low = 0; low < pstr->valid_len; low++)
+ pstr->offsets[low] = pstr->offsets[low + offset] - offset;
+ }
+ else
+ {
+ /* Otherwise, just find out how long the partial multibyte
+ character at offset is and fill it with WEOF/255. */
+ pstr->len = pstr->raw_len - idx + offset;
+ pstr->stop = pstr->raw_stop - idx + offset;
+ pstr->offsets_needed = 0;
+ while (mid > 0 && pstr->offsets[mid - 1] == offset)
+ --mid;
+ while (mid < pstr->valid_len)
+ if (pstr->wcs[mid] != WEOF)
+ break;
+ else
+ ++mid;
+ if (mid == pstr->valid_len)
+ pstr->valid_len = 0;
+ else
+ {
+ pstr->valid_len = pstr->offsets[mid] - offset;
+ if (pstr->valid_len)
+ {
+ for (low = 0; low < pstr->valid_len; ++low)
+ pstr->wcs[low] = WEOF;
+ memset (pstr->mbs, 255, pstr->valid_len);
+ }
+ }
+ pstr->valid_raw_len = pstr->valid_len;
+ }
+ }
+ else
+#endif
+ {
+ pstr->tip_context = re_string_context_at (pstr, offset - 1,
+ eflags);
+#ifdef RE_ENABLE_I18N
+ if (pstr->mb_cur_max > 1)
+ memmove (pstr->wcs, pstr->wcs + offset,
+ (pstr->valid_len - offset) * sizeof (wint_t));
+#endif /* RE_ENABLE_I18N */
+ if (BE (pstr->mbs_allocated, 0))
+ memmove (pstr->mbs, pstr->mbs + offset,
+ pstr->valid_len - offset);
+ pstr->valid_len -= offset;
+ pstr->valid_raw_len -= offset;
+#if defined DEBUG && DEBUG
+ assert (pstr->valid_len > 0);
+#endif
+ }
+ }
+ else
+ {
+#ifdef RE_ENABLE_I18N
+ /* No, skip all characters until IDX. */
+ Idx prev_valid_len = pstr->valid_len;
+
+ if (BE (pstr->offsets_needed, 0))
+ {
+ pstr->len = pstr->raw_len - idx + offset;
+ pstr->stop = pstr->raw_stop - idx + offset;
+ pstr->offsets_needed = 0;
+ }
+#endif
+ pstr->valid_len = 0;
+#ifdef RE_ENABLE_I18N
+ if (pstr->mb_cur_max > 1)
+ {
+ Idx wcs_idx;
+ wint_t wc = WEOF;
+
+ if (pstr->is_utf8)
+ {
+ const unsigned char *raw, *p, *end;
+
+ /* Special case UTF-8. Multi-byte chars start with any
+ byte other than 0x80 - 0xbf. */
+ raw = pstr->raw_mbs + pstr->raw_mbs_idx;
+ end = raw + (offset - pstr->mb_cur_max);
+ if (end < pstr->raw_mbs)
+ end = pstr->raw_mbs;
+ p = raw + offset - 1;
+#ifdef _LIBC
+ /* We know the wchar_t encoding is UCS4, so for the simple
+ case, ASCII characters, skip the conversion step. */
+ if (isascii (*p) && BE (pstr->trans == NULL, 1))
+ {
+ memset (&pstr->cur_state, '\0', sizeof (mbstate_t));
+ /* pstr->valid_len = 0; */
+ wc = (wchar_t) *p;
+ }
+ else
+#endif
+ for (; p >= end; --p)
+ if ((*p & 0xc0) != 0x80)
+ {
+ mbstate_t cur_state;
+ wchar_t wc2;
+ Idx mlen = raw + pstr->len - p;
+ unsigned char buf[6];
+ size_t mbclen;
+
+ const unsigned char *pp = p;
+ if (BE (pstr->trans != NULL, 0))
+ {
+ int i = mlen < 6 ? mlen : 6;
+ while (--i >= 0)
+ buf[i] = pstr->trans[p[i]];
+ pp = buf;
+ }
+ /* XXX Don't use mbrtowc, we know which conversion
+ to use (UTF-8 -> UCS4). */
+ memset (&cur_state, 0, sizeof (cur_state));
+ mbclen = __mbrtowc (&wc2, (const char *) pp, mlen,
+ &cur_state);
+ if (raw + offset - p <= mbclen
+ && mbclen < (size_t) -2)
+ {
+ memset (&pstr->cur_state, '\0',
+ sizeof (mbstate_t));
+ pstr->valid_len = mbclen - (raw + offset - p);
+ wc = wc2;
+ }
+ break;
+ }
+ }
+
+ if (wc == WEOF)
+ pstr->valid_len = re_string_skip_chars (pstr, idx, &wc) - idx;
+ if (wc == WEOF)
+ pstr->tip_context
+ = re_string_context_at (pstr, prev_valid_len - 1, eflags);
+ else
+ pstr->tip_context = ((BE (pstr->word_ops_used != 0, 0)
+ && IS_WIDE_WORD_CHAR (wc))
+ ? CONTEXT_WORD
+ : ((IS_WIDE_NEWLINE (wc)
+ && pstr->newline_anchor)
+ ? CONTEXT_NEWLINE : 0));
+ if (BE (pstr->valid_len, 0))
+ {
+ for (wcs_idx = 0; wcs_idx < pstr->valid_len; ++wcs_idx)
+ pstr->wcs[wcs_idx] = WEOF;
+ if (pstr->mbs_allocated)
+ memset (pstr->mbs, 255, pstr->valid_len);
+ }
+ pstr->valid_raw_len = pstr->valid_len;
+ }
+ else
+#endif /* RE_ENABLE_I18N */
+ {
+ int c = pstr->raw_mbs[pstr->raw_mbs_idx + offset - 1];
+ pstr->valid_raw_len = 0;
+ if (pstr->trans)
+ c = pstr->trans[c];
+ pstr->tip_context = (bitset_contain (pstr->word_char, c)
+ ? CONTEXT_WORD
+ : ((IS_NEWLINE (c) && pstr->newline_anchor)
+ ? CONTEXT_NEWLINE : 0));
+ }
+ }
+ if (!BE (pstr->mbs_allocated, 0))
+ pstr->mbs += offset;
+ }
+ pstr->raw_mbs_idx = idx;
+ pstr->len -= offset;
+ pstr->stop -= offset;
+
+ /* Then build the buffers. */
+#ifdef RE_ENABLE_I18N
+ if (pstr->mb_cur_max > 1)
+ {
+ if (pstr->icase)
+ {
+ reg_errcode_t ret = build_wcs_upper_buffer (pstr);
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+ }
+ else
+ build_wcs_buffer (pstr);
+ }
+ else
+#endif /* RE_ENABLE_I18N */
+ if (BE (pstr->mbs_allocated, 0))
+ {
+ if (pstr->icase)
+ build_upper_buffer (pstr);
+ else if (pstr->trans != NULL)
+ re_string_translate_buffer (pstr);
+ }
+ else
+ pstr->valid_len = pstr->len;
+
+ pstr->cur_idx = 0;
+ return REG_NOERROR;
+}
+
+static unsigned char
+__attribute__ ((pure))
+re_string_peek_byte_case (const re_string_t *pstr, Idx idx)
+{
+ int ch;
+ Idx off;
+
+ /* Handle the common (easiest) cases first. */
+ if (BE (!pstr->mbs_allocated, 1))
+ return re_string_peek_byte (pstr, idx);
+
+#ifdef RE_ENABLE_I18N
+ if (pstr->mb_cur_max > 1
+ && ! re_string_is_single_byte_char (pstr, pstr->cur_idx + idx))
+ return re_string_peek_byte (pstr, idx);
+#endif
+
+ off = pstr->cur_idx + idx;
+#ifdef RE_ENABLE_I18N
+ if (pstr->offsets_needed)
+ off = pstr->offsets[off];
+#endif
+
+ ch = pstr->raw_mbs[pstr->raw_mbs_idx + off];
+
+#ifdef RE_ENABLE_I18N
+ /* Ensure that e.g. for tr_TR.UTF-8 BACKSLASH DOTLESS SMALL LETTER I
+ this function returns CAPITAL LETTER I instead of first byte of
+ DOTLESS SMALL LETTER I. The latter would confuse the parser,
+ since peek_byte_case doesn't advance cur_idx in any way. */
+ if (pstr->offsets_needed && !isascii (ch))
+ return re_string_peek_byte (pstr, idx);
+#endif
+
+ return ch;
+}
+
+static unsigned char
+re_string_fetch_byte_case (re_string_t *pstr)
+{
+ if (BE (!pstr->mbs_allocated, 1))
+ return re_string_fetch_byte (pstr);
+
+#ifdef RE_ENABLE_I18N
+ if (pstr->offsets_needed)
+ {
+ Idx off;
+ int ch;
+
+ /* For tr_TR.UTF-8 [[:islower:]] there is
+ [[: CAPITAL LETTER I WITH DOT lower:]] in mbs. Skip
+ in that case the whole multi-byte character and return
+ the original letter. On the other side, with
+ [[: DOTLESS SMALL LETTER I return [[:I, as doing
+ anything else would complicate things too much. */
+
+ if (!re_string_first_byte (pstr, pstr->cur_idx))
+ return re_string_fetch_byte (pstr);
+
+ off = pstr->offsets[pstr->cur_idx];
+ ch = pstr->raw_mbs[pstr->raw_mbs_idx + off];
+
+ if (! isascii (ch))
+ return re_string_fetch_byte (pstr);
+
+ re_string_skip_bytes (pstr,
+ re_string_char_size_at (pstr, pstr->cur_idx));
+ return ch;
+ }
+#endif
+
+ return pstr->raw_mbs[pstr->raw_mbs_idx + pstr->cur_idx++];
+}
+
+static void
+re_string_destruct (re_string_t *pstr)
+{
+#ifdef RE_ENABLE_I18N
+ re_free (pstr->wcs);
+ re_free (pstr->offsets);
+#endif /* RE_ENABLE_I18N */
+ if (pstr->mbs_allocated)
+ re_free (pstr->mbs);
+}
+
+/* Return the context at IDX in INPUT. */
+
+static unsigned int
+re_string_context_at (const re_string_t *input, Idx idx, int eflags)
+{
+ int c;
+ if (BE (idx < 0, 0))
+ /* In this case, we use the value stored in input->tip_context,
+ since we can't know the character in input->mbs[-1] here. */
+ return input->tip_context;
+ if (BE (idx == input->len, 0))
+ return ((eflags & REG_NOTEOL) ? CONTEXT_ENDBUF
+ : CONTEXT_NEWLINE | CONTEXT_ENDBUF);
+#ifdef RE_ENABLE_I18N
+ if (input->mb_cur_max > 1)
+ {
+ wint_t wc;
+ Idx wc_idx = idx;
+ while(input->wcs[wc_idx] == WEOF)
+ {
+#if defined DEBUG && DEBUG
+ /* It must not happen. */
+ assert (wc_idx >= 0);
+#endif
+ --wc_idx;
+ if (wc_idx < 0)
+ return input->tip_context;
+ }
+ wc = input->wcs[wc_idx];
+ if (BE (input->word_ops_used != 0, 0) && IS_WIDE_WORD_CHAR (wc))
+ return CONTEXT_WORD;
+ return (IS_WIDE_NEWLINE (wc) && input->newline_anchor
+ ? CONTEXT_NEWLINE : 0);
+ }
+ else
+#endif
+ {
+ c = re_string_byte_at (input, idx);
+ if (bitset_contain (input->word_char, c))
+ return CONTEXT_WORD;
+ return IS_NEWLINE (c) && input->newline_anchor ? CONTEXT_NEWLINE : 0;
+ }
+}
+
+/* Functions for set operation. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_node_set_alloc (re_node_set *set, Idx size)
+{
+ set->alloc = size;
+ set->nelem = 0;
+ set->elems = re_malloc (Idx, size);
+ if (BE (set->elems == NULL, 0) && (MALLOC_0_IS_NONNULL || size != 0))
+ return REG_ESPACE;
+ return REG_NOERROR;
+}
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_node_set_init_1 (re_node_set *set, Idx elem)
+{
+ set->alloc = 1;
+ set->nelem = 1;
+ set->elems = re_malloc (Idx, 1);
+ if (BE (set->elems == NULL, 0))
+ {
+ set->alloc = set->nelem = 0;
+ return REG_ESPACE;
+ }
+ set->elems[0] = elem;
+ return REG_NOERROR;
+}
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_node_set_init_2 (re_node_set *set, Idx elem1, Idx elem2)
+{
+ set->alloc = 2;
+ set->elems = re_malloc (Idx, 2);
+ if (BE (set->elems == NULL, 0))
+ return REG_ESPACE;
+ if (elem1 == elem2)
+ {
+ set->nelem = 1;
+ set->elems[0] = elem1;
+ }
+ else
+ {
+ set->nelem = 2;
+ if (elem1 < elem2)
+ {
+ set->elems[0] = elem1;
+ set->elems[1] = elem2;
+ }
+ else
+ {
+ set->elems[0] = elem2;
+ set->elems[1] = elem1;
+ }
+ }
+ return REG_NOERROR;
+}
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_node_set_init_copy (re_node_set *dest, const re_node_set *src)
+{
+ dest->nelem = src->nelem;
+ if (src->nelem > 0)
+ {
+ dest->alloc = dest->nelem;
+ dest->elems = re_malloc (Idx, dest->alloc);
+ if (BE (dest->elems == NULL, 0))
+ {
+ dest->alloc = dest->nelem = 0;
+ return REG_ESPACE;
+ }
+ memcpy (dest->elems, src->elems, src->nelem * sizeof (Idx));
+ }
+ else
+ re_node_set_init_empty (dest);
+ return REG_NOERROR;
+}
+
+/* Calculate the intersection of the sets SRC1 and SRC2. And merge it to
+ DEST. Return value indicate the error code or REG_NOERROR if succeeded.
+ Note: We assume dest->elems is NULL, when dest->alloc is 0. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_node_set_add_intersect (re_node_set *dest, const re_node_set *src1,
+ const re_node_set *src2)
+{
+ Idx i1, i2, is, id, delta, sbase;
+ if (src1->nelem == 0 || src2->nelem == 0)
+ return REG_NOERROR;
+
+ /* We need dest->nelem + 2 * elems_in_intersection; this is a
+ conservative estimate. */
+ if (src1->nelem + src2->nelem + dest->nelem > dest->alloc)
+ {
+ Idx new_alloc = src1->nelem + src2->nelem + dest->alloc;
+ Idx *new_elems = re_realloc (dest->elems, Idx, new_alloc);
+ if (BE (new_elems == NULL, 0))
+ return REG_ESPACE;
+ dest->elems = new_elems;
+ dest->alloc = new_alloc;
+ }
+
+ /* Find the items in the intersection of SRC1 and SRC2, and copy
+ into the top of DEST those that are not already in DEST itself. */
+ sbase = dest->nelem + src1->nelem + src2->nelem;
+ i1 = src1->nelem - 1;
+ i2 = src2->nelem - 1;
+ id = dest->nelem - 1;
+ for (;;)
+ {
+ if (src1->elems[i1] == src2->elems[i2])
+ {
+ /* Try to find the item in DEST. Maybe we could binary search? */
+ while (id >= 0 && dest->elems[id] > src1->elems[i1])
+ --id;
+
+ if (id < 0 || dest->elems[id] != src1->elems[i1])
+ dest->elems[--sbase] = src1->elems[i1];
+
+ if (--i1 < 0 || --i2 < 0)
+ break;
+ }
+
+ /* Lower the highest of the two items. */
+ else if (src1->elems[i1] < src2->elems[i2])
+ {
+ if (--i2 < 0)
+ break;
+ }
+ else
+ {
+ if (--i1 < 0)
+ break;
+ }
+ }
+
+ id = dest->nelem - 1;
+ is = dest->nelem + src1->nelem + src2->nelem - 1;
+ delta = is - sbase + 1;
+
+ /* Now copy. When DELTA becomes zero, the remaining
+ DEST elements are already in place; this is more or
+ less the same loop that is in re_node_set_merge. */
+ dest->nelem += delta;
+ if (delta > 0 && id >= 0)
+ for (;;)
+ {
+ if (dest->elems[is] > dest->elems[id])
+ {
+ /* Copy from the top. */
+ dest->elems[id + delta--] = dest->elems[is--];
+ if (delta == 0)
+ break;
+ }
+ else
+ {
+ /* Slide from the bottom. */
+ dest->elems[id + delta] = dest->elems[id];
+ if (--id < 0)
+ break;
+ }
+ }
+
+ /* Copy remaining SRC elements. */
+ memcpy (dest->elems, dest->elems + sbase, delta * sizeof (Idx));
+
+ return REG_NOERROR;
+}
+
+/* Calculate the union set of the sets SRC1 and SRC2. And store it to
+ DEST. Return value indicate the error code or REG_NOERROR if succeeded. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_node_set_init_union (re_node_set *dest, const re_node_set *src1,
+ const re_node_set *src2)
+{
+ Idx i1, i2, id;
+ if (src1 != NULL && src1->nelem > 0 && src2 != NULL && src2->nelem > 0)
+ {
+ dest->alloc = src1->nelem + src2->nelem;
+ dest->elems = re_malloc (Idx, dest->alloc);
+ if (BE (dest->elems == NULL, 0))
+ return REG_ESPACE;
+ }
+ else
+ {
+ if (src1 != NULL && src1->nelem > 0)
+ return re_node_set_init_copy (dest, src1);
+ else if (src2 != NULL && src2->nelem > 0)
+ return re_node_set_init_copy (dest, src2);
+ else
+ re_node_set_init_empty (dest);
+ return REG_NOERROR;
+ }
+ for (i1 = i2 = id = 0 ; i1 < src1->nelem && i2 < src2->nelem ;)
+ {
+ if (src1->elems[i1] > src2->elems[i2])
+ {
+ dest->elems[id++] = src2->elems[i2++];
+ continue;
+ }
+ if (src1->elems[i1] == src2->elems[i2])
+ ++i2;
+ dest->elems[id++] = src1->elems[i1++];
+ }
+ if (i1 < src1->nelem)
+ {
+ memcpy (dest->elems + id, src1->elems + i1,
+ (src1->nelem - i1) * sizeof (Idx));
+ id += src1->nelem - i1;
+ }
+ else if (i2 < src2->nelem)
+ {
+ memcpy (dest->elems + id, src2->elems + i2,
+ (src2->nelem - i2) * sizeof (Idx));
+ id += src2->nelem - i2;
+ }
+ dest->nelem = id;
+ return REG_NOERROR;
+}
+
+/* Calculate the union set of the sets DEST and SRC. And store it to
+ DEST. Return value indicate the error code or REG_NOERROR if succeeded. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_node_set_merge (re_node_set *dest, const re_node_set *src)
+{
+ Idx is, id, sbase, delta;
+ if (src == NULL || src->nelem == 0)
+ return REG_NOERROR;
+ if (dest->alloc < 2 * src->nelem + dest->nelem)
+ {
+ Idx new_alloc = 2 * (src->nelem + dest->alloc);
+ Idx *new_buffer = re_realloc (dest->elems, Idx, new_alloc);
+ if (BE (new_buffer == NULL, 0))
+ return REG_ESPACE;
+ dest->elems = new_buffer;
+ dest->alloc = new_alloc;
+ }
+
+ if (BE (dest->nelem == 0, 0))
+ {
+ dest->nelem = src->nelem;
+ memcpy (dest->elems, src->elems, src->nelem * sizeof (Idx));
+ return REG_NOERROR;
+ }
+
+ /* Copy into the top of DEST the items of SRC that are not
+ found in DEST. Maybe we could binary search in DEST? */
+ for (sbase = dest->nelem + 2 * src->nelem,
+ is = src->nelem - 1, id = dest->nelem - 1; is >= 0 && id >= 0; )
+ {
+ if (dest->elems[id] == src->elems[is])
+ is--, id--;
+ else if (dest->elems[id] < src->elems[is])
+ dest->elems[--sbase] = src->elems[is--];
+ else /* if (dest->elems[id] > src->elems[is]) */
+ --id;
+ }
+
+ if (is >= 0)
+ {
+ /* If DEST is exhausted, the remaining items of SRC must be unique. */
+ sbase -= is + 1;
+ memcpy (dest->elems + sbase, src->elems, (is + 1) * sizeof (Idx));
+ }
+
+ id = dest->nelem - 1;
+ is = dest->nelem + 2 * src->nelem - 1;
+ delta = is - sbase + 1;
+ if (delta == 0)
+ return REG_NOERROR;
+
+ /* Now copy. When DELTA becomes zero, the remaining
+ DEST elements are already in place. */
+ dest->nelem += delta;
+ for (;;)
+ {
+ if (dest->elems[is] > dest->elems[id])
+ {
+ /* Copy from the top. */
+ dest->elems[id + delta--] = dest->elems[is--];
+ if (delta == 0)
+ break;
+ }
+ else
+ {
+ /* Slide from the bottom. */
+ dest->elems[id + delta] = dest->elems[id];
+ if (--id < 0)
+ {
+ /* Copy remaining SRC elements. */
+ memcpy (dest->elems, dest->elems + sbase,
+ delta * sizeof (Idx));
+ break;
+ }
+ }
+ }
+
+ return REG_NOERROR;
+}
+
+/* Insert the new element ELEM to the re_node_set* SET.
+ SET should not already have ELEM.
+ Return true if successful. */
+
+static bool
+__attribute_warn_unused_result__
+re_node_set_insert (re_node_set *set, Idx elem)
+{
+ Idx idx;
+ /* In case the set is empty. */
+ if (set->alloc == 0)
+ return BE (re_node_set_init_1 (set, elem) == REG_NOERROR, 1);
+
+ if (BE (set->nelem, 0) == 0)
+ {
+ /* We already guaranteed above that set->alloc != 0. */
+ set->elems[0] = elem;
+ ++set->nelem;
+ return true;
+ }
+
+ /* Realloc if we need. */
+ if (set->alloc == set->nelem)
+ {
+ Idx *new_elems;
+ set->alloc = set->alloc * 2;
+ new_elems = re_realloc (set->elems, Idx, set->alloc);
+ if (BE (new_elems == NULL, 0))
+ return false;
+ set->elems = new_elems;
+ }
+
+ /* Move the elements which follows the new element. Test the
+ first element separately to skip a check in the inner loop. */
+ if (elem < set->elems[0])
+ {
+ idx = 0;
+ for (idx = set->nelem; idx > 0; idx--)
+ set->elems[idx] = set->elems[idx - 1];
+ }
+ else
+ {
+ for (idx = set->nelem; set->elems[idx - 1] > elem; idx--)
+ set->elems[idx] = set->elems[idx - 1];
+ }
+
+ /* Insert the new element. */
+ set->elems[idx] = elem;
+ ++set->nelem;
+ return true;
+}
+
+/* Insert the new element ELEM to the re_node_set* SET.
+ SET should not already have any element greater than or equal to ELEM.
+ Return true if successful. */
+
+static bool
+__attribute_warn_unused_result__
+re_node_set_insert_last (re_node_set *set, Idx elem)
+{
+ /* Realloc if we need. */
+ if (set->alloc == set->nelem)
+ {
+ Idx *new_elems;
+ set->alloc = (set->alloc + 1) * 2;
+ new_elems = re_realloc (set->elems, Idx, set->alloc);
+ if (BE (new_elems == NULL, 0))
+ return false;
+ set->elems = new_elems;
+ }
+
+ /* Insert the new element. */
+ set->elems[set->nelem++] = elem;
+ return true;
+}
+
+/* Compare two node sets SET1 and SET2.
+ Return true if SET1 and SET2 are equivalent. */
+
+static bool
+__attribute__ ((pure))
+re_node_set_compare (const re_node_set *set1, const re_node_set *set2)
+{
+ Idx i;
+ if (set1 == NULL || set2 == NULL || set1->nelem != set2->nelem)
+ return false;
+ for (i = set1->nelem ; --i >= 0 ; )
+ if (set1->elems[i] != set2->elems[i])
+ return false;
+ return true;
+}
+
+/* Return (idx + 1) if SET contains the element ELEM, return 0 otherwise. */
+
+static Idx
+__attribute__ ((pure))
+re_node_set_contains (const re_node_set *set, Idx elem)
+{
+ __re_size_t idx, right, mid;
+ if (set->nelem <= 0)
+ return 0;
+
+ /* Binary search the element. */
+ idx = 0;
+ right = set->nelem - 1;
+ while (idx < right)
+ {
+ mid = (idx + right) / 2;
+ if (set->elems[mid] < elem)
+ idx = mid + 1;
+ else
+ right = mid;
+ }
+ return set->elems[idx] == elem ? idx + 1 : 0;
+}
+
+static void
+re_node_set_remove_at (re_node_set *set, Idx idx)
+{
+ if (idx < 0 || idx >= set->nelem)
+ return;
+ --set->nelem;
+ for (; idx < set->nelem; idx++)
+ set->elems[idx] = set->elems[idx + 1];
+}
+
+
+/* Add the token TOKEN to dfa->nodes, and return the index of the token.
+ Or return -1 if an error occurred. */
+
+static Idx
+re_dfa_add_node (re_dfa_t *dfa, re_token_t token)
+{
+ if (BE (dfa->nodes_len >= dfa->nodes_alloc, 0))
+ {
+ size_t new_nodes_alloc = dfa->nodes_alloc * 2;
+ Idx *new_nexts, *new_indices;
+ re_node_set *new_edests, *new_eclosures;
+ re_token_t *new_nodes;
+
+ /* Avoid overflows in realloc. */
+ const size_t max_object_size = MAX (sizeof (re_token_t),
+ MAX (sizeof (re_node_set),
+ sizeof (Idx)));
+ if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) < new_nodes_alloc, 0))
+ return -1;
+
+ new_nodes = re_realloc (dfa->nodes, re_token_t, new_nodes_alloc);
+ if (BE (new_nodes == NULL, 0))
+ return -1;
+ dfa->nodes = new_nodes;
+ new_nexts = re_realloc (dfa->nexts, Idx, new_nodes_alloc);
+ new_indices = re_realloc (dfa->org_indices, Idx, new_nodes_alloc);
+ new_edests = re_realloc (dfa->edests, re_node_set, new_nodes_alloc);
+ new_eclosures = re_realloc (dfa->eclosures, re_node_set,
new_nodes_alloc);
+ if (BE (new_nexts == NULL || new_indices == NULL
+ || new_edests == NULL || new_eclosures == NULL, 0))
+ {
+ re_free (new_nexts);
+ re_free (new_indices);
+ re_free (new_edests);
+ re_free (new_eclosures);
+ return -1;
+ }
+ dfa->nexts = new_nexts;
+ dfa->org_indices = new_indices;
+ dfa->edests = new_edests;
+ dfa->eclosures = new_eclosures;
+ dfa->nodes_alloc = new_nodes_alloc;
+ }
+ dfa->nodes[dfa->nodes_len] = token;
+ dfa->nodes[dfa->nodes_len].constraint = 0;
+#ifdef RE_ENABLE_I18N
+ dfa->nodes[dfa->nodes_len].accept_mb =
+ ((token.type == OP_PERIOD && dfa->mb_cur_max > 1)
+ || token.type == COMPLEX_BRACKET);
+#endif
+ dfa->nexts[dfa->nodes_len] = -1;
+ re_node_set_init_empty (dfa->edests + dfa->nodes_len);
+ re_node_set_init_empty (dfa->eclosures + dfa->nodes_len);
+ return dfa->nodes_len++;
+}
+
+static re_hashval_t
+calc_state_hash (const re_node_set *nodes, unsigned int context)
+{
+ re_hashval_t hash = nodes->nelem + context;
+ Idx i;
+ for (i = 0 ; i < nodes->nelem ; i++)
+ hash += nodes->elems[i];
+ return hash;
+}
+
+/* Search for the state whose node_set is equivalent to NODES.
+ Return the pointer to the state, if we found it in the DFA.
+ Otherwise create the new one and return it. In case of an error
+ return NULL and set the error code in ERR.
+ Note: - We assume NULL as the invalid state, then it is possible that
+ return value is NULL and ERR is REG_NOERROR.
+ - We never return non-NULL value in case of any errors, it is for
+ optimization. */
+
+static re_dfastate_t *
+__attribute_warn_unused_result__
+re_acquire_state (reg_errcode_t *err, const re_dfa_t *dfa,
+ const re_node_set *nodes)
+{
+ re_hashval_t hash;
+ re_dfastate_t *new_state;
+ struct re_state_table_entry *spot;
+ Idx i;
+#if defined GCC_LINT || defined lint
+ /* Suppress bogus uninitialized-variable warnings. */
+ *err = REG_NOERROR;
+#endif
+ if (BE (nodes->nelem == 0, 0))
+ {
+ *err = REG_NOERROR;
+ return NULL;
+ }
+ hash = calc_state_hash (nodes, 0);
+ spot = dfa->state_table + (hash & dfa->state_hash_mask);
+
+ for (i = 0 ; i < spot->num ; i++)
+ {
+ re_dfastate_t *state = spot->array[i];
+ if (hash != state->hash)
+ continue;
+ if (re_node_set_compare (&state->nodes, nodes))
+ return state;
+ }
+
+ /* There are no appropriate state in the dfa, create the new one. */
+ new_state = create_ci_newstate (dfa, nodes, hash);
+ if (BE (new_state == NULL, 0))
+ *err = REG_ESPACE;
+
+ return new_state;
+}
+
+/* Search for the state whose node_set is equivalent to NODES and
+ whose context is equivalent to CONTEXT.
+ Return the pointer to the state, if we found it in the DFA.
+ Otherwise create the new one and return it. In case of an error
+ return NULL and set the error code in ERR.
+ Note: - We assume NULL as the invalid state, then it is possible that
+ return value is NULL and ERR is REG_NOERROR.
+ - We never return non-NULL value in case of any errors, it is for
+ optimization. */
+
+static re_dfastate_t *
+__attribute_warn_unused_result__
+re_acquire_state_context (reg_errcode_t *err, const re_dfa_t *dfa,
+ const re_node_set *nodes, unsigned int context)
+{
+ re_hashval_t hash;
+ re_dfastate_t *new_state;
+ struct re_state_table_entry *spot;
+ Idx i;
+#if defined GCC_LINT || defined lint
+ /* Suppress bogus uninitialized-variable warnings. */
+ *err = REG_NOERROR;
+#endif
+ if (nodes->nelem == 0)
+ {
+ *err = REG_NOERROR;
+ return NULL;
+ }
+ hash = calc_state_hash (nodes, context);
+ spot = dfa->state_table + (hash & dfa->state_hash_mask);
+
+ for (i = 0 ; i < spot->num ; i++)
+ {
+ re_dfastate_t *state = spot->array[i];
+ if (state->hash == hash
+ && state->context == context
+ && re_node_set_compare (state->entrance_nodes, nodes))
+ return state;
+ }
+ /* There are no appropriate state in 'dfa', create the new one. */
+ new_state = create_cd_newstate (dfa, nodes, context, hash);
+ if (BE (new_state == NULL, 0))
+ *err = REG_ESPACE;
+
+ return new_state;
+}
+
+/* Finish initialization of the new state NEWSTATE, and using its hash value
+ HASH put in the appropriate bucket of DFA's state table. Return value
+ indicates the error code if failed. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+register_state (const re_dfa_t *dfa, re_dfastate_t *newstate,
+ re_hashval_t hash)
+{
+ struct re_state_table_entry *spot;
+ reg_errcode_t err;
+ Idx i;
+
+ newstate->hash = hash;
+ err = re_node_set_alloc (&newstate->non_eps_nodes, newstate->nodes.nelem);
+ if (BE (err != REG_NOERROR, 0))
+ return REG_ESPACE;
+ for (i = 0; i < newstate->nodes.nelem; i++)
+ {
+ Idx elem = newstate->nodes.elems[i];
+ if (!IS_EPSILON_NODE (dfa->nodes[elem].type))
+ if (! re_node_set_insert_last (&newstate->non_eps_nodes, elem))
+ return REG_ESPACE;
+ }
+
+ spot = dfa->state_table + (hash & dfa->state_hash_mask);
+ if (BE (spot->alloc <= spot->num, 0))
+ {
+ Idx new_alloc = 2 * spot->num + 2;
+ re_dfastate_t **new_array = re_realloc (spot->array, re_dfastate_t *,
+ new_alloc);
+ if (BE (new_array == NULL, 0))
+ return REG_ESPACE;
+ spot->array = new_array;
+ spot->alloc = new_alloc;
+ }
+ spot->array[spot->num++] = newstate;
+ return REG_NOERROR;
+}
+
+static void
+free_state (re_dfastate_t *state)
+{
+ re_node_set_free (&state->non_eps_nodes);
+ re_node_set_free (&state->inveclosure);
+ if (state->entrance_nodes != &state->nodes)
+ {
+ re_node_set_free (state->entrance_nodes);
+ re_free (state->entrance_nodes);
+ }
+ re_node_set_free (&state->nodes);
+ re_free (state->word_trtable);
+ re_free (state->trtable);
+ re_free (state);
+}
+
+/* Create the new state which is independent of contexts.
+ Return the new state if succeeded, otherwise return NULL. */
+
+static re_dfastate_t *
+__attribute_warn_unused_result__
+create_ci_newstate (const re_dfa_t *dfa, const re_node_set *nodes,
+ re_hashval_t hash)
+{
+ Idx i;
+ reg_errcode_t err;
+ re_dfastate_t *newstate;
+
+ newstate = (re_dfastate_t *) calloc (sizeof (re_dfastate_t), 1);
+ if (BE (newstate == NULL, 0))
+ return NULL;
+ err = re_node_set_init_copy (&newstate->nodes, nodes);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_free (newstate);
+ return NULL;
+ }
+
+ newstate->entrance_nodes = &newstate->nodes;
+ for (i = 0 ; i < nodes->nelem ; i++)
+ {
+ re_token_t *node = dfa->nodes + nodes->elems[i];
+ re_token_type_t type = node->type;
+ if (type == CHARACTER && !node->constraint)
+ continue;
+#ifdef RE_ENABLE_I18N
+ newstate->accept_mb |= node->accept_mb;
+#endif /* RE_ENABLE_I18N */
+
+ /* If the state has the halt node, the state is a halt state. */
+ if (type == END_OF_RE)
+ newstate->halt = 1;
+ else if (type == OP_BACK_REF)
+ newstate->has_backref = 1;
+ else if (type == ANCHOR || node->constraint)
+ newstate->has_constraint = 1;
+ }
+ err = register_state (dfa, newstate, hash);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ free_state (newstate);
+ newstate = NULL;
+ }
+ return newstate;
+}
+
+/* Create the new state which is depend on the context CONTEXT.
+ Return the new state if succeeded, otherwise return NULL. */
+
+static re_dfastate_t *
+__attribute_warn_unused_result__
+create_cd_newstate (const re_dfa_t *dfa, const re_node_set *nodes,
+ unsigned int context, re_hashval_t hash)
+{
+ Idx i, nctx_nodes = 0;
+ reg_errcode_t err;
+ re_dfastate_t *newstate;
+
+ newstate = (re_dfastate_t *) calloc (sizeof (re_dfastate_t), 1);
+ if (BE (newstate == NULL, 0))
+ return NULL;
+ err = re_node_set_init_copy (&newstate->nodes, nodes);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_free (newstate);
+ return NULL;
+ }
+
+ newstate->context = context;
+ newstate->entrance_nodes = &newstate->nodes;
+
+ for (i = 0 ; i < nodes->nelem ; i++)
+ {
+ re_token_t *node = dfa->nodes + nodes->elems[i];
+ re_token_type_t type = node->type;
+ unsigned int constraint = node->constraint;
+
+ if (type == CHARACTER && !constraint)
+ continue;
+#ifdef RE_ENABLE_I18N
+ newstate->accept_mb |= node->accept_mb;
+#endif /* RE_ENABLE_I18N */
+
+ /* If the state has the halt node, the state is a halt state. */
+ if (type == END_OF_RE)
+ newstate->halt = 1;
+ else if (type == OP_BACK_REF)
+ newstate->has_backref = 1;
+
+ if (constraint)
+ {
+ if (newstate->entrance_nodes == &newstate->nodes)
+ {
+ newstate->entrance_nodes = re_malloc (re_node_set, 1);
+ if (BE (newstate->entrance_nodes == NULL, 0))
+ {
+ free_state (newstate);
+ return NULL;
+ }
+ if (re_node_set_init_copy (newstate->entrance_nodes, nodes)
+ != REG_NOERROR)
+ return NULL;
+ nctx_nodes = 0;
+ newstate->has_constraint = 1;
+ }
+
+ if (NOT_SATISFY_PREV_CONSTRAINT (constraint,context))
+ {
+ re_node_set_remove_at (&newstate->nodes, i - nctx_nodes);
+ ++nctx_nodes;
+ }
+ }
+ }
+ err = register_state (dfa, newstate, hash);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ free_state (newstate);
+ newstate = NULL;
+ }
+ return newstate;
+}
diff --git a/lib/regex_internal.h b/lib/regex_internal.h
new file mode 100644
index 0000000..7bbe802
--- /dev/null
+++ b/lib/regex_internal.h
@@ -0,0 +1,911 @@
+/* Extended regular expression matching and search library.
+ Copyright (C) 2002-2018 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+ Contributed by Isamu Hasegawa <address@hidden>.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU General Public
+ License as published by the Free Software Foundation; either
+ version 3 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ General Public License for more details.
+
+ You should have received a copy of the GNU General Public
+ License along with the GNU C Library; if not, see
+ <https://www.gnu.org/licenses/>. */
+
+#ifndef _REGEX_INTERNAL_H
+#define _REGEX_INTERNAL_H 1
+
+#include <assert.h>
+#include <ctype.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include <langinfo.h>
+#include <locale.h>
+#include <wchar.h>
+#include <wctype.h>
+#include <stdbool.h>
+#include <stdint.h>
+
+/* Properties of integers. Although Gnulib has intprops.h, glibc does
+ without for now. */
+#ifndef _LIBC
+# include "intprops.h"
+#else
+/* True if the real type T is signed. */
+# define TYPE_SIGNED(t) (! ((t) 0 < (t) -1))
+
+/* True if adding the nonnegative Idx values A and B would overflow.
+ If false, set *R to A + B. A, B, and R may be evaluated more than
+ once, or zero times. Although this is not a full implementation of
+ Gnulib INT_ADD_WRAPV, it is good enough for glibc regex code.
+ FIXME: This implementation is a fragile stopgap, and this file would
+ be simpler and more robust if intprops.h were migrated into glibc. */
+# define INT_ADD_WRAPV(a, b, r) \
+ (IDX_MAX - (a) < (b) ? true : (*(r) = (a) + (b), false))
+#endif
+
+#ifdef _LIBC
+# include <libc-lock.h>
+# define lock_define(name) __libc_lock_define (, name)
+# define lock_init(lock) (__libc_lock_init (lock), 0)
+# define lock_fini(lock) ((void) 0)
+# define lock_lock(lock) __libc_lock_lock (lock)
+# define lock_unlock(lock) __libc_lock_unlock (lock)
+#elif defined GNULIB_LOCK && !defined USE_UNLOCKED_IO
+# include "glthread/lock.h"
+ /* Use gl_lock_define if empty macro arguments are known to work.
+ Otherwise, fall back on less-portable substitutes. */
+# if ((defined __GNUC__ && !defined __STRICT_ANSI__) \
+ || (defined __STDC_VERSION__ && 199901L <= __STDC_VERSION__))
+# define lock_define(name) gl_lock_define (, name)
+# elif USE_POSIX_THREADS
+# define lock_define(name) pthread_mutex_t name;
+# elif USE_PTH_THREADS
+# define lock_define(name) pth_mutex_t name;
+# elif USE_SOLARIS_THREADS
+# define lock_define(name) mutex_t name;
+# elif USE_WINDOWS_THREADS
+# define lock_define(name) gl_lock_t name;
+# else
+# define lock_define(name)
+# endif
+# define lock_init(lock) glthread_lock_init (&(lock))
+# define lock_fini(lock) glthread_lock_destroy (&(lock))
+# define lock_lock(lock) glthread_lock_lock (&(lock))
+# define lock_unlock(lock) glthread_lock_unlock (&(lock))
+#elif defined GNULIB_PTHREAD && !defined USE_UNLOCKED_IO
+# include <pthread.h>
+# define lock_define(name) pthread_mutex_t name;
+# define lock_init(lock) pthread_mutex_init (&(lock), 0)
+# define lock_fini(lock) pthread_mutex_destroy (&(lock))
+# define lock_lock(lock) pthread_mutex_lock (&(lock))
+# define lock_unlock(lock) pthread_mutex_unlock (&(lock))
+#else
+# define lock_define(name)
+# define lock_init(lock) 0
+# define lock_fini(lock) ((void) 0)
+ /* The 'dfa' avoids an "unused variable 'dfa'" warning from GCC. */
+# define lock_lock(lock) ((void) dfa)
+# define lock_unlock(lock) ((void) 0)
+#endif
+
+/* In case that the system doesn't have isblank(). */
+#if !defined _LIBC && ! (defined isblank || (HAVE_ISBLANK &&
HAVE_DECL_ISBLANK))
+# define isblank(ch) ((ch) == ' ' || (ch) == '\t')
+#endif
+
+#ifdef _LIBC
+# ifndef _RE_DEFINE_LOCALE_FUNCTIONS
+# define _RE_DEFINE_LOCALE_FUNCTIONS 1
+# include <locale/localeinfo.h>
+# include <locale/coll-lookup.h>
+# endif
+#endif
+
+/* This is for other GNU distributions with internationalized messages. */
+#if (HAVE_LIBINTL_H && ENABLE_NLS) || defined _LIBC
+# include <libintl.h>
+# ifdef _LIBC
+# undef gettext
+# define gettext(msgid) \
+ __dcgettext (_libc_intl_domainname, msgid, LC_MESSAGES)
+# endif
+#else
+# undef gettext
+# define gettext(msgid) (msgid)
+#endif
+
+#ifndef gettext_noop
+/* This define is so xgettext can find the internationalizable
+ strings. */
+# define gettext_noop(String) String
+#endif
+
+#if (defined MB_CUR_MAX && HAVE_WCTYPE_H && HAVE_ISWCTYPE) || _LIBC
+# define RE_ENABLE_I18N
+#endif
+
+#define BE(expr, val) __builtin_expect (expr, val)
+
+/* Number of ASCII characters. */
+#define ASCII_CHARS 0x80
+
+/* Number of single byte characters. */
+#define SBC_MAX (UCHAR_MAX + 1)
+
+#define COLL_ELEM_LEN_MAX 8
+
+/* The character which represents newline. */
+#define NEWLINE_CHAR '\n'
+#define WIDE_NEWLINE_CHAR L'\n'
+
+/* Rename to standard API for using out of glibc. */
+#ifndef _LIBC
+# undef __wctype
+# undef __iswctype
+# define __wctype wctype
+# define __iswalnum iswalnum
+# define __iswctype iswctype
+# define __towlower towlower
+# define __towupper towupper
+# define __btowc btowc
+# define __mbrtowc mbrtowc
+# define __wcrtomb wcrtomb
+# define __regfree regfree
+# define attribute_hidden
+#endif /* not _LIBC */
+
+#if __GNUC__ < 3 + (__GNUC_MINOR__ < 1)
+# define __attribute__(arg)
+#endif
+
+#ifndef SSIZE_MAX
+# define SSIZE_MAX ((ssize_t) (SIZE_MAX / 2))
+#endif
+
+/* The type of indexes into strings. This is signed, not size_t,
+ since the API requires indexes to fit in regoff_t anyway, and using
+ signed integers makes the code a bit smaller and presumably faster.
+ The traditional GNU regex implementation uses int for indexes.
+ The POSIX-compatible implementation uses a possibly-wider type.
+ The name 'Idx' is three letters to minimize the hassle of
+ reindenting a lot of regex code that formerly used 'int'. */
+typedef regoff_t Idx;
+#ifdef _REGEX_LARGE_OFFSETS
+# define IDX_MAX SSIZE_MAX
+#else
+# define IDX_MAX INT_MAX
+#endif
+
+/* A hash value, suitable for computing hash tables. */
+typedef __re_size_t re_hashval_t;
+
+/* An integer used to represent a set of bits. It must be unsigned,
+ and must be at least as wide as unsigned int. */
+typedef unsigned long int bitset_word_t;
+/* All bits set in a bitset_word_t. */
+#define BITSET_WORD_MAX ULONG_MAX
+
+/* Number of bits in a bitset_word_t. For portability to hosts with
+ padding bits, do not use '(sizeof (bitset_word_t) * CHAR_BIT)';
+ instead, deduce it directly from BITSET_WORD_MAX. Avoid
+ greater-than-32-bit integers and unconditional shifts by more than
+ 31 bits, as they're not portable. */
+#if BITSET_WORD_MAX == 0xffffffffUL
+# define BITSET_WORD_BITS 32
+#elif BITSET_WORD_MAX >> 31 >> 4 == 1
+# define BITSET_WORD_BITS 36
+#elif BITSET_WORD_MAX >> 31 >> 16 == 1
+# define BITSET_WORD_BITS 48
+#elif BITSET_WORD_MAX >> 31 >> 28 == 1
+# define BITSET_WORD_BITS 60
+#elif BITSET_WORD_MAX >> 31 >> 31 >> 1 == 1
+# define BITSET_WORD_BITS 64
+#elif BITSET_WORD_MAX >> 31 >> 31 >> 9 == 1
+# define BITSET_WORD_BITS 72
+#elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 3 == 1
+# define BITSET_WORD_BITS 128
+#elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 7 == 1
+# define BITSET_WORD_BITS 256
+#elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 7 > 1
+# define BITSET_WORD_BITS 257 /* any value > SBC_MAX will do here */
+# if BITSET_WORD_BITS <= SBC_MAX
+# error "Invalid SBC_MAX"
+# endif
+#else
+# error "Add case for new bitset_word_t size"
+#endif
+
+/* Number of bitset_word_t values in a bitset_t. */
+#define BITSET_WORDS ((SBC_MAX + BITSET_WORD_BITS - 1) / BITSET_WORD_BITS)
+
+typedef bitset_word_t bitset_t[BITSET_WORDS];
+typedef bitset_word_t *re_bitset_ptr_t;
+typedef const bitset_word_t *re_const_bitset_ptr_t;
+
+#define PREV_WORD_CONSTRAINT 0x0001
+#define PREV_NOTWORD_CONSTRAINT 0x0002
+#define NEXT_WORD_CONSTRAINT 0x0004
+#define NEXT_NOTWORD_CONSTRAINT 0x0008
+#define PREV_NEWLINE_CONSTRAINT 0x0010
+#define NEXT_NEWLINE_CONSTRAINT 0x0020
+#define PREV_BEGBUF_CONSTRAINT 0x0040
+#define NEXT_ENDBUF_CONSTRAINT 0x0080
+#define WORD_DELIM_CONSTRAINT 0x0100
+#define NOT_WORD_DELIM_CONSTRAINT 0x0200
+
+typedef enum
+{
+ INSIDE_WORD = PREV_WORD_CONSTRAINT | NEXT_WORD_CONSTRAINT,
+ WORD_FIRST = PREV_NOTWORD_CONSTRAINT | NEXT_WORD_CONSTRAINT,
+ WORD_LAST = PREV_WORD_CONSTRAINT | NEXT_NOTWORD_CONSTRAINT,
+ INSIDE_NOTWORD = PREV_NOTWORD_CONSTRAINT | NEXT_NOTWORD_CONSTRAINT,
+ LINE_FIRST = PREV_NEWLINE_CONSTRAINT,
+ LINE_LAST = NEXT_NEWLINE_CONSTRAINT,
+ BUF_FIRST = PREV_BEGBUF_CONSTRAINT,
+ BUF_LAST = NEXT_ENDBUF_CONSTRAINT,
+ WORD_DELIM = WORD_DELIM_CONSTRAINT,
+ NOT_WORD_DELIM = NOT_WORD_DELIM_CONSTRAINT
+} re_context_type;
+
+typedef struct
+{
+ Idx alloc;
+ Idx nelem;
+ Idx *elems;
+} re_node_set;
+
+typedef enum
+{
+ NON_TYPE = 0,
+
+ /* Node type, These are used by token, node, tree. */
+ CHARACTER = 1,
+ END_OF_RE = 2,
+ SIMPLE_BRACKET = 3,
+ OP_BACK_REF = 4,
+ OP_PERIOD = 5,
+#ifdef RE_ENABLE_I18N
+ COMPLEX_BRACKET = 6,
+ OP_UTF8_PERIOD = 7,
+#endif /* RE_ENABLE_I18N */
+
+ /* We define EPSILON_BIT as a macro so that OP_OPEN_SUBEXP is used
+ when the debugger shows values of this enum type. */
+#define EPSILON_BIT 8
+ OP_OPEN_SUBEXP = EPSILON_BIT | 0,
+ OP_CLOSE_SUBEXP = EPSILON_BIT | 1,
+ OP_ALT = EPSILON_BIT | 2,
+ OP_DUP_ASTERISK = EPSILON_BIT | 3,
+ ANCHOR = EPSILON_BIT | 4,
+
+ /* Tree type, these are used only by tree. */
+ CONCAT = 16,
+ SUBEXP = 17,
+
+ /* Token type, these are used only by token. */
+ OP_DUP_PLUS = 18,
+ OP_DUP_QUESTION,
+ OP_OPEN_BRACKET,
+ OP_CLOSE_BRACKET,
+ OP_CHARSET_RANGE,
+ OP_OPEN_DUP_NUM,
+ OP_CLOSE_DUP_NUM,
+ OP_NON_MATCH_LIST,
+ OP_OPEN_COLL_ELEM,
+ OP_CLOSE_COLL_ELEM,
+ OP_OPEN_EQUIV_CLASS,
+ OP_CLOSE_EQUIV_CLASS,
+ OP_OPEN_CHAR_CLASS,
+ OP_CLOSE_CHAR_CLASS,
+ OP_WORD,
+ OP_NOTWORD,
+ OP_SPACE,
+ OP_NOTSPACE,
+ BACK_SLASH
+
+} re_token_type_t;
+
+#ifdef RE_ENABLE_I18N
+typedef struct
+{
+ /* Multibyte characters. */
+ wchar_t *mbchars;
+
+ /* Collating symbols. */
+# ifdef _LIBC
+ int32_t *coll_syms;
+# endif
+
+ /* Equivalence classes. */
+# ifdef _LIBC
+ int32_t *equiv_classes;
+# endif
+
+ /* Range expressions. */
+# ifdef _LIBC
+ uint32_t *range_starts;
+ uint32_t *range_ends;
+# else /* not _LIBC */
+ wchar_t *range_starts;
+ wchar_t *range_ends;
+# endif /* not _LIBC */
+
+ /* Character classes. */
+ wctype_t *char_classes;
+
+ /* If this character set is the non-matching list. */
+ unsigned int non_match : 1;
+
+ /* # of multibyte characters. */
+ Idx nmbchars;
+
+ /* # of collating symbols. */
+ Idx ncoll_syms;
+
+ /* # of equivalence classes. */
+ Idx nequiv_classes;
+
+ /* # of range expressions. */
+ Idx nranges;
+
+ /* # of character classes. */
+ Idx nchar_classes;
+} re_charset_t;
+#endif /* RE_ENABLE_I18N */
+
+typedef struct
+{
+ union
+ {
+ unsigned char c; /* for CHARACTER */
+ re_bitset_ptr_t sbcset; /* for SIMPLE_BRACKET */
+#ifdef RE_ENABLE_I18N
+ re_charset_t *mbcset; /* for COMPLEX_BRACKET */
+#endif /* RE_ENABLE_I18N */
+ Idx idx; /* for BACK_REF */
+ re_context_type ctx_type; /* for ANCHOR */
+ } opr;
+#if __GNUC__ >= 2 && !defined __STRICT_ANSI__
+ re_token_type_t type : 8;
+#else
+ re_token_type_t type;
+#endif
+ unsigned int constraint : 10; /* context constraint */
+ unsigned int duplicated : 1;
+ unsigned int opt_subexp : 1;
+#ifdef RE_ENABLE_I18N
+ unsigned int accept_mb : 1;
+ /* These 2 bits can be moved into the union if needed (e.g. if running out
+ of bits; move opr.c to opr.c.c and move the flags to opr.c.flags). */
+ unsigned int mb_partial : 1;
+#endif
+ unsigned int word_char : 1;
+} re_token_t;
+
+#define IS_EPSILON_NODE(type) ((type) & EPSILON_BIT)
+
+struct re_string_t
+{
+ /* Indicate the raw buffer which is the original string passed as an
+ argument of regexec(), re_search(), etc.. */
+ const unsigned char *raw_mbs;
+ /* Store the multibyte string. In case of "case insensitive mode" like
+ REG_ICASE, upper cases of the string are stored, otherwise MBS points
+ the same address that RAW_MBS points. */
+ unsigned char *mbs;
+#ifdef RE_ENABLE_I18N
+ /* Store the wide character string which is corresponding to MBS. */
+ wint_t *wcs;
+ Idx *offsets;
+ mbstate_t cur_state;
+#endif
+ /* Index in RAW_MBS. Each character mbs[i] corresponds to
+ raw_mbs[raw_mbs_idx + i]. */
+ Idx raw_mbs_idx;
+ /* The length of the valid characters in the buffers. */
+ Idx valid_len;
+ /* The corresponding number of bytes in raw_mbs array. */
+ Idx valid_raw_len;
+ /* The length of the buffers MBS and WCS. */
+ Idx bufs_len;
+ /* The index in MBS, which is updated by re_string_fetch_byte. */
+ Idx cur_idx;
+ /* length of RAW_MBS array. */
+ Idx raw_len;
+ /* This is RAW_LEN - RAW_MBS_IDX + VALID_LEN - VALID_RAW_LEN. */
+ Idx len;
+ /* End of the buffer may be shorter than its length in the cases such
+ as re_match_2, re_search_2. Then, we use STOP for end of the buffer
+ instead of LEN. */
+ Idx raw_stop;
+ /* This is RAW_STOP - RAW_MBS_IDX adjusted through OFFSETS. */
+ Idx stop;
+
+ /* The context of mbs[0]. We store the context independently, since
+ the context of mbs[0] may be different from raw_mbs[0], which is
+ the beginning of the input string. */
+ unsigned int tip_context;
+ /* The translation passed as a part of an argument of re_compile_pattern. */
+ RE_TRANSLATE_TYPE trans;
+ /* Copy of re_dfa_t's word_char. */
+ re_const_bitset_ptr_t word_char;
+ /* true if REG_ICASE. */
+ unsigned char icase;
+ unsigned char is_utf8;
+ unsigned char map_notascii;
+ unsigned char mbs_allocated;
+ unsigned char offsets_needed;
+ unsigned char newline_anchor;
+ unsigned char word_ops_used;
+ int mb_cur_max;
+};
+typedef struct re_string_t re_string_t;
+
+
+struct re_dfa_t;
+typedef struct re_dfa_t re_dfa_t;
+
+#ifndef _LIBC
+# define IS_IN(libc) false
+#endif
+
+#define re_string_peek_byte(pstr, offset) \
+ ((pstr)->mbs[(pstr)->cur_idx + offset])
+#define re_string_fetch_byte(pstr) \
+ ((pstr)->mbs[(pstr)->cur_idx++])
+#define re_string_first_byte(pstr, idx) \
+ ((idx) == (pstr)->valid_len || (pstr)->wcs[idx] != WEOF)
+#define re_string_is_single_byte_char(pstr, idx) \
+ ((pstr)->wcs[idx] != WEOF && ((pstr)->valid_len == (idx) + 1 \
+ || (pstr)->wcs[(idx) + 1] != WEOF))
+#define re_string_eoi(pstr) ((pstr)->stop <= (pstr)->cur_idx)
+#define re_string_cur_idx(pstr) ((pstr)->cur_idx)
+#define re_string_get_buffer(pstr) ((pstr)->mbs)
+#define re_string_length(pstr) ((pstr)->len)
+#define re_string_byte_at(pstr,idx) ((pstr)->mbs[idx])
+#define re_string_skip_bytes(pstr,idx) ((pstr)->cur_idx += (idx))
+#define re_string_set_index(pstr,idx) ((pstr)->cur_idx = (idx))
+
+#if defined _LIBC || HAVE_ALLOCA
+# include <alloca.h>
+#endif
+
+#ifndef _LIBC
+# if HAVE_ALLOCA
+/* The OS usually guarantees only one guard page at the bottom of the stack,
+ and a page size can be as small as 4096 bytes. So we cannot safely
+ allocate anything larger than 4096 bytes. Also care for the possibility
+ of a few compiler-allocated temporary stack slots. */
+# define __libc_use_alloca(n) ((n) < 4032)
+# else
+/* alloca is implemented with malloc, so just use malloc. */
+# define __libc_use_alloca(n) 0
+# undef alloca
+# define alloca(n) malloc (n)
+# endif
+#endif
+
+#ifdef _LIBC
+# define MALLOC_0_IS_NONNULL 1
+#elif !defined MALLOC_0_IS_NONNULL
+# define MALLOC_0_IS_NONNULL 0
+#endif
+
+#ifndef MAX
+# define MAX(a,b) ((a) < (b) ? (b) : (a))
+#endif
+#ifndef MIN
+# define MIN(a,b) ((a) < (b) ? (a) : (b))
+#endif
+
+#define re_malloc(t,n) ((t *) malloc ((n) * sizeof (t)))
+#define re_realloc(p,t,n) ((t *) realloc (p, (n) * sizeof (t)))
+#define re_free(p) free (p)
+
+struct bin_tree_t
+{
+ struct bin_tree_t *parent;
+ struct bin_tree_t *left;
+ struct bin_tree_t *right;
+ struct bin_tree_t *first;
+ struct bin_tree_t *next;
+
+ re_token_t token;
+
+ /* 'node_idx' is the index in dfa->nodes, if 'type' == 0.
+ Otherwise 'type' indicate the type of this node. */
+ Idx node_idx;
+};
+typedef struct bin_tree_t bin_tree_t;
+
+#define BIN_TREE_STORAGE_SIZE \
+ ((1024 - sizeof (void *)) / sizeof (bin_tree_t))
+
+struct bin_tree_storage_t
+{
+ struct bin_tree_storage_t *next;
+ bin_tree_t data[BIN_TREE_STORAGE_SIZE];
+};
+typedef struct bin_tree_storage_t bin_tree_storage_t;
+
+#define CONTEXT_WORD 1
+#define CONTEXT_NEWLINE (CONTEXT_WORD << 1)
+#define CONTEXT_BEGBUF (CONTEXT_NEWLINE << 1)
+#define CONTEXT_ENDBUF (CONTEXT_BEGBUF << 1)
+
+#define IS_WORD_CONTEXT(c) ((c) & CONTEXT_WORD)
+#define IS_NEWLINE_CONTEXT(c) ((c) & CONTEXT_NEWLINE)
+#define IS_BEGBUF_CONTEXT(c) ((c) & CONTEXT_BEGBUF)
+#define IS_ENDBUF_CONTEXT(c) ((c) & CONTEXT_ENDBUF)
+#define IS_ORDINARY_CONTEXT(c) ((c) == 0)
+
+#define IS_WORD_CHAR(ch) (isalnum (ch) || (ch) == '_')
+#define IS_NEWLINE(ch) ((ch) == NEWLINE_CHAR)
+#define IS_WIDE_WORD_CHAR(ch) (__iswalnum (ch) || (ch) == L'_')
+#define IS_WIDE_NEWLINE(ch) ((ch) == WIDE_NEWLINE_CHAR)
+
+#define NOT_SATISFY_PREV_CONSTRAINT(constraint,context) \
+ ((((constraint) & PREV_WORD_CONSTRAINT) && !IS_WORD_CONTEXT (context)) \
+ || ((constraint & PREV_NOTWORD_CONSTRAINT) && IS_WORD_CONTEXT (context)) \
+ || ((constraint & PREV_NEWLINE_CONSTRAINT) && !IS_NEWLINE_CONTEXT (context))\
+ || ((constraint & PREV_BEGBUF_CONSTRAINT) && !IS_BEGBUF_CONTEXT (context)))
+
+#define NOT_SATISFY_NEXT_CONSTRAINT(constraint,context) \
+ ((((constraint) & NEXT_WORD_CONSTRAINT) && !IS_WORD_CONTEXT (context)) \
+ || (((constraint) & NEXT_NOTWORD_CONSTRAINT) && IS_WORD_CONTEXT (context)) \
+ || (((constraint) & NEXT_NEWLINE_CONSTRAINT) && !IS_NEWLINE_CONTEXT
(context)) \
+ || (((constraint) & NEXT_ENDBUF_CONSTRAINT) && !IS_ENDBUF_CONTEXT (context)))
+
+struct re_dfastate_t
+{
+ re_hashval_t hash;
+ re_node_set nodes;
+ re_node_set non_eps_nodes;
+ re_node_set inveclosure;
+ re_node_set *entrance_nodes;
+ struct re_dfastate_t **trtable, **word_trtable;
+ unsigned int context : 4;
+ unsigned int halt : 1;
+ /* If this state can accept "multi byte".
+ Note that we refer to multibyte characters, and multi character
+ collating elements as "multi byte". */
+ unsigned int accept_mb : 1;
+ /* If this state has backreference node(s). */
+ unsigned int has_backref : 1;
+ unsigned int has_constraint : 1;
+};
+typedef struct re_dfastate_t re_dfastate_t;
+
+struct re_state_table_entry
+{
+ Idx num;
+ Idx alloc;
+ re_dfastate_t **array;
+};
+
+/* Array type used in re_sub_match_last_t and re_sub_match_top_t. */
+
+typedef struct
+{
+ Idx next_idx;
+ Idx alloc;
+ re_dfastate_t **array;
+} state_array_t;
+
+/* Store information about the node NODE whose type is OP_CLOSE_SUBEXP. */
+
+typedef struct
+{
+ Idx node;
+ Idx str_idx; /* The position NODE match at. */
+ state_array_t path;
+} re_sub_match_last_t;
+
+/* Store information about the node NODE whose type is OP_OPEN_SUBEXP.
+ And information about the node, whose type is OP_CLOSE_SUBEXP,
+ corresponding to NODE is stored in LASTS. */
+
+typedef struct
+{
+ Idx str_idx;
+ Idx node;
+ state_array_t *path;
+ Idx alasts; /* Allocation size of LASTS. */
+ Idx nlasts; /* The number of LASTS. */
+ re_sub_match_last_t **lasts;
+} re_sub_match_top_t;
+
+struct re_backref_cache_entry
+{
+ Idx node;
+ Idx str_idx;
+ Idx subexp_from;
+ Idx subexp_to;
+ char more;
+ char unused;
+ unsigned short int eps_reachable_subexps_map;
+};
+
+typedef struct
+{
+ /* The string object corresponding to the input string. */
+ re_string_t input;
+#if defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L)
+ const re_dfa_t *const dfa;
+#else
+ const re_dfa_t *dfa;
+#endif
+ /* EFLAGS of the argument of regexec. */
+ int eflags;
+ /* Where the matching ends. */
+ Idx match_last;
+ Idx last_node;
+ /* The state log used by the matcher. */
+ re_dfastate_t **state_log;
+ Idx state_log_top;
+ /* Back reference cache. */
+ Idx nbkref_ents;
+ Idx abkref_ents;
+ struct re_backref_cache_entry *bkref_ents;
+ int max_mb_elem_len;
+ Idx nsub_tops;
+ Idx asub_tops;
+ re_sub_match_top_t **sub_tops;
+} re_match_context_t;
+
+typedef struct
+{
+ re_dfastate_t **sifted_states;
+ re_dfastate_t **limited_states;
+ Idx last_node;
+ Idx last_str_idx;
+ re_node_set limits;
+} re_sift_context_t;
+
+struct re_fail_stack_ent_t
+{
+ Idx idx;
+ Idx node;
+ regmatch_t *regs;
+ re_node_set eps_via_nodes;
+};
+
+struct re_fail_stack_t
+{
+ Idx num;
+ Idx alloc;
+ struct re_fail_stack_ent_t *stack;
+};
+
+struct re_dfa_t
+{
+ re_token_t *nodes;
+ size_t nodes_alloc;
+ size_t nodes_len;
+ Idx *nexts;
+ Idx *org_indices;
+ re_node_set *edests;
+ re_node_set *eclosures;
+ re_node_set *inveclosures;
+ struct re_state_table_entry *state_table;
+ re_dfastate_t *init_state;
+ re_dfastate_t *init_state_word;
+ re_dfastate_t *init_state_nl;
+ re_dfastate_t *init_state_begbuf;
+ bin_tree_t *str_tree;
+ bin_tree_storage_t *str_tree_storage;
+ re_bitset_ptr_t sb_char;
+ int str_tree_storage_idx;
+
+ /* number of subexpressions 're_nsub' is in regex_t. */
+ re_hashval_t state_hash_mask;
+ Idx init_node;
+ Idx nbackref; /* The number of backreference in this dfa. */
+
+ /* Bitmap expressing which backreference is used. */
+ bitset_word_t used_bkref_map;
+ bitset_word_t completed_bkref_map;
+
+ unsigned int has_plural_match : 1;
+ /* If this dfa has "multibyte node", which is a backreference or
+ a node which can accept multibyte character or multi character
+ collating element. */
+ unsigned int has_mb_node : 1;
+ unsigned int is_utf8 : 1;
+ unsigned int map_notascii : 1;
+ unsigned int word_ops_used : 1;
+ int mb_cur_max;
+ bitset_t word_char;
+ reg_syntax_t syntax;
+ Idx *subexp_map;
+#ifdef DEBUG
+ char* re_str;
+#endif
+ lock_define (lock)
+};
+
+#define re_node_set_init_empty(set) memset (set, '\0', sizeof (re_node_set))
+#define re_node_set_remove(set,id) \
+ (re_node_set_remove_at (set, re_node_set_contains (set, id) - 1))
+#define re_node_set_empty(p) ((p)->nelem = 0)
+#define re_node_set_free(set) re_free ((set)->elems)
+
+
+typedef enum
+{
+ SB_CHAR,
+ MB_CHAR,
+ EQUIV_CLASS,
+ COLL_SYM,
+ CHAR_CLASS
+} bracket_elem_type;
+
+typedef struct
+{
+ bracket_elem_type type;
+ union
+ {
+ unsigned char ch;
+ unsigned char *name;
+ wchar_t wch;
+ } opr;
+} bracket_elem_t;
+
+
+/* Functions for bitset_t operation. */
+
+static inline void
+bitset_set (bitset_t set, Idx i)
+{
+ set[i / BITSET_WORD_BITS] |= (bitset_word_t) 1 << i % BITSET_WORD_BITS;
+}
+
+static inline void
+bitset_clear (bitset_t set, Idx i)
+{
+ set[i / BITSET_WORD_BITS] &= ~ ((bitset_word_t) 1 << i % BITSET_WORD_BITS);
+}
+
+static inline bool
+bitset_contain (const bitset_t set, Idx i)
+{
+ return (set[i / BITSET_WORD_BITS] >> i % BITSET_WORD_BITS) & 1;
+}
+
+static inline void
+bitset_empty (bitset_t set)
+{
+ memset (set, '\0', sizeof (bitset_t));
+}
+
+static inline void
+bitset_set_all (bitset_t set)
+{
+ memset (set, -1, sizeof (bitset_word_t) * (SBC_MAX / BITSET_WORD_BITS));
+ if (SBC_MAX % BITSET_WORD_BITS != 0)
+ set[BITSET_WORDS - 1] =
+ ((bitset_word_t) 1 << SBC_MAX % BITSET_WORD_BITS) - 1;
+}
+
+static inline void
+bitset_copy (bitset_t dest, const bitset_t src)
+{
+ memcpy (dest, src, sizeof (bitset_t));
+}
+
+static inline void
+bitset_not (bitset_t set)
+{
+ int bitset_i;
+ for (bitset_i = 0; bitset_i < SBC_MAX / BITSET_WORD_BITS; ++bitset_i)
+ set[bitset_i] = ~set[bitset_i];
+ if (SBC_MAX % BITSET_WORD_BITS != 0)
+ set[BITSET_WORDS - 1] =
+ ((((bitset_word_t) 1 << SBC_MAX % BITSET_WORD_BITS) - 1)
+ & ~set[BITSET_WORDS - 1]);
+}
+
+static inline void
+bitset_merge (bitset_t dest, const bitset_t src)
+{
+ int bitset_i;
+ for (bitset_i = 0; bitset_i < BITSET_WORDS; ++bitset_i)
+ dest[bitset_i] |= src[bitset_i];
+}
+
+static inline void
+bitset_mask (bitset_t dest, const bitset_t src)
+{
+ int bitset_i;
+ for (bitset_i = 0; bitset_i < BITSET_WORDS; ++bitset_i)
+ dest[bitset_i] &= src[bitset_i];
+}
+
+#ifdef RE_ENABLE_I18N
+/* Functions for re_string. */
+static int
+__attribute__ ((pure, unused))
+re_string_char_size_at (const re_string_t *pstr, Idx idx)
+{
+ int byte_idx;
+ if (pstr->mb_cur_max == 1)
+ return 1;
+ for (byte_idx = 1; idx + byte_idx < pstr->valid_len; ++byte_idx)
+ if (pstr->wcs[idx + byte_idx] != WEOF)
+ break;
+ return byte_idx;
+}
+
+static wint_t
+__attribute__ ((pure, unused))
+re_string_wchar_at (const re_string_t *pstr, Idx idx)
+{
+ if (pstr->mb_cur_max == 1)
+ return (wint_t) pstr->mbs[idx];
+ return (wint_t) pstr->wcs[idx];
+}
+
+# ifdef _LIBC
+# include <locale/weight.h>
+# endif
+
+static int
+__attribute__ ((pure, unused))
+re_string_elem_size_at (const re_string_t *pstr, Idx idx)
+{
+# ifdef _LIBC
+ const unsigned char *p, *extra;
+ const int32_t *table, *indirect;
+ uint_fast32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
+
+ if (nrules != 0)
+ {
+ table = (const int32_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB);
+ extra = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAMB);
+ indirect = (const int32_t *) _NL_CURRENT (LC_COLLATE,
+ _NL_COLLATE_INDIRECTMB);
+ p = pstr->mbs + idx;
+ findidx (table, indirect, extra, &p, pstr->len - idx);
+ return p - pstr->mbs - idx;
+ }
+ else
+# endif /* _LIBC */
+ return 1;
+}
+#endif /* RE_ENABLE_I18N */
+
+#ifndef __GNUC_PREREQ
+# if defined __GNUC__ && defined __GNUC_MINOR__
+# define __GNUC_PREREQ(maj, min) \
+ ((__GNUC__ << 16) + __GNUC_MINOR__ >= ((maj) << 16) + (min))
+# else
+# define __GNUC_PREREQ(maj, min) 0
+# endif
+#endif
+
+#if __GNUC_PREREQ (3,4)
+# undef __attribute_warn_unused_result__
+# define __attribute_warn_unused_result__ \
+ __attribute__ ((__warn_unused_result__))
+#else
+# define __attribute_warn_unused_result__ /* empty */
+#endif
+
+#ifndef FALLTHROUGH
+# if __GNUC__ < 7
+# define FALLTHROUGH ((void) 0)
+# else
+# define FALLTHROUGH __attribute__ ((__fallthrough__))
+# endif
+#endif
+
+#endif /* _REGEX_INTERNAL_H */
diff --git a/lib/regexec.c b/lib/regexec.c
new file mode 100644
index 0000000..6591311
--- /dev/null
+++ b/lib/regexec.c
@@ -0,0 +1,4324 @@
+/* Extended regular expression matching and search library.
+ Copyright (C) 2002-2018 Free Software Foundation, Inc.
+ This file is part of the GNU C Library.
+ Contributed by Isamu Hasegawa <address@hidden>.
+
+ The GNU C Library is free software; you can redistribute it and/or
+ modify it under the terms of the GNU General Public
+ License as published by the Free Software Foundation; either
+ version 3 of the License, or (at your option) any later version.
+
+ The GNU C Library is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ General Public License for more details.
+
+ You should have received a copy of the GNU General Public
+ License along with the GNU C Library; if not, see
+ <https://www.gnu.org/licenses/>. */
+
+static reg_errcode_t match_ctx_init (re_match_context_t *cache, int eflags,
+ Idx n);
+static void match_ctx_clean (re_match_context_t *mctx);
+static void match_ctx_free (re_match_context_t *cache);
+static reg_errcode_t match_ctx_add_entry (re_match_context_t *cache, Idx node,
+ Idx str_idx, Idx from, Idx to);
+static Idx search_cur_bkref_entry (const re_match_context_t *mctx, Idx
str_idx);
+static reg_errcode_t match_ctx_add_subtop (re_match_context_t *mctx, Idx node,
+ Idx str_idx);
+static re_sub_match_last_t * match_ctx_add_sublast (re_sub_match_top_t *subtop,
+ Idx node, Idx str_idx);
+static void sift_ctx_init (re_sift_context_t *sctx, re_dfastate_t **sifted_sts,
+ re_dfastate_t **limited_sts, Idx last_node,
+ Idx last_str_idx);
+static reg_errcode_t re_search_internal (const regex_t *preg,
+ const char *string, Idx length,
+ Idx start, Idx last_start, Idx stop,
+ size_t nmatch, regmatch_t pmatch[],
+ int eflags);
+static regoff_t re_search_2_stub (struct re_pattern_buffer *bufp,
+ const char *string1, Idx length1,
+ const char *string2, Idx length2,
+ Idx start, regoff_t range,
+ struct re_registers *regs,
+ Idx stop, bool ret_len);
+static regoff_t re_search_stub (struct re_pattern_buffer *bufp,
+ const char *string, Idx length, Idx start,
+ regoff_t range, Idx stop,
+ struct re_registers *regs,
+ bool ret_len);
+static unsigned re_copy_regs (struct re_registers *regs, regmatch_t *pmatch,
+ Idx nregs, int regs_allocated);
+static reg_errcode_t prune_impossible_nodes (re_match_context_t *mctx);
+static Idx check_matching (re_match_context_t *mctx, bool fl_longest_match,
+ Idx *p_match_first);
+static Idx check_halt_state_context (const re_match_context_t *mctx,
+ const re_dfastate_t *state, Idx idx);
+static void update_regs (const re_dfa_t *dfa, regmatch_t *pmatch,
+ regmatch_t *prev_idx_match, Idx cur_node,
+ Idx cur_idx, Idx nmatch);
+static reg_errcode_t push_fail_stack (struct re_fail_stack_t *fs,
+ Idx str_idx, Idx dest_node, Idx nregs,
+ regmatch_t *regs,
+ re_node_set *eps_via_nodes);
+static reg_errcode_t set_regs (const regex_t *preg,
+ const re_match_context_t *mctx,
+ size_t nmatch, regmatch_t *pmatch,
+ bool fl_backtrack);
+static reg_errcode_t free_fail_stack_return (struct re_fail_stack_t *fs);
+
+#ifdef RE_ENABLE_I18N
+static int sift_states_iter_mb (const re_match_context_t *mctx,
+ re_sift_context_t *sctx,
+ Idx node_idx, Idx str_idx, Idx max_str_idx);
+#endif /* RE_ENABLE_I18N */
+static reg_errcode_t sift_states_backward (const re_match_context_t *mctx,
+ re_sift_context_t *sctx);
+static reg_errcode_t build_sifted_states (const re_match_context_t *mctx,
+ re_sift_context_t *sctx, Idx str_idx,
+ re_node_set *cur_dest);
+static reg_errcode_t update_cur_sifted_state (const re_match_context_t *mctx,
+ re_sift_context_t *sctx,
+ Idx str_idx,
+ re_node_set *dest_nodes);
+static reg_errcode_t add_epsilon_src_nodes (const re_dfa_t *dfa,
+ re_node_set *dest_nodes,
+ const re_node_set *candidates);
+static bool check_dst_limits (const re_match_context_t *mctx,
+ const re_node_set *limits,
+ Idx dst_node, Idx dst_idx, Idx src_node,
+ Idx src_idx);
+static int check_dst_limits_calc_pos_1 (const re_match_context_t *mctx,
+ int boundaries, Idx subexp_idx,
+ Idx from_node, Idx bkref_idx);
+static int check_dst_limits_calc_pos (const re_match_context_t *mctx,
+ Idx limit, Idx subexp_idx,
+ Idx node, Idx str_idx,
+ Idx bkref_idx);
+static reg_errcode_t check_subexp_limits (const re_dfa_t *dfa,
+ re_node_set *dest_nodes,
+ const re_node_set *candidates,
+ re_node_set *limits,
+ struct re_backref_cache_entry
*bkref_ents,
+ Idx str_idx);
+static reg_errcode_t sift_states_bkref (const re_match_context_t *mctx,
+ re_sift_context_t *sctx,
+ Idx str_idx, const re_node_set
*candidates);
+static reg_errcode_t merge_state_array (const re_dfa_t *dfa,
+ re_dfastate_t **dst,
+ re_dfastate_t **src, Idx num);
+static re_dfastate_t *find_recover_state (reg_errcode_t *err,
+ re_match_context_t *mctx);
+static re_dfastate_t *transit_state (reg_errcode_t *err,
+ re_match_context_t *mctx,
+ re_dfastate_t *state);
+static re_dfastate_t *merge_state_with_log (reg_errcode_t *err,
+ re_match_context_t *mctx,
+ re_dfastate_t *next_state);
+static reg_errcode_t check_subexp_matching_top (re_match_context_t *mctx,
+ re_node_set *cur_nodes,
+ Idx str_idx);
+#if 0
+static re_dfastate_t *transit_state_sb (reg_errcode_t *err,
+ re_match_context_t *mctx,
+ re_dfastate_t *pstate);
+#endif
+#ifdef RE_ENABLE_I18N
+static reg_errcode_t transit_state_mb (re_match_context_t *mctx,
+ re_dfastate_t *pstate);
+#endif /* RE_ENABLE_I18N */
+static reg_errcode_t transit_state_bkref (re_match_context_t *mctx,
+ const re_node_set *nodes);
+static reg_errcode_t get_subexp (re_match_context_t *mctx,
+ Idx bkref_node, Idx bkref_str_idx);
+static reg_errcode_t get_subexp_sub (re_match_context_t *mctx,
+ const re_sub_match_top_t *sub_top,
+ re_sub_match_last_t *sub_last,
+ Idx bkref_node, Idx bkref_str);
+static Idx find_subexp_node (const re_dfa_t *dfa, const re_node_set *nodes,
+ Idx subexp_idx, int type);
+static reg_errcode_t check_arrival (re_match_context_t *mctx,
+ state_array_t *path, Idx top_node,
+ Idx top_str, Idx last_node, Idx last_str,
+ int type);
+static reg_errcode_t check_arrival_add_next_nodes (re_match_context_t *mctx,
+ Idx str_idx,
+ re_node_set *cur_nodes,
+ re_node_set *next_nodes);
+static reg_errcode_t check_arrival_expand_ecl (const re_dfa_t *dfa,
+ re_node_set *cur_nodes,
+ Idx ex_subexp, int type);
+static reg_errcode_t check_arrival_expand_ecl_sub (const re_dfa_t *dfa,
+ re_node_set *dst_nodes,
+ Idx target, Idx ex_subexp,
+ int type);
+static reg_errcode_t expand_bkref_cache (re_match_context_t *mctx,
+ re_node_set *cur_nodes, Idx cur_str,
+ Idx subexp_num, int type);
+static bool build_trtable (const re_dfa_t *dfa, re_dfastate_t *state);
+#ifdef RE_ENABLE_I18N
+static int check_node_accept_bytes (const re_dfa_t *dfa, Idx node_idx,
+ const re_string_t *input, Idx idx);
+# ifdef _LIBC
+static unsigned int find_collation_sequence_value (const unsigned char *mbs,
+ size_t name_len);
+# endif /* _LIBC */
+#endif /* RE_ENABLE_I18N */
+static Idx group_nodes_into_DFAstates (const re_dfa_t *dfa,
+ const re_dfastate_t *state,
+ re_node_set *states_node,
+ bitset_t *states_ch);
+static bool check_node_accept (const re_match_context_t *mctx,
+ const re_token_t *node, Idx idx);
+static reg_errcode_t extend_buffers (re_match_context_t *mctx, int min_len);
+
+/* Entry point for POSIX code. */
+
+/* regexec searches for a given pattern, specified by PREG, in the
+ string STRING.
+
+ If NMATCH is zero or REG_NOSUB was set in the cflags argument to
+ 'regcomp', we ignore PMATCH. Otherwise, we assume PMATCH has at
+ least NMATCH elements, and we set them to the offsets of the
+ corresponding matched substrings.
+
+ EFLAGS specifies "execution flags" which affect matching: if
+ REG_NOTBOL is set, then ^ does not match at the beginning of the
+ string; if REG_NOTEOL is set, then $ does not match at the end.
+
+ We return 0 if we find a match and REG_NOMATCH if not. */
+
+int
+regexec (const regex_t *_Restrict_ preg, const char *_Restrict_ string,
+ size_t nmatch, regmatch_t pmatch[], int eflags)
+{
+ reg_errcode_t err;
+ Idx start, length;
+ re_dfa_t *dfa = preg->buffer;
+
+ if (eflags & ~(REG_NOTBOL | REG_NOTEOL | REG_STARTEND))
+ return REG_BADPAT;
+
+ if (eflags & REG_STARTEND)
+ {
+ start = pmatch[0].rm_so;
+ length = pmatch[0].rm_eo;
+ }
+ else
+ {
+ start = 0;
+ length = strlen (string);
+ }
+
+ lock_lock (dfa->lock);
+ if (preg->no_sub)
+ err = re_search_internal (preg, string, length, start, length,
+ length, 0, NULL, eflags);
+ else
+ err = re_search_internal (preg, string, length, start, length,
+ length, nmatch, pmatch, eflags);
+ lock_unlock (dfa->lock);
+ return err != REG_NOERROR;
+}
+
+#ifdef _LIBC
+libc_hidden_def (__regexec)
+
+# include <shlib-compat.h>
+versioned_symbol (libc, __regexec, regexec, GLIBC_2_3_4);
+
+# if SHLIB_COMPAT (libc, GLIBC_2_0, GLIBC_2_3_4)
+__typeof__ (__regexec) __compat_regexec;
+
+int
+attribute_compat_text_section
+__compat_regexec (const regex_t *_Restrict_ preg,
+ const char *_Restrict_ string, size_t nmatch,
+ regmatch_t pmatch[], int eflags)
+{
+ return regexec (preg, string, nmatch, pmatch,
+ eflags & (REG_NOTBOL | REG_NOTEOL));
+}
+compat_symbol (libc, __compat_regexec, regexec, GLIBC_2_0);
+# endif
+#endif
+
+/* Entry points for GNU code. */
+
+/* re_match, re_search, re_match_2, re_search_2
+
+ The former two functions operate on STRING with length LENGTH,
+ while the later two operate on concatenation of STRING1 and STRING2
+ with lengths LENGTH1 and LENGTH2, respectively.
+
+ re_match() matches the compiled pattern in BUFP against the string,
+ starting at index START.
+
+ re_search() first tries matching at index START, then it tries to match
+ starting from index START + 1, and so on. The last start position tried
+ is START + RANGE. (Thus RANGE = 0 forces re_search to operate the same
+ way as re_match().)
+
+ The parameter STOP of re_{match,search}_2 specifies that no match exceeding
+ the first STOP characters of the concatenation of the strings should be
+ concerned.
+
+ If REGS is not NULL, and BUFP->no_sub is not set, the offsets of the match
+ and all groups is stored in REGS. (For the "_2" variants, the offsets are
+ computed relative to the concatenation, not relative to the individual
+ strings.)
+
+ On success, re_match* functions return the length of the match, re_search*
+ return the position of the start of the match. Return value -1 means no
+ match was found and -2 indicates an internal error. */
+
+regoff_t
+re_match (struct re_pattern_buffer *bufp, const char *string, Idx length,
+ Idx start, struct re_registers *regs)
+{
+ return re_search_stub (bufp, string, length, start, 0, length, regs, true);
+}
+#ifdef _LIBC
+weak_alias (__re_match, re_match)
+#endif
+
+regoff_t
+re_search (struct re_pattern_buffer *bufp, const char *string, Idx length,
+ Idx start, regoff_t range, struct re_registers *regs)
+{
+ return re_search_stub (bufp, string, length, start, range, length, regs,
+ false);
+}
+#ifdef _LIBC
+weak_alias (__re_search, re_search)
+#endif
+
+regoff_t
+re_match_2 (struct re_pattern_buffer *bufp, const char *string1, Idx length1,
+ const char *string2, Idx length2, Idx start,
+ struct re_registers *regs, Idx stop)
+{
+ return re_search_2_stub (bufp, string1, length1, string2, length2,
+ start, 0, regs, stop, true);
+}
+#ifdef _LIBC
+weak_alias (__re_match_2, re_match_2)
+#endif
+
+regoff_t
+re_search_2 (struct re_pattern_buffer *bufp, const char *string1, Idx length1,
+ const char *string2, Idx length2, Idx start, regoff_t range,
+ struct re_registers *regs, Idx stop)
+{
+ return re_search_2_stub (bufp, string1, length1, string2, length2,
+ start, range, regs, stop, false);
+}
+#ifdef _LIBC
+weak_alias (__re_search_2, re_search_2)
+#endif
+
+static regoff_t
+re_search_2_stub (struct re_pattern_buffer *bufp, const char *string1,
+ Idx length1, const char *string2, Idx length2, Idx start,
+ regoff_t range, struct re_registers *regs,
+ Idx stop, bool ret_len)
+{
+ const char *str;
+ regoff_t rval;
+ Idx len;
+ char *s = NULL;
+
+ if (BE ((length1 < 0 || length2 < 0 || stop < 0
+ || INT_ADD_WRAPV (length1, length2, &len)),
+ 0))
+ return -2;
+
+ /* Concatenate the strings. */
+ if (length2 > 0)
+ if (length1 > 0)
+ {
+ s = re_malloc (char, len);
+
+ if (BE (s == NULL, 0))
+ return -2;
+#ifdef _LIBC
+ memcpy (__mempcpy (s, string1, length1), string2, length2);
+#else
+ memcpy (s, string1, length1);
+ memcpy (s + length1, string2, length2);
+#endif
+ str = s;
+ }
+ else
+ str = string2;
+ else
+ str = string1;
+
+ rval = re_search_stub (bufp, str, len, start, range, stop, regs,
+ ret_len);
+ re_free (s);
+ return rval;
+}
+
+/* The parameters have the same meaning as those of re_search.
+ Additional parameters:
+ If RET_LEN is true the length of the match is returned (re_match style);
+ otherwise the position of the match is returned. */
+
+static regoff_t
+re_search_stub (struct re_pattern_buffer *bufp, const char *string, Idx length,
+ Idx start, regoff_t range, Idx stop, struct re_registers *regs,
+ bool ret_len)
+{
+ reg_errcode_t result;
+ regmatch_t *pmatch;
+ Idx nregs;
+ regoff_t rval;
+ int eflags = 0;
+ re_dfa_t *dfa = bufp->buffer;
+ Idx last_start = start + range;
+
+ /* Check for out-of-range. */
+ if (BE (start < 0 || start > length, 0))
+ return -1;
+ if (BE (length < last_start || (0 <= range && last_start < start), 0))
+ last_start = length;
+ else if (BE (last_start < 0 || (range < 0 && start <= last_start), 0))
+ last_start = 0;
+
+ lock_lock (dfa->lock);
+
+ eflags |= (bufp->not_bol) ? REG_NOTBOL : 0;
+ eflags |= (bufp->not_eol) ? REG_NOTEOL : 0;
+
+ /* Compile fastmap if we haven't yet. */
+ if (start < last_start && bufp->fastmap != NULL && !bufp->fastmap_accurate)
+ re_compile_fastmap (bufp);
+
+ if (BE (bufp->no_sub, 0))
+ regs = NULL;
+
+ /* We need at least 1 register. */
+ if (regs == NULL)
+ nregs = 1;
+ else if (BE (bufp->regs_allocated == REGS_FIXED
+ && regs->num_regs <= bufp->re_nsub, 0))
+ {
+ nregs = regs->num_regs;
+ if (BE (nregs < 1, 0))
+ {
+ /* Nothing can be copied to regs. */
+ regs = NULL;
+ nregs = 1;
+ }
+ }
+ else
+ nregs = bufp->re_nsub + 1;
+ pmatch = re_malloc (regmatch_t, nregs);
+ if (BE (pmatch == NULL, 0))
+ {
+ rval = -2;
+ goto out;
+ }
+
+ result = re_search_internal (bufp, string, length, start, last_start, stop,
+ nregs, pmatch, eflags);
+
+ rval = 0;
+
+ /* I hope we needn't fill their regs with -1's when no match was found. */
+ if (result != REG_NOERROR)
+ rval = result == REG_NOMATCH ? -1 : -2;
+ else if (regs != NULL)
+ {
+ /* If caller wants register contents data back, copy them. */
+ bufp->regs_allocated = re_copy_regs (regs, pmatch, nregs,
+ bufp->regs_allocated);
+ if (BE (bufp->regs_allocated == REGS_UNALLOCATED, 0))
+ rval = -2;
+ }
+
+ if (BE (rval == 0, 1))
+ {
+ if (ret_len)
+ {
+ assert (pmatch[0].rm_so == start);
+ rval = pmatch[0].rm_eo - start;
+ }
+ else
+ rval = pmatch[0].rm_so;
+ }
+ re_free (pmatch);
+ out:
+ lock_unlock (dfa->lock);
+ return rval;
+}
+
+static unsigned
+re_copy_regs (struct re_registers *regs, regmatch_t *pmatch, Idx nregs,
+ int regs_allocated)
+{
+ int rval = REGS_REALLOCATE;
+ Idx i;
+ Idx need_regs = nregs + 1;
+ /* We need one extra element beyond 'num_regs' for the '-1' marker GNU code
+ uses. */
+
+ /* Have the register data arrays been allocated? */
+ if (regs_allocated == REGS_UNALLOCATED)
+ { /* No. So allocate them with malloc. */
+ regs->start = re_malloc (regoff_t, need_regs);
+ if (BE (regs->start == NULL, 0))
+ return REGS_UNALLOCATED;
+ regs->end = re_malloc (regoff_t, need_regs);
+ if (BE (regs->end == NULL, 0))
+ {
+ re_free (regs->start);
+ return REGS_UNALLOCATED;
+ }
+ regs->num_regs = need_regs;
+ }
+ else if (regs_allocated == REGS_REALLOCATE)
+ { /* Yes. If we need more elements than were already
+ allocated, reallocate them. If we need fewer, just
+ leave it alone. */
+ if (BE (need_regs > regs->num_regs, 0))
+ {
+ regoff_t *new_start = re_realloc (regs->start, regoff_t, need_regs);
+ regoff_t *new_end;
+ if (BE (new_start == NULL, 0))
+ return REGS_UNALLOCATED;
+ new_end = re_realloc (regs->end, regoff_t, need_regs);
+ if (BE (new_end == NULL, 0))
+ {
+ re_free (new_start);
+ return REGS_UNALLOCATED;
+ }
+ regs->start = new_start;
+ regs->end = new_end;
+ regs->num_regs = need_regs;
+ }
+ }
+ else
+ {
+ assert (regs_allocated == REGS_FIXED);
+ /* This function may not be called with REGS_FIXED and nregs too big. */
+ assert (regs->num_regs >= nregs);
+ rval = REGS_FIXED;
+ }
+
+ /* Copy the regs. */
+ for (i = 0; i < nregs; ++i)
+ {
+ regs->start[i] = pmatch[i].rm_so;
+ regs->end[i] = pmatch[i].rm_eo;
+ }
+ for ( ; i < regs->num_regs; ++i)
+ regs->start[i] = regs->end[i] = -1;
+
+ return rval;
+}
+
+/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
+ ENDS. Subsequent matches using PATTERN_BUFFER and REGS will use
+ this memory for recording register information. STARTS and ENDS
+ must be allocated using the malloc library routine, and must each
+ be at least NUM_REGS * sizeof (regoff_t) bytes long.
+
+ If NUM_REGS == 0, then subsequent matches should allocate their own
+ register data.
+
+ Unless this function is called, the first search or match using
+ PATTERN_BUFFER will allocate its own register data, without
+ freeing the old data. */
+
+void
+re_set_registers (struct re_pattern_buffer *bufp, struct re_registers *regs,
+ __re_size_t num_regs, regoff_t *starts, regoff_t *ends)
+{
+ if (num_regs)
+ {
+ bufp->regs_allocated = REGS_REALLOCATE;
+ regs->num_regs = num_regs;
+ regs->start = starts;
+ regs->end = ends;
+ }
+ else
+ {
+ bufp->regs_allocated = REGS_UNALLOCATED;
+ regs->num_regs = 0;
+ regs->start = regs->end = NULL;
+ }
+}
+#ifdef _LIBC
+weak_alias (__re_set_registers, re_set_registers)
+#endif
+
+/* Entry points compatible with 4.2 BSD regex library. We don't define
+ them unless specifically requested. */
+
+#if defined _REGEX_RE_COMP || defined _LIBC
+int
+# ifdef _LIBC
+weak_function
+# endif
+re_exec (const char *s)
+{
+ return 0 == regexec (&re_comp_buf, s, 0, NULL, 0);
+}
+#endif /* _REGEX_RE_COMP */
+
+/* Internal entry point. */
+
+/* Searches for a compiled pattern PREG in the string STRING, whose
+ length is LENGTH. NMATCH, PMATCH, and EFLAGS have the same
+ meaning as with regexec. LAST_START is START + RANGE, where
+ START and RANGE have the same meaning as with re_search.
+ Return REG_NOERROR if we find a match, and REG_NOMATCH if not,
+ otherwise return the error code.
+ Note: We assume front end functions already check ranges.
+ (0 <= LAST_START && LAST_START <= LENGTH) */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+re_search_internal (const regex_t *preg, const char *string, Idx length,
+ Idx start, Idx last_start, Idx stop, size_t nmatch,
+ regmatch_t pmatch[], int eflags)
+{
+ reg_errcode_t err;
+ const re_dfa_t *dfa = preg->buffer;
+ Idx left_lim, right_lim;
+ int incr;
+ bool fl_longest_match;
+ int match_kind;
+ Idx match_first;
+ Idx match_last = -1;
+ Idx extra_nmatch;
+ bool sb;
+ int ch;
+#if defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L)
+ re_match_context_t mctx = { .dfa = dfa };
+#else
+ re_match_context_t mctx;
+#endif
+ char *fastmap = ((preg->fastmap != NULL && preg->fastmap_accurate
+ && start != last_start && !preg->can_be_null)
+ ? preg->fastmap : NULL);
+ RE_TRANSLATE_TYPE t = preg->translate;
+
+#if !(defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >=
199901L))
+ memset (&mctx, '\0', sizeof (re_match_context_t));
+ mctx.dfa = dfa;
+#endif
+
+ extra_nmatch = (nmatch > preg->re_nsub) ? nmatch - (preg->re_nsub + 1) : 0;
+ nmatch -= extra_nmatch;
+
+ /* Check if the DFA haven't been compiled. */
+ if (BE (preg->used == 0 || dfa->init_state == NULL
+ || dfa->init_state_word == NULL || dfa->init_state_nl == NULL
+ || dfa->init_state_begbuf == NULL, 0))
+ return REG_NOMATCH;
+
+#ifdef DEBUG
+ /* We assume front-end functions already check them. */
+ assert (0 <= last_start && last_start <= length);
+#endif
+
+ /* If initial states with non-begbuf contexts have no elements,
+ the regex must be anchored. If preg->newline_anchor is set,
+ we'll never use init_state_nl, so do not check it. */
+ if (dfa->init_state->nodes.nelem == 0
+ && dfa->init_state_word->nodes.nelem == 0
+ && (dfa->init_state_nl->nodes.nelem == 0
+ || !preg->newline_anchor))
+ {
+ if (start != 0 && last_start != 0)
+ return REG_NOMATCH;
+ start = last_start = 0;
+ }
+
+ /* We must check the longest matching, if nmatch > 0. */
+ fl_longest_match = (nmatch != 0 || dfa->nbackref);
+
+ err = re_string_allocate (&mctx.input, string, length, dfa->nodes_len + 1,
+ preg->translate, (preg->syntax & RE_ICASE) != 0,
+ dfa);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+ mctx.input.stop = stop;
+ mctx.input.raw_stop = stop;
+ mctx.input.newline_anchor = preg->newline_anchor;
+
+ err = match_ctx_init (&mctx, eflags, dfa->nbackref * 2);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+
+ /* We will log all the DFA states through which the dfa pass,
+ if nmatch > 1, or this dfa has "multibyte node", which is a
+ back-reference or a node which can accept multibyte character or
+ multi character collating element. */
+ if (nmatch > 1 || dfa->has_mb_node)
+ {
+ /* Avoid overflow. */
+ if (BE ((MIN (IDX_MAX, SIZE_MAX / sizeof (re_dfastate_t *))
+ <= mctx.input.bufs_len), 0))
+ {
+ err = REG_ESPACE;
+ goto free_return;
+ }
+
+ mctx.state_log = re_malloc (re_dfastate_t *, mctx.input.bufs_len + 1);
+ if (BE (mctx.state_log == NULL, 0))
+ {
+ err = REG_ESPACE;
+ goto free_return;
+ }
+ }
+ else
+ mctx.state_log = NULL;
+
+ match_first = start;
+ mctx.input.tip_context = (eflags & REG_NOTBOL) ? CONTEXT_BEGBUF
+ : CONTEXT_NEWLINE | CONTEXT_BEGBUF;
+
+ /* Check incrementally whether the input string matches. */
+ incr = (last_start < start) ? -1 : 1;
+ left_lim = (last_start < start) ? last_start : start;
+ right_lim = (last_start < start) ? start : last_start;
+ sb = dfa->mb_cur_max == 1;
+ match_kind =
+ (fastmap
+ ? ((sb || !(preg->syntax & RE_ICASE || t) ? 4 : 0)
+ | (start <= last_start ? 2 : 0)
+ | (t != NULL ? 1 : 0))
+ : 8);
+
+ for (;; match_first += incr)
+ {
+ err = REG_NOMATCH;
+ if (match_first < left_lim || right_lim < match_first)
+ goto free_return;
+
+ /* Advance as rapidly as possible through the string, until we
+ find a plausible place to start matching. This may be done
+ with varying efficiency, so there are various possibilities:
+ only the most common of them are specialized, in order to
+ save on code size. We use a switch statement for speed. */
+ switch (match_kind)
+ {
+ case 8:
+ /* No fastmap. */
+ break;
+
+ case 7:
+ /* Fastmap with single-byte translation, match forward. */
+ while (BE (match_first < right_lim, 1)
+ && !fastmap[t[(unsigned char) string[match_first]]])
+ ++match_first;
+ goto forward_match_found_start_or_reached_end;
+
+ case 6:
+ /* Fastmap without translation, match forward. */
+ while (BE (match_first < right_lim, 1)
+ && !fastmap[(unsigned char) string[match_first]])
+ ++match_first;
+
+ forward_match_found_start_or_reached_end:
+ if (BE (match_first == right_lim, 0))
+ {
+ ch = match_first >= length
+ ? 0 : (unsigned char) string[match_first];
+ if (!fastmap[t ? t[ch] : ch])
+ goto free_return;
+ }
+ break;
+
+ case 4:
+ case 5:
+ /* Fastmap without multi-byte translation, match backwards. */
+ while (match_first >= left_lim)
+ {
+ ch = match_first >= length
+ ? 0 : (unsigned char) string[match_first];
+ if (fastmap[t ? t[ch] : ch])
+ break;
+ --match_first;
+ }
+ if (match_first < left_lim)
+ goto free_return;
+ break;
+
+ default:
+ /* In this case, we can't determine easily the current byte,
+ since it might be a component byte of a multibyte
+ character. Then we use the constructed buffer instead. */
+ for (;;)
+ {
+ /* If MATCH_FIRST is out of the valid range, reconstruct the
+ buffers. */
+ __re_size_t offset = match_first - mctx.input.raw_mbs_idx;
+ if (BE (offset >= (__re_size_t) mctx.input.valid_raw_len, 0))
+ {
+ err = re_string_reconstruct (&mctx.input, match_first,
+ eflags);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+
+ offset = match_first - mctx.input.raw_mbs_idx;
+ }
+ /* If MATCH_FIRST is out of the buffer, leave it as '\0'.
+ Note that MATCH_FIRST must not be smaller than 0. */
+ ch = (match_first >= length
+ ? 0 : re_string_byte_at (&mctx.input, offset));
+ if (fastmap[ch])
+ break;
+ match_first += incr;
+ if (match_first < left_lim || match_first > right_lim)
+ {
+ err = REG_NOMATCH;
+ goto free_return;
+ }
+ }
+ break;
+ }
+
+ /* Reconstruct the buffers so that the matcher can assume that
+ the matching starts from the beginning of the buffer. */
+ err = re_string_reconstruct (&mctx.input, match_first, eflags);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+
+#ifdef RE_ENABLE_I18N
+ /* Don't consider this char as a possible match start if it part,
+ yet isn't the head, of a multibyte character. */
+ if (!sb && !re_string_first_byte (&mctx.input, 0))
+ continue;
+#endif
+
+ /* It seems to be appropriate one, then use the matcher. */
+ /* We assume that the matching starts from 0. */
+ mctx.state_log_top = mctx.nbkref_ents = mctx.max_mb_elem_len = 0;
+ match_last = check_matching (&mctx, fl_longest_match,
+ start <= last_start ? &match_first : NULL);
+ if (match_last != -1)
+ {
+ if (BE (match_last == -2, 0))
+ {
+ err = REG_ESPACE;
+ goto free_return;
+ }
+ else
+ {
+ mctx.match_last = match_last;
+ if ((!preg->no_sub && nmatch > 1) || dfa->nbackref)
+ {
+ re_dfastate_t *pstate = mctx.state_log[match_last];
+ mctx.last_node = check_halt_state_context (&mctx, pstate,
+ match_last);
+ }
+ if ((!preg->no_sub && nmatch > 1 && dfa->has_plural_match)
+ || dfa->nbackref)
+ {
+ err = prune_impossible_nodes (&mctx);
+ if (err == REG_NOERROR)
+ break;
+ if (BE (err != REG_NOMATCH, 0))
+ goto free_return;
+ match_last = -1;
+ }
+ else
+ break; /* We found a match. */
+ }
+ }
+
+ match_ctx_clean (&mctx);
+ }
+
+#ifdef DEBUG
+ assert (match_last != -1);
+ assert (err == REG_NOERROR);
+#endif
+
+ /* Set pmatch[] if we need. */
+ if (nmatch > 0)
+ {
+ Idx reg_idx;
+
+ /* Initialize registers. */
+ for (reg_idx = 1; reg_idx < nmatch; ++reg_idx)
+ pmatch[reg_idx].rm_so = pmatch[reg_idx].rm_eo = -1;
+
+ /* Set the points where matching start/end. */
+ pmatch[0].rm_so = 0;
+ pmatch[0].rm_eo = mctx.match_last;
+ /* FIXME: This function should fail if mctx.match_last exceeds
+ the maximum possible regoff_t value. We need a new error
+ code REG_OVERFLOW. */
+
+ if (!preg->no_sub && nmatch > 1)
+ {
+ err = set_regs (preg, &mctx, nmatch, pmatch,
+ dfa->has_plural_match && dfa->nbackref > 0);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+ }
+
+ /* At last, add the offset to each register, since we slid
+ the buffers so that we could assume that the matching starts
+ from 0. */
+ for (reg_idx = 0; reg_idx < nmatch; ++reg_idx)
+ if (pmatch[reg_idx].rm_so != -1)
+ {
+#ifdef RE_ENABLE_I18N
+ if (BE (mctx.input.offsets_needed != 0, 0))
+ {
+ pmatch[reg_idx].rm_so =
+ (pmatch[reg_idx].rm_so == mctx.input.valid_len
+ ? mctx.input.valid_raw_len
+ : mctx.input.offsets[pmatch[reg_idx].rm_so]);
+ pmatch[reg_idx].rm_eo =
+ (pmatch[reg_idx].rm_eo == mctx.input.valid_len
+ ? mctx.input.valid_raw_len
+ : mctx.input.offsets[pmatch[reg_idx].rm_eo]);
+ }
+#else
+ assert (mctx.input.offsets_needed == 0);
+#endif
+ pmatch[reg_idx].rm_so += match_first;
+ pmatch[reg_idx].rm_eo += match_first;
+ }
+ for (reg_idx = 0; reg_idx < extra_nmatch; ++reg_idx)
+ {
+ pmatch[nmatch + reg_idx].rm_so = -1;
+ pmatch[nmatch + reg_idx].rm_eo = -1;
+ }
+
+ if (dfa->subexp_map)
+ for (reg_idx = 0; reg_idx + 1 < nmatch; reg_idx++)
+ if (dfa->subexp_map[reg_idx] != reg_idx)
+ {
+ pmatch[reg_idx + 1].rm_so
+ = pmatch[dfa->subexp_map[reg_idx] + 1].rm_so;
+ pmatch[reg_idx + 1].rm_eo
+ = pmatch[dfa->subexp_map[reg_idx] + 1].rm_eo;
+ }
+ }
+
+ free_return:
+ re_free (mctx.state_log);
+ if (dfa->nbackref)
+ match_ctx_free (&mctx);
+ re_string_destruct (&mctx.input);
+ return err;
+}
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+prune_impossible_nodes (re_match_context_t *mctx)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ Idx halt_node, match_last;
+ reg_errcode_t ret;
+ re_dfastate_t **sifted_states;
+ re_dfastate_t **lim_states = NULL;
+ re_sift_context_t sctx;
+#ifdef DEBUG
+ assert (mctx->state_log != NULL);
+#endif
+ match_last = mctx->match_last;
+ halt_node = mctx->last_node;
+
+ /* Avoid overflow. */
+ if (BE (MIN (IDX_MAX, SIZE_MAX / sizeof (re_dfastate_t *)) <= match_last, 0))
+ return REG_ESPACE;
+
+ sifted_states = re_malloc (re_dfastate_t *, match_last + 1);
+ if (BE (sifted_states == NULL, 0))
+ {
+ ret = REG_ESPACE;
+ goto free_return;
+ }
+ if (dfa->nbackref)
+ {
+ lim_states = re_malloc (re_dfastate_t *, match_last + 1);
+ if (BE (lim_states == NULL, 0))
+ {
+ ret = REG_ESPACE;
+ goto free_return;
+ }
+ while (1)
+ {
+ memset (lim_states, '\0',
+ sizeof (re_dfastate_t *) * (match_last + 1));
+ sift_ctx_init (&sctx, sifted_states, lim_states, halt_node,
+ match_last);
+ ret = sift_states_backward (mctx, &sctx);
+ re_node_set_free (&sctx.limits);
+ if (BE (ret != REG_NOERROR, 0))
+ goto free_return;
+ if (sifted_states[0] != NULL || lim_states[0] != NULL)
+ break;
+ do
+ {
+ --match_last;
+ if (match_last < 0)
+ {
+ ret = REG_NOMATCH;
+ goto free_return;
+ }
+ } while (mctx->state_log[match_last] == NULL
+ || !mctx->state_log[match_last]->halt);
+ halt_node = check_halt_state_context (mctx,
+ mctx->state_log[match_last],
+ match_last);
+ }
+ ret = merge_state_array (dfa, sifted_states, lim_states,
+ match_last + 1);
+ re_free (lim_states);
+ lim_states = NULL;
+ if (BE (ret != REG_NOERROR, 0))
+ goto free_return;
+ }
+ else
+ {
+ sift_ctx_init (&sctx, sifted_states, lim_states, halt_node, match_last);
+ ret = sift_states_backward (mctx, &sctx);
+ re_node_set_free (&sctx.limits);
+ if (BE (ret != REG_NOERROR, 0))
+ goto free_return;
+ if (sifted_states[0] == NULL)
+ {
+ ret = REG_NOMATCH;
+ goto free_return;
+ }
+ }
+ re_free (mctx->state_log);
+ mctx->state_log = sifted_states;
+ sifted_states = NULL;
+ mctx->last_node = halt_node;
+ mctx->match_last = match_last;
+ ret = REG_NOERROR;
+ free_return:
+ re_free (sifted_states);
+ re_free (lim_states);
+ return ret;
+}
+
+/* Acquire an initial state and return it.
+ We must select appropriate initial state depending on the context,
+ since initial states may have constraints like "\<", "^", etc.. */
+
+static inline re_dfastate_t *
+__attribute__ ((always_inline))
+acquire_init_state_context (reg_errcode_t *err, const re_match_context_t *mctx,
+ Idx idx)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ if (dfa->init_state->has_constraint)
+ {
+ unsigned int context;
+ context = re_string_context_at (&mctx->input, idx - 1, mctx->eflags);
+ if (IS_WORD_CONTEXT (context))
+ return dfa->init_state_word;
+ else if (IS_ORDINARY_CONTEXT (context))
+ return dfa->init_state;
+ else if (IS_BEGBUF_CONTEXT (context) && IS_NEWLINE_CONTEXT (context))
+ return dfa->init_state_begbuf;
+ else if (IS_NEWLINE_CONTEXT (context))
+ return dfa->init_state_nl;
+ else if (IS_BEGBUF_CONTEXT (context))
+ {
+ /* It is relatively rare case, then calculate on demand. */
+ return re_acquire_state_context (err, dfa,
+ dfa->init_state->entrance_nodes,
+ context);
+ }
+ else
+ /* Must not happen? */
+ return dfa->init_state;
+ }
+ else
+ return dfa->init_state;
+}
+
+/* Check whether the regular expression match input string INPUT or not,
+ and return the index where the matching end. Return -1 if
+ there is no match, and return -2 in case of an error.
+ FL_LONGEST_MATCH means we want the POSIX longest matching.
+ If P_MATCH_FIRST is not NULL, and the match fails, it is set to the
+ next place where we may want to try matching.
+ Note that the matcher assumes that the matching starts from the current
+ index of the buffer. */
+
+static Idx
+__attribute_warn_unused_result__
+check_matching (re_match_context_t *mctx, bool fl_longest_match,
+ Idx *p_match_first)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ reg_errcode_t err;
+ Idx match = 0;
+ Idx match_last = -1;
+ Idx cur_str_idx = re_string_cur_idx (&mctx->input);
+ re_dfastate_t *cur_state;
+ bool at_init_state = p_match_first != NULL;
+ Idx next_start_idx = cur_str_idx;
+
+ err = REG_NOERROR;
+ cur_state = acquire_init_state_context (&err, mctx, cur_str_idx);
+ /* An initial state must not be NULL (invalid). */
+ if (BE (cur_state == NULL, 0))
+ {
+ assert (err == REG_ESPACE);
+ return -2;
+ }
+
+ if (mctx->state_log != NULL)
+ {
+ mctx->state_log[cur_str_idx] = cur_state;
+
+ /* Check OP_OPEN_SUBEXP in the initial state in case that we use them
+ later. E.g. Processing back references. */
+ if (BE (dfa->nbackref, 0))
+ {
+ at_init_state = false;
+ err = check_subexp_matching_top (mctx, &cur_state->nodes, 0);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+
+ if (cur_state->has_backref)
+ {
+ err = transit_state_bkref (mctx, &cur_state->nodes);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ }
+ }
+
+ /* If the RE accepts NULL string. */
+ if (BE (cur_state->halt, 0))
+ {
+ if (!cur_state->has_constraint
+ || check_halt_state_context (mctx, cur_state, cur_str_idx))
+ {
+ if (!fl_longest_match)
+ return cur_str_idx;
+ else
+ {
+ match_last = cur_str_idx;
+ match = 1;
+ }
+ }
+ }
+
+ while (!re_string_eoi (&mctx->input))
+ {
+ re_dfastate_t *old_state = cur_state;
+ Idx next_char_idx = re_string_cur_idx (&mctx->input) + 1;
+
+ if ((BE (next_char_idx >= mctx->input.bufs_len, 0)
+ && mctx->input.bufs_len < mctx->input.len)
+ || (BE (next_char_idx >= mctx->input.valid_len, 0)
+ && mctx->input.valid_len < mctx->input.len))
+ {
+ err = extend_buffers (mctx, next_char_idx + 1);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ assert (err == REG_ESPACE);
+ return -2;
+ }
+ }
+
+ cur_state = transit_state (&err, mctx, cur_state);
+ if (mctx->state_log != NULL)
+ cur_state = merge_state_with_log (&err, mctx, cur_state);
+
+ if (cur_state == NULL)
+ {
+ /* Reached the invalid state or an error. Try to recover a valid
+ state using the state log, if available and if we have not
+ already found a valid (even if not the longest) match. */
+ if (BE (err != REG_NOERROR, 0))
+ return -2;
+
+ if (mctx->state_log == NULL
+ || (match && !fl_longest_match)
+ || (cur_state = find_recover_state (&err, mctx)) == NULL)
+ break;
+ }
+
+ if (BE (at_init_state, 0))
+ {
+ if (old_state == cur_state)
+ next_start_idx = next_char_idx;
+ else
+ at_init_state = false;
+ }
+
+ if (cur_state->halt)
+ {
+ /* Reached a halt state.
+ Check the halt state can satisfy the current context. */
+ if (!cur_state->has_constraint
+ || check_halt_state_context (mctx, cur_state,
+ re_string_cur_idx (&mctx->input)))
+ {
+ /* We found an appropriate halt state. */
+ match_last = re_string_cur_idx (&mctx->input);
+ match = 1;
+
+ /* We found a match, do not modify match_first below. */
+ p_match_first = NULL;
+ if (!fl_longest_match)
+ break;
+ }
+ }
+ }
+
+ if (p_match_first)
+ *p_match_first += next_start_idx;
+
+ return match_last;
+}
+
+/* Check NODE match the current context. */
+
+static bool
+check_halt_node_context (const re_dfa_t *dfa, Idx node, unsigned int context)
+{
+ re_token_type_t type = dfa->nodes[node].type;
+ unsigned int constraint = dfa->nodes[node].constraint;
+ if (type != END_OF_RE)
+ return false;
+ if (!constraint)
+ return true;
+ if (NOT_SATISFY_NEXT_CONSTRAINT (constraint, context))
+ return false;
+ return true;
+}
+
+/* Check the halt state STATE match the current context.
+ Return 0 if not match, if the node, STATE has, is a halt node and
+ match the context, return the node. */
+
+static Idx
+check_halt_state_context (const re_match_context_t *mctx,
+ const re_dfastate_t *state, Idx idx)
+{
+ Idx i;
+ unsigned int context;
+#ifdef DEBUG
+ assert (state->halt);
+#endif
+ context = re_string_context_at (&mctx->input, idx, mctx->eflags);
+ for (i = 0; i < state->nodes.nelem; ++i)
+ if (check_halt_node_context (mctx->dfa, state->nodes.elems[i], context))
+ return state->nodes.elems[i];
+ return 0;
+}
+
+/* Compute the next node to which "NFA" transit from NODE("NFA" is a NFA
+ corresponding to the DFA).
+ Return the destination node, and update EPS_VIA_NODES;
+ return -1 in case of errors. */
+
+static Idx
+proceed_next_node (const re_match_context_t *mctx, Idx nregs, regmatch_t *regs,
+ Idx *pidx, Idx node, re_node_set *eps_via_nodes,
+ struct re_fail_stack_t *fs)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ Idx i;
+ bool ok;
+ if (IS_EPSILON_NODE (dfa->nodes[node].type))
+ {
+ re_node_set *cur_nodes = &mctx->state_log[*pidx]->nodes;
+ re_node_set *edests = &dfa->edests[node];
+ Idx dest_node;
+ ok = re_node_set_insert (eps_via_nodes, node);
+ if (BE (! ok, 0))
+ return -2;
+ /* Pick up a valid destination, or return -1 if none
+ is found. */
+ for (dest_node = -1, i = 0; i < edests->nelem; ++i)
+ {
+ Idx candidate = edests->elems[i];
+ if (!re_node_set_contains (cur_nodes, candidate))
+ continue;
+ if (dest_node == -1)
+ dest_node = candidate;
+
+ else
+ {
+ /* In order to avoid infinite loop like "(a*)*", return the second
+ epsilon-transition if the first was already considered. */
+ if (re_node_set_contains (eps_via_nodes, dest_node))
+ return candidate;
+
+ /* Otherwise, push the second epsilon-transition on the fail
stack. */
+ else if (fs != NULL
+ && push_fail_stack (fs, *pidx, candidate, nregs, regs,
+ eps_via_nodes))
+ return -2;
+
+ /* We know we are going to exit. */
+ break;
+ }
+ }
+ return dest_node;
+ }
+ else
+ {
+ Idx naccepted = 0;
+ re_token_type_t type = dfa->nodes[node].type;
+
+#ifdef RE_ENABLE_I18N
+ if (dfa->nodes[node].accept_mb)
+ naccepted = check_node_accept_bytes (dfa, node, &mctx->input, *pidx);
+ else
+#endif /* RE_ENABLE_I18N */
+ if (type == OP_BACK_REF)
+ {
+ Idx subexp_idx = dfa->nodes[node].opr.idx + 1;
+ naccepted = regs[subexp_idx].rm_eo - regs[subexp_idx].rm_so;
+ if (fs != NULL)
+ {
+ if (regs[subexp_idx].rm_so == -1 || regs[subexp_idx].rm_eo == -1)
+ return -1;
+ else if (naccepted)
+ {
+ char *buf = (char *) re_string_get_buffer (&mctx->input);
+ if (memcmp (buf + regs[subexp_idx].rm_so, buf + *pidx,
+ naccepted) != 0)
+ return -1;
+ }
+ }
+
+ if (naccepted == 0)
+ {
+ Idx dest_node;
+ ok = re_node_set_insert (eps_via_nodes, node);
+ if (BE (! ok, 0))
+ return -2;
+ dest_node = dfa->edests[node].elems[0];
+ if (re_node_set_contains (&mctx->state_log[*pidx]->nodes,
+ dest_node))
+ return dest_node;
+ }
+ }
+
+ if (naccepted != 0
+ || check_node_accept (mctx, dfa->nodes + node, *pidx))
+ {
+ Idx dest_node = dfa->nexts[node];
+ *pidx = (naccepted == 0) ? *pidx + 1 : *pidx + naccepted;
+ if (fs && (*pidx > mctx->match_last || mctx->state_log[*pidx] == NULL
+ || !re_node_set_contains (&mctx->state_log[*pidx]->nodes,
+ dest_node)))
+ return -1;
+ re_node_set_empty (eps_via_nodes);
+ return dest_node;
+ }
+ }
+ return -1;
+}
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+push_fail_stack (struct re_fail_stack_t *fs, Idx str_idx, Idx dest_node,
+ Idx nregs, regmatch_t *regs, re_node_set *eps_via_nodes)
+{
+ reg_errcode_t err;
+ Idx num = fs->num++;
+ if (fs->num == fs->alloc)
+ {
+ struct re_fail_stack_ent_t *new_array;
+ new_array = re_realloc (fs->stack, struct re_fail_stack_ent_t,
+ fs->alloc * 2);
+ if (new_array == NULL)
+ return REG_ESPACE;
+ fs->alloc *= 2;
+ fs->stack = new_array;
+ }
+ fs->stack[num].idx = str_idx;
+ fs->stack[num].node = dest_node;
+ fs->stack[num].regs = re_malloc (regmatch_t, nregs);
+ if (fs->stack[num].regs == NULL)
+ return REG_ESPACE;
+ memcpy (fs->stack[num].regs, regs, sizeof (regmatch_t) * nregs);
+ err = re_node_set_init_copy (&fs->stack[num].eps_via_nodes, eps_via_nodes);
+ return err;
+}
+
+static Idx
+pop_fail_stack (struct re_fail_stack_t *fs, Idx *pidx, Idx nregs,
+ regmatch_t *regs, re_node_set *eps_via_nodes)
+{
+ Idx num = --fs->num;
+ assert (num >= 0);
+ *pidx = fs->stack[num].idx;
+ memcpy (regs, fs->stack[num].regs, sizeof (regmatch_t) * nregs);
+ re_node_set_free (eps_via_nodes);
+ re_free (fs->stack[num].regs);
+ *eps_via_nodes = fs->stack[num].eps_via_nodes;
+ return fs->stack[num].node;
+}
+
+/* Set the positions where the subexpressions are starts/ends to registers
+ PMATCH.
+ Note: We assume that pmatch[0] is already set, and
+ pmatch[i].rm_so == pmatch[i].rm_eo == -1 for 0 < i < nmatch. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+set_regs (const regex_t *preg, const re_match_context_t *mctx, size_t nmatch,
+ regmatch_t *pmatch, bool fl_backtrack)
+{
+ const re_dfa_t *dfa = preg->buffer;
+ Idx idx, cur_node;
+ re_node_set eps_via_nodes;
+ struct re_fail_stack_t *fs;
+ struct re_fail_stack_t fs_body = { 0, 2, NULL };
+ regmatch_t *prev_idx_match;
+ bool prev_idx_match_malloced = false;
+
+#ifdef DEBUG
+ assert (nmatch > 1);
+ assert (mctx->state_log != NULL);
+#endif
+ if (fl_backtrack)
+ {
+ fs = &fs_body;
+ fs->stack = re_malloc (struct re_fail_stack_ent_t, fs->alloc);
+ if (fs->stack == NULL)
+ return REG_ESPACE;
+ }
+ else
+ fs = NULL;
+
+ cur_node = dfa->init_node;
+ re_node_set_init_empty (&eps_via_nodes);
+
+ if (__libc_use_alloca (nmatch * sizeof (regmatch_t)))
+ prev_idx_match = (regmatch_t *) alloca (nmatch * sizeof (regmatch_t));
+ else
+ {
+ prev_idx_match = re_malloc (regmatch_t, nmatch);
+ if (prev_idx_match == NULL)
+ {
+ free_fail_stack_return (fs);
+ return REG_ESPACE;
+ }
+ prev_idx_match_malloced = true;
+ }
+ memcpy (prev_idx_match, pmatch, sizeof (regmatch_t) * nmatch);
+
+ for (idx = pmatch[0].rm_so; idx <= pmatch[0].rm_eo ;)
+ {
+ update_regs (dfa, pmatch, prev_idx_match, cur_node, idx, nmatch);
+
+ if (idx == pmatch[0].rm_eo && cur_node == mctx->last_node)
+ {
+ Idx reg_idx;
+ if (fs)
+ {
+ for (reg_idx = 0; reg_idx < nmatch; ++reg_idx)
+ if (pmatch[reg_idx].rm_so > -1 && pmatch[reg_idx].rm_eo == -1)
+ break;
+ if (reg_idx == nmatch)
+ {
+ re_node_set_free (&eps_via_nodes);
+ if (prev_idx_match_malloced)
+ re_free (prev_idx_match);
+ return free_fail_stack_return (fs);
+ }
+ cur_node = pop_fail_stack (fs, &idx, nmatch, pmatch,
+ &eps_via_nodes);
+ }
+ else
+ {
+ re_node_set_free (&eps_via_nodes);
+ if (prev_idx_match_malloced)
+ re_free (prev_idx_match);
+ return REG_NOERROR;
+ }
+ }
+
+ /* Proceed to next node. */
+ cur_node = proceed_next_node (mctx, nmatch, pmatch, &idx, cur_node,
+ &eps_via_nodes, fs);
+
+ if (BE (cur_node < 0, 0))
+ {
+ if (BE (cur_node == -2, 0))
+ {
+ re_node_set_free (&eps_via_nodes);
+ if (prev_idx_match_malloced)
+ re_free (prev_idx_match);
+ free_fail_stack_return (fs);
+ return REG_ESPACE;
+ }
+ if (fs)
+ cur_node = pop_fail_stack (fs, &idx, nmatch, pmatch,
+ &eps_via_nodes);
+ else
+ {
+ re_node_set_free (&eps_via_nodes);
+ if (prev_idx_match_malloced)
+ re_free (prev_idx_match);
+ return REG_NOMATCH;
+ }
+ }
+ }
+ re_node_set_free (&eps_via_nodes);
+ if (prev_idx_match_malloced)
+ re_free (prev_idx_match);
+ return free_fail_stack_return (fs);
+}
+
+static reg_errcode_t
+free_fail_stack_return (struct re_fail_stack_t *fs)
+{
+ if (fs)
+ {
+ Idx fs_idx;
+ for (fs_idx = 0; fs_idx < fs->num; ++fs_idx)
+ {
+ re_node_set_free (&fs->stack[fs_idx].eps_via_nodes);
+ re_free (fs->stack[fs_idx].regs);
+ }
+ re_free (fs->stack);
+ }
+ return REG_NOERROR;
+}
+
+static void
+update_regs (const re_dfa_t *dfa, regmatch_t *pmatch,
+ regmatch_t *prev_idx_match, Idx cur_node, Idx cur_idx, Idx nmatch)
+{
+ int type = dfa->nodes[cur_node].type;
+ if (type == OP_OPEN_SUBEXP)
+ {
+ Idx reg_num = dfa->nodes[cur_node].opr.idx + 1;
+
+ /* We are at the first node of this sub expression. */
+ if (reg_num < nmatch)
+ {
+ pmatch[reg_num].rm_so = cur_idx;
+ pmatch[reg_num].rm_eo = -1;
+ }
+ }
+ else if (type == OP_CLOSE_SUBEXP)
+ {
+ Idx reg_num = dfa->nodes[cur_node].opr.idx + 1;
+ if (reg_num < nmatch)
+ {
+ /* We are at the last node of this sub expression. */
+ if (pmatch[reg_num].rm_so < cur_idx)
+ {
+ pmatch[reg_num].rm_eo = cur_idx;
+ /* This is a non-empty match or we are not inside an optional
+ subexpression. Accept this right away. */
+ memcpy (prev_idx_match, pmatch, sizeof (regmatch_t) * nmatch);
+ }
+ else
+ {
+ if (dfa->nodes[cur_node].opt_subexp
+ && prev_idx_match[reg_num].rm_so != -1)
+ /* We transited through an empty match for an optional
+ subexpression, like (a?)*, and this is not the subexp's
+ first match. Copy back the old content of the registers
+ so that matches of an inner subexpression are undone as
+ well, like in ((a?))*. */
+ memcpy (pmatch, prev_idx_match, sizeof (regmatch_t) * nmatch);
+ else
+ /* We completed a subexpression, but it may be part of
+ an optional one, so do not update PREV_IDX_MATCH. */
+ pmatch[reg_num].rm_eo = cur_idx;
+ }
+ }
+ }
+}
+
+/* This function checks the STATE_LOG from the SCTX->last_str_idx to 0
+ and sift the nodes in each states according to the following rules.
+ Updated state_log will be wrote to STATE_LOG.
+
+ Rules: We throw away the Node 'a' in the STATE_LOG[STR_IDX] if...
+ 1. When STR_IDX == MATCH_LAST(the last index in the state_log):
+ If 'a' isn't the LAST_NODE and 'a' can't epsilon transit to
+ the LAST_NODE, we throw away the node 'a'.
+ 2. When 0 <= STR_IDX < MATCH_LAST and 'a' accepts
+ string 's' and transit to 'b':
+ i. If 'b' isn't in the STATE_LOG[STR_IDX+strlen('s')], we throw
+ away the node 'a'.
+ ii. If 'b' is in the STATE_LOG[STR_IDX+strlen('s')] but 'b' is
+ thrown away, we throw away the node 'a'.
+ 3. When 0 <= STR_IDX < MATCH_LAST and 'a' epsilon transit to 'b':
+ i. If 'b' isn't in the STATE_LOG[STR_IDX], we throw away the
+ node 'a'.
+ ii. If 'b' is in the STATE_LOG[STR_IDX] but 'b' is thrown away,
+ we throw away the node 'a'. */
+
+#define STATE_NODE_CONTAINS(state,node) \
+ ((state) != NULL && re_node_set_contains (&(state)->nodes, node))
+
+static reg_errcode_t
+sift_states_backward (const re_match_context_t *mctx, re_sift_context_t *sctx)
+{
+ reg_errcode_t err;
+ int null_cnt = 0;
+ Idx str_idx = sctx->last_str_idx;
+ re_node_set cur_dest;
+
+#ifdef DEBUG
+ assert (mctx->state_log != NULL && mctx->state_log[str_idx] != NULL);
+#endif
+
+ /* Build sifted state_log[str_idx]. It has the nodes which can epsilon
+ transit to the last_node and the last_node itself. */
+ err = re_node_set_init_1 (&cur_dest, sctx->last_node);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ err = update_cur_sifted_state (mctx, sctx, str_idx, &cur_dest);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+
+ /* Then check each states in the state_log. */
+ while (str_idx > 0)
+ {
+ /* Update counters. */
+ null_cnt = (sctx->sifted_states[str_idx] == NULL) ? null_cnt + 1 : 0;
+ if (null_cnt > mctx->max_mb_elem_len)
+ {
+ memset (sctx->sifted_states, '\0',
+ sizeof (re_dfastate_t *) * str_idx);
+ re_node_set_free (&cur_dest);
+ return REG_NOERROR;
+ }
+ re_node_set_empty (&cur_dest);
+ --str_idx;
+
+ if (mctx->state_log[str_idx])
+ {
+ err = build_sifted_states (mctx, sctx, str_idx, &cur_dest);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+ }
+
+ /* Add all the nodes which satisfy the following conditions:
+ - It can epsilon transit to a node in CUR_DEST.
+ - It is in CUR_SRC.
+ And update state_log. */
+ err = update_cur_sifted_state (mctx, sctx, str_idx, &cur_dest);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+ }
+ err = REG_NOERROR;
+ free_return:
+ re_node_set_free (&cur_dest);
+ return err;
+}
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+build_sifted_states (const re_match_context_t *mctx, re_sift_context_t *sctx,
+ Idx str_idx, re_node_set *cur_dest)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ const re_node_set *cur_src = &mctx->state_log[str_idx]->non_eps_nodes;
+ Idx i;
+
+ /* Then build the next sifted state.
+ We build the next sifted state on 'cur_dest', and update
+ 'sifted_states[str_idx]' with 'cur_dest'.
+ Note:
+ 'cur_dest' is the sifted state from 'state_log[str_idx + 1]'.
+ 'cur_src' points the node_set of the old 'state_log[str_idx]'
+ (with the epsilon nodes pre-filtered out). */
+ for (i = 0; i < cur_src->nelem; i++)
+ {
+ Idx prev_node = cur_src->elems[i];
+ int naccepted = 0;
+ bool ok;
+
+#ifdef DEBUG
+ re_token_type_t type = dfa->nodes[prev_node].type;
+ assert (!IS_EPSILON_NODE (type));
+#endif
+#ifdef RE_ENABLE_I18N
+ /* If the node may accept "multi byte". */
+ if (dfa->nodes[prev_node].accept_mb)
+ naccepted = sift_states_iter_mb (mctx, sctx, prev_node,
+ str_idx, sctx->last_str_idx);
+#endif /* RE_ENABLE_I18N */
+
+ /* We don't check backreferences here.
+ See update_cur_sifted_state(). */
+ if (!naccepted
+ && check_node_accept (mctx, dfa->nodes + prev_node, str_idx)
+ && STATE_NODE_CONTAINS (sctx->sifted_states[str_idx + 1],
+ dfa->nexts[prev_node]))
+ naccepted = 1;
+
+ if (naccepted == 0)
+ continue;
+
+ if (sctx->limits.nelem)
+ {
+ Idx to_idx = str_idx + naccepted;
+ if (check_dst_limits (mctx, &sctx->limits,
+ dfa->nexts[prev_node], to_idx,
+ prev_node, str_idx))
+ continue;
+ }
+ ok = re_node_set_insert (cur_dest, prev_node);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ }
+
+ return REG_NOERROR;
+}
+
+/* Helper functions. */
+
+static reg_errcode_t
+clean_state_log_if_needed (re_match_context_t *mctx, Idx next_state_log_idx)
+{
+ Idx top = mctx->state_log_top;
+
+ if ((next_state_log_idx >= mctx->input.bufs_len
+ && mctx->input.bufs_len < mctx->input.len)
+ || (next_state_log_idx >= mctx->input.valid_len
+ && mctx->input.valid_len < mctx->input.len))
+ {
+ reg_errcode_t err;
+ err = extend_buffers (mctx, next_state_log_idx + 1);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+
+ if (top < next_state_log_idx)
+ {
+ memset (mctx->state_log + top + 1, '\0',
+ sizeof (re_dfastate_t *) * (next_state_log_idx - top));
+ mctx->state_log_top = next_state_log_idx;
+ }
+ return REG_NOERROR;
+}
+
+static reg_errcode_t
+merge_state_array (const re_dfa_t *dfa, re_dfastate_t **dst,
+ re_dfastate_t **src, Idx num)
+{
+ Idx st_idx;
+ reg_errcode_t err;
+ for (st_idx = 0; st_idx < num; ++st_idx)
+ {
+ if (dst[st_idx] == NULL)
+ dst[st_idx] = src[st_idx];
+ else if (src[st_idx] != NULL)
+ {
+ re_node_set merged_set;
+ err = re_node_set_init_union (&merged_set, &dst[st_idx]->nodes,
+ &src[st_idx]->nodes);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ dst[st_idx] = re_acquire_state (&err, dfa, &merged_set);
+ re_node_set_free (&merged_set);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ }
+ return REG_NOERROR;
+}
+
+static reg_errcode_t
+update_cur_sifted_state (const re_match_context_t *mctx,
+ re_sift_context_t *sctx, Idx str_idx,
+ re_node_set *dest_nodes)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ reg_errcode_t err = REG_NOERROR;
+ const re_node_set *candidates;
+ candidates = ((mctx->state_log[str_idx] == NULL) ? NULL
+ : &mctx->state_log[str_idx]->nodes);
+
+ if (dest_nodes->nelem == 0)
+ sctx->sifted_states[str_idx] = NULL;
+ else
+ {
+ if (candidates)
+ {
+ /* At first, add the nodes which can epsilon transit to a node in
+ DEST_NODE. */
+ err = add_epsilon_src_nodes (dfa, dest_nodes, candidates);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+
+ /* Then, check the limitations in the current sift_context. */
+ if (sctx->limits.nelem)
+ {
+ err = check_subexp_limits (dfa, dest_nodes, candidates,
&sctx->limits,
+ mctx->bkref_ents, str_idx);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ }
+
+ sctx->sifted_states[str_idx] = re_acquire_state (&err, dfa, dest_nodes);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+
+ if (candidates && mctx->state_log[str_idx]->has_backref)
+ {
+ err = sift_states_bkref (mctx, sctx, str_idx, candidates);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ return REG_NOERROR;
+}
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+add_epsilon_src_nodes (const re_dfa_t *dfa, re_node_set *dest_nodes,
+ const re_node_set *candidates)
+{
+ reg_errcode_t err = REG_NOERROR;
+ Idx i;
+
+ re_dfastate_t *state = re_acquire_state (&err, dfa, dest_nodes);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+
+ if (!state->inveclosure.alloc)
+ {
+ err = re_node_set_alloc (&state->inveclosure, dest_nodes->nelem);
+ if (BE (err != REG_NOERROR, 0))
+ return REG_ESPACE;
+ for (i = 0; i < dest_nodes->nelem; i++)
+ {
+ err = re_node_set_merge (&state->inveclosure,
+ dfa->inveclosures + dest_nodes->elems[i]);
+ if (BE (err != REG_NOERROR, 0))
+ return REG_ESPACE;
+ }
+ }
+ return re_node_set_add_intersect (dest_nodes, candidates,
+ &state->inveclosure);
+}
+
+static reg_errcode_t
+sub_epsilon_src_nodes (const re_dfa_t *dfa, Idx node, re_node_set *dest_nodes,
+ const re_node_set *candidates)
+{
+ Idx ecl_idx;
+ reg_errcode_t err;
+ re_node_set *inv_eclosure = dfa->inveclosures + node;
+ re_node_set except_nodes;
+ re_node_set_init_empty (&except_nodes);
+ for (ecl_idx = 0; ecl_idx < inv_eclosure->nelem; ++ecl_idx)
+ {
+ Idx cur_node = inv_eclosure->elems[ecl_idx];
+ if (cur_node == node)
+ continue;
+ if (IS_EPSILON_NODE (dfa->nodes[cur_node].type))
+ {
+ Idx edst1 = dfa->edests[cur_node].elems[0];
+ Idx edst2 = ((dfa->edests[cur_node].nelem > 1)
+ ? dfa->edests[cur_node].elems[1] : -1);
+ if ((!re_node_set_contains (inv_eclosure, edst1)
+ && re_node_set_contains (dest_nodes, edst1))
+ || (edst2 > 0
+ && !re_node_set_contains (inv_eclosure, edst2)
+ && re_node_set_contains (dest_nodes, edst2)))
+ {
+ err = re_node_set_add_intersect (&except_nodes, candidates,
+ dfa->inveclosures + cur_node);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&except_nodes);
+ return err;
+ }
+ }
+ }
+ }
+ for (ecl_idx = 0; ecl_idx < inv_eclosure->nelem; ++ecl_idx)
+ {
+ Idx cur_node = inv_eclosure->elems[ecl_idx];
+ if (!re_node_set_contains (&except_nodes, cur_node))
+ {
+ Idx idx = re_node_set_contains (dest_nodes, cur_node) - 1;
+ re_node_set_remove_at (dest_nodes, idx);
+ }
+ }
+ re_node_set_free (&except_nodes);
+ return REG_NOERROR;
+}
+
+static bool
+check_dst_limits (const re_match_context_t *mctx, const re_node_set *limits,
+ Idx dst_node, Idx dst_idx, Idx src_node, Idx src_idx)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ Idx lim_idx, src_pos, dst_pos;
+
+ Idx dst_bkref_idx = search_cur_bkref_entry (mctx, dst_idx);
+ Idx src_bkref_idx = search_cur_bkref_entry (mctx, src_idx);
+ for (lim_idx = 0; lim_idx < limits->nelem; ++lim_idx)
+ {
+ Idx subexp_idx;
+ struct re_backref_cache_entry *ent;
+ ent = mctx->bkref_ents + limits->elems[lim_idx];
+ subexp_idx = dfa->nodes[ent->node].opr.idx;
+
+ dst_pos = check_dst_limits_calc_pos (mctx, limits->elems[lim_idx],
+ subexp_idx, dst_node, dst_idx,
+ dst_bkref_idx);
+ src_pos = check_dst_limits_calc_pos (mctx, limits->elems[lim_idx],
+ subexp_idx, src_node, src_idx,
+ src_bkref_idx);
+
+ /* In case of:
+ <src> <dst> ( <subexp> )
+ ( <subexp> ) <src> <dst>
+ ( <subexp1> <src> <subexp2> <dst> <subexp3> ) */
+ if (src_pos == dst_pos)
+ continue; /* This is unrelated limitation. */
+ else
+ return true;
+ }
+ return false;
+}
+
+static int
+check_dst_limits_calc_pos_1 (const re_match_context_t *mctx, int boundaries,
+ Idx subexp_idx, Idx from_node, Idx bkref_idx)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ const re_node_set *eclosures = dfa->eclosures + from_node;
+ Idx node_idx;
+
+ /* Else, we are on the boundary: examine the nodes on the epsilon
+ closure. */
+ for (node_idx = 0; node_idx < eclosures->nelem; ++node_idx)
+ {
+ Idx node = eclosures->elems[node_idx];
+ switch (dfa->nodes[node].type)
+ {
+ case OP_BACK_REF:
+ if (bkref_idx != -1)
+ {
+ struct re_backref_cache_entry *ent = mctx->bkref_ents + bkref_idx;
+ do
+ {
+ Idx dst;
+ int cpos;
+
+ if (ent->node != node)
+ continue;
+
+ if (subexp_idx < BITSET_WORD_BITS
+ && !(ent->eps_reachable_subexps_map
+ & ((bitset_word_t) 1 << subexp_idx)))
+ continue;
+
+ /* Recurse trying to reach the OP_OPEN_SUBEXP and
+ OP_CLOSE_SUBEXP cases below. But, if the
+ destination node is the same node as the source
+ node, don't recurse because it would cause an
+ infinite loop: a regex that exhibits this behavior
+ is ()\1*\1* */
+ dst = dfa->edests[node].elems[0];
+ if (dst == from_node)
+ {
+ if (boundaries & 1)
+ return -1;
+ else /* if (boundaries & 2) */
+ return 0;
+ }
+
+ cpos =
+ check_dst_limits_calc_pos_1 (mctx, boundaries, subexp_idx,
+ dst, bkref_idx);
+ if (cpos == -1 /* && (boundaries & 1) */)
+ return -1;
+ if (cpos == 0 && (boundaries & 2))
+ return 0;
+
+ if (subexp_idx < BITSET_WORD_BITS)
+ ent->eps_reachable_subexps_map
+ &= ~((bitset_word_t) 1 << subexp_idx);
+ }
+ while (ent++->more);
+ }
+ break;
+
+ case OP_OPEN_SUBEXP:
+ if ((boundaries & 1) && subexp_idx == dfa->nodes[node].opr.idx)
+ return -1;
+ break;
+
+ case OP_CLOSE_SUBEXP:
+ if ((boundaries & 2) && subexp_idx == dfa->nodes[node].opr.idx)
+ return 0;
+ break;
+
+ default:
+ break;
+ }
+ }
+
+ return (boundaries & 2) ? 1 : 0;
+}
+
+static int
+check_dst_limits_calc_pos (const re_match_context_t *mctx, Idx limit,
+ Idx subexp_idx, Idx from_node, Idx str_idx,
+ Idx bkref_idx)
+{
+ struct re_backref_cache_entry *lim = mctx->bkref_ents + limit;
+ int boundaries;
+
+ /* If we are outside the range of the subexpression, return -1 or 1. */
+ if (str_idx < lim->subexp_from)
+ return -1;
+
+ if (lim->subexp_to < str_idx)
+ return 1;
+
+ /* If we are within the subexpression, return 0. */
+ boundaries = (str_idx == lim->subexp_from);
+ boundaries |= (str_idx == lim->subexp_to) << 1;
+ if (boundaries == 0)
+ return 0;
+
+ /* Else, examine epsilon closure. */
+ return check_dst_limits_calc_pos_1 (mctx, boundaries, subexp_idx,
+ from_node, bkref_idx);
+}
+
+/* Check the limitations of sub expressions LIMITS, and remove the nodes
+ which are against limitations from DEST_NODES. */
+
+static reg_errcode_t
+check_subexp_limits (const re_dfa_t *dfa, re_node_set *dest_nodes,
+ const re_node_set *candidates, re_node_set *limits,
+ struct re_backref_cache_entry *bkref_ents, Idx str_idx)
+{
+ reg_errcode_t err;
+ Idx node_idx, lim_idx;
+
+ for (lim_idx = 0; lim_idx < limits->nelem; ++lim_idx)
+ {
+ Idx subexp_idx;
+ struct re_backref_cache_entry *ent;
+ ent = bkref_ents + limits->elems[lim_idx];
+
+ if (str_idx <= ent->subexp_from || ent->str_idx < str_idx)
+ continue; /* This is unrelated limitation. */
+
+ subexp_idx = dfa->nodes[ent->node].opr.idx;
+ if (ent->subexp_to == str_idx)
+ {
+ Idx ops_node = -1;
+ Idx cls_node = -1;
+ for (node_idx = 0; node_idx < dest_nodes->nelem; ++node_idx)
+ {
+ Idx node = dest_nodes->elems[node_idx];
+ re_token_type_t type = dfa->nodes[node].type;
+ if (type == OP_OPEN_SUBEXP
+ && subexp_idx == dfa->nodes[node].opr.idx)
+ ops_node = node;
+ else if (type == OP_CLOSE_SUBEXP
+ && subexp_idx == dfa->nodes[node].opr.idx)
+ cls_node = node;
+ }
+
+ /* Check the limitation of the open subexpression. */
+ /* Note that (ent->subexp_to = str_idx != ent->subexp_from). */
+ if (ops_node >= 0)
+ {
+ err = sub_epsilon_src_nodes (dfa, ops_node, dest_nodes,
+ candidates);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+
+ /* Check the limitation of the close subexpression. */
+ if (cls_node >= 0)
+ for (node_idx = 0; node_idx < dest_nodes->nelem; ++node_idx)
+ {
+ Idx node = dest_nodes->elems[node_idx];
+ if (!re_node_set_contains (dfa->inveclosures + node,
+ cls_node)
+ && !re_node_set_contains (dfa->eclosures + node,
+ cls_node))
+ {
+ /* It is against this limitation.
+ Remove it form the current sifted state. */
+ err = sub_epsilon_src_nodes (dfa, node, dest_nodes,
+ candidates);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ --node_idx;
+ }
+ }
+ }
+ else /* (ent->subexp_to != str_idx) */
+ {
+ for (node_idx = 0; node_idx < dest_nodes->nelem; ++node_idx)
+ {
+ Idx node = dest_nodes->elems[node_idx];
+ re_token_type_t type = dfa->nodes[node].type;
+ if (type == OP_CLOSE_SUBEXP || type == OP_OPEN_SUBEXP)
+ {
+ if (subexp_idx != dfa->nodes[node].opr.idx)
+ continue;
+ /* It is against this limitation.
+ Remove it form the current sifted state. */
+ err = sub_epsilon_src_nodes (dfa, node, dest_nodes,
+ candidates);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ }
+ }
+ }
+ return REG_NOERROR;
+}
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+sift_states_bkref (const re_match_context_t *mctx, re_sift_context_t *sctx,
+ Idx str_idx, const re_node_set *candidates)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ reg_errcode_t err;
+ Idx node_idx, node;
+ re_sift_context_t local_sctx;
+ Idx first_idx = search_cur_bkref_entry (mctx, str_idx);
+
+ if (first_idx == -1)
+ return REG_NOERROR;
+
+ local_sctx.sifted_states = NULL; /* Mark that it hasn't been initialized. */
+
+ for (node_idx = 0; node_idx < candidates->nelem; ++node_idx)
+ {
+ Idx enabled_idx;
+ re_token_type_t type;
+ struct re_backref_cache_entry *entry;
+ node = candidates->elems[node_idx];
+ type = dfa->nodes[node].type;
+ /* Avoid infinite loop for the REs like "()\1+". */
+ if (node == sctx->last_node && str_idx == sctx->last_str_idx)
+ continue;
+ if (type != OP_BACK_REF)
+ continue;
+
+ entry = mctx->bkref_ents + first_idx;
+ enabled_idx = first_idx;
+ do
+ {
+ Idx subexp_len;
+ Idx to_idx;
+ Idx dst_node;
+ bool ok;
+ re_dfastate_t *cur_state;
+
+ if (entry->node != node)
+ continue;
+ subexp_len = entry->subexp_to - entry->subexp_from;
+ to_idx = str_idx + subexp_len;
+ dst_node = (subexp_len ? dfa->nexts[node]
+ : dfa->edests[node].elems[0]);
+
+ if (to_idx > sctx->last_str_idx
+ || sctx->sifted_states[to_idx] == NULL
+ || !STATE_NODE_CONTAINS (sctx->sifted_states[to_idx], dst_node)
+ || check_dst_limits (mctx, &sctx->limits, node,
+ str_idx, dst_node, to_idx))
+ continue;
+
+ if (local_sctx.sifted_states == NULL)
+ {
+ local_sctx = *sctx;
+ err = re_node_set_init_copy (&local_sctx.limits, &sctx->limits);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+ }
+ local_sctx.last_node = node;
+ local_sctx.last_str_idx = str_idx;
+ ok = re_node_set_insert (&local_sctx.limits, enabled_idx);
+ if (BE (! ok, 0))
+ {
+ err = REG_ESPACE;
+ goto free_return;
+ }
+ cur_state = local_sctx.sifted_states[str_idx];
+ err = sift_states_backward (mctx, &local_sctx);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+ if (sctx->limited_states != NULL)
+ {
+ err = merge_state_array (dfa, sctx->limited_states,
+ local_sctx.sifted_states,
+ str_idx + 1);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+ }
+ local_sctx.sifted_states[str_idx] = cur_state;
+ re_node_set_remove (&local_sctx.limits, enabled_idx);
+
+ /* mctx->bkref_ents may have changed, reload the pointer. */
+ entry = mctx->bkref_ents + enabled_idx;
+ }
+ while (enabled_idx++, entry++->more);
+ }
+ err = REG_NOERROR;
+ free_return:
+ if (local_sctx.sifted_states != NULL)
+ {
+ re_node_set_free (&local_sctx.limits);
+ }
+
+ return err;
+}
+
+
+#ifdef RE_ENABLE_I18N
+static int
+sift_states_iter_mb (const re_match_context_t *mctx, re_sift_context_t *sctx,
+ Idx node_idx, Idx str_idx, Idx max_str_idx)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ int naccepted;
+ /* Check the node can accept "multi byte". */
+ naccepted = check_node_accept_bytes (dfa, node_idx, &mctx->input, str_idx);
+ if (naccepted > 0 && str_idx + naccepted <= max_str_idx &&
+ !STATE_NODE_CONTAINS (sctx->sifted_states[str_idx + naccepted],
+ dfa->nexts[node_idx]))
+ /* The node can't accept the "multi byte", or the
+ destination was already thrown away, then the node
+ could't accept the current input "multi byte". */
+ naccepted = 0;
+ /* Otherwise, it is sure that the node could accept
+ 'naccepted' bytes input. */
+ return naccepted;
+}
+#endif /* RE_ENABLE_I18N */
+
+
+/* Functions for state transition. */
+
+/* Return the next state to which the current state STATE will transit by
+ accepting the current input byte, and update STATE_LOG if necessary.
+ If STATE can accept a multibyte char/collating element/back reference
+ update the destination of STATE_LOG. */
+
+static re_dfastate_t *
+__attribute_warn_unused_result__
+transit_state (reg_errcode_t *err, re_match_context_t *mctx,
+ re_dfastate_t *state)
+{
+ re_dfastate_t **trtable;
+ unsigned char ch;
+
+#ifdef RE_ENABLE_I18N
+ /* If the current state can accept multibyte. */
+ if (BE (state->accept_mb, 0))
+ {
+ *err = transit_state_mb (mctx, state);
+ if (BE (*err != REG_NOERROR, 0))
+ return NULL;
+ }
+#endif /* RE_ENABLE_I18N */
+
+ /* Then decide the next state with the single byte. */
+#if 0
+ if (0)
+ /* don't use transition table */
+ return transit_state_sb (err, mctx, state);
+#endif
+
+ /* Use transition table */
+ ch = re_string_fetch_byte (&mctx->input);
+ for (;;)
+ {
+ trtable = state->trtable;
+ if (BE (trtable != NULL, 1))
+ return trtable[ch];
+
+ trtable = state->word_trtable;
+ if (BE (trtable != NULL, 1))
+ {
+ unsigned int context;
+ context
+ = re_string_context_at (&mctx->input,
+ re_string_cur_idx (&mctx->input) - 1,
+ mctx->eflags);
+ if (IS_WORD_CONTEXT (context))
+ return trtable[ch + SBC_MAX];
+ else
+ return trtable[ch];
+ }
+
+ if (!build_trtable (mctx->dfa, state))
+ {
+ *err = REG_ESPACE;
+ return NULL;
+ }
+
+ /* Retry, we now have a transition table. */
+ }
+}
+
+/* Update the state_log if we need */
+static re_dfastate_t *
+merge_state_with_log (reg_errcode_t *err, re_match_context_t *mctx,
+ re_dfastate_t *next_state)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ Idx cur_idx = re_string_cur_idx (&mctx->input);
+
+ if (cur_idx > mctx->state_log_top)
+ {
+ mctx->state_log[cur_idx] = next_state;
+ mctx->state_log_top = cur_idx;
+ }
+ else if (mctx->state_log[cur_idx] == 0)
+ {
+ mctx->state_log[cur_idx] = next_state;
+ }
+ else
+ {
+ re_dfastate_t *pstate;
+ unsigned int context;
+ re_node_set next_nodes, *log_nodes, *table_nodes = NULL;
+ /* If (state_log[cur_idx] != 0), it implies that cur_idx is
+ the destination of a multibyte char/collating element/
+ back reference. Then the next state is the union set of
+ these destinations and the results of the transition table. */
+ pstate = mctx->state_log[cur_idx];
+ log_nodes = pstate->entrance_nodes;
+ if (next_state != NULL)
+ {
+ table_nodes = next_state->entrance_nodes;
+ *err = re_node_set_init_union (&next_nodes, table_nodes,
+ log_nodes);
+ if (BE (*err != REG_NOERROR, 0))
+ return NULL;
+ }
+ else
+ next_nodes = *log_nodes;
+ /* Note: We already add the nodes of the initial state,
+ then we don't need to add them here. */
+
+ context = re_string_context_at (&mctx->input,
+ re_string_cur_idx (&mctx->input) - 1,
+ mctx->eflags);
+ next_state = mctx->state_log[cur_idx]
+ = re_acquire_state_context (err, dfa, &next_nodes, context);
+ /* We don't need to check errors here, since the return value of
+ this function is next_state and ERR is already set. */
+
+ if (table_nodes != NULL)
+ re_node_set_free (&next_nodes);
+ }
+
+ if (BE (dfa->nbackref, 0) && next_state != NULL)
+ {
+ /* Check OP_OPEN_SUBEXP in the current state in case that we use them
+ later. We must check them here, since the back references in the
+ next state might use them. */
+ *err = check_subexp_matching_top (mctx, &next_state->nodes,
+ cur_idx);
+ if (BE (*err != REG_NOERROR, 0))
+ return NULL;
+
+ /* If the next state has back references. */
+ if (next_state->has_backref)
+ {
+ *err = transit_state_bkref (mctx, &next_state->nodes);
+ if (BE (*err != REG_NOERROR, 0))
+ return NULL;
+ next_state = mctx->state_log[cur_idx];
+ }
+ }
+
+ return next_state;
+}
+
+/* Skip bytes in the input that correspond to part of a
+ multi-byte match, then look in the log for a state
+ from which to restart matching. */
+static re_dfastate_t *
+find_recover_state (reg_errcode_t *err, re_match_context_t *mctx)
+{
+ re_dfastate_t *cur_state;
+ do
+ {
+ Idx max = mctx->state_log_top;
+ Idx cur_str_idx = re_string_cur_idx (&mctx->input);
+
+ do
+ {
+ if (++cur_str_idx > max)
+ return NULL;
+ re_string_skip_bytes (&mctx->input, 1);
+ }
+ while (mctx->state_log[cur_str_idx] == NULL);
+
+ cur_state = merge_state_with_log (err, mctx, NULL);
+ }
+ while (*err == REG_NOERROR && cur_state == NULL);
+ return cur_state;
+}
+
+/* Helper functions for transit_state. */
+
+/* From the node set CUR_NODES, pick up the nodes whose types are
+ OP_OPEN_SUBEXP and which have corresponding back references in the regular
+ expression. And register them to use them later for evaluating the
+ corresponding back references. */
+
+static reg_errcode_t
+check_subexp_matching_top (re_match_context_t *mctx, re_node_set *cur_nodes,
+ Idx str_idx)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ Idx node_idx;
+ reg_errcode_t err;
+
+ /* TODO: This isn't efficient.
+ Because there might be more than one nodes whose types are
+ OP_OPEN_SUBEXP and whose index is SUBEXP_IDX, we must check all
+ nodes.
+ E.g. RE: (a){2} */
+ for (node_idx = 0; node_idx < cur_nodes->nelem; ++node_idx)
+ {
+ Idx node = cur_nodes->elems[node_idx];
+ if (dfa->nodes[node].type == OP_OPEN_SUBEXP
+ && dfa->nodes[node].opr.idx < BITSET_WORD_BITS
+ && (dfa->used_bkref_map
+ & ((bitset_word_t) 1 << dfa->nodes[node].opr.idx)))
+ {
+ err = match_ctx_add_subtop (mctx, node, str_idx);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ }
+ return REG_NOERROR;
+}
+
+#if 0
+/* Return the next state to which the current state STATE will transit by
+ accepting the current input byte. */
+
+static re_dfastate_t *
+transit_state_sb (reg_errcode_t *err, re_match_context_t *mctx,
+ re_dfastate_t *state)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ re_node_set next_nodes;
+ re_dfastate_t *next_state;
+ Idx node_cnt, cur_str_idx = re_string_cur_idx (&mctx->input);
+ unsigned int context;
+
+ *err = re_node_set_alloc (&next_nodes, state->nodes.nelem + 1);
+ if (BE (*err != REG_NOERROR, 0))
+ return NULL;
+ for (node_cnt = 0; node_cnt < state->nodes.nelem; ++node_cnt)
+ {
+ Idx cur_node = state->nodes.elems[node_cnt];
+ if (check_node_accept (mctx, dfa->nodes + cur_node, cur_str_idx))
+ {
+ *err = re_node_set_merge (&next_nodes,
+ dfa->eclosures + dfa->nexts[cur_node]);
+ if (BE (*err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&next_nodes);
+ return NULL;
+ }
+ }
+ }
+ context = re_string_context_at (&mctx->input, cur_str_idx, mctx->eflags);
+ next_state = re_acquire_state_context (err, dfa, &next_nodes, context);
+ /* We don't need to check errors here, since the return value of
+ this function is next_state and ERR is already set. */
+
+ re_node_set_free (&next_nodes);
+ re_string_skip_bytes (&mctx->input, 1);
+ return next_state;
+}
+#endif
+
+#ifdef RE_ENABLE_I18N
+static reg_errcode_t
+transit_state_mb (re_match_context_t *mctx, re_dfastate_t *pstate)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ reg_errcode_t err;
+ Idx i;
+
+ for (i = 0; i < pstate->nodes.nelem; ++i)
+ {
+ re_node_set dest_nodes, *new_nodes;
+ Idx cur_node_idx = pstate->nodes.elems[i];
+ int naccepted;
+ Idx dest_idx;
+ unsigned int context;
+ re_dfastate_t *dest_state;
+
+ if (!dfa->nodes[cur_node_idx].accept_mb)
+ continue;
+
+ if (dfa->nodes[cur_node_idx].constraint)
+ {
+ context = re_string_context_at (&mctx->input,
+ re_string_cur_idx (&mctx->input),
+ mctx->eflags);
+ if (NOT_SATISFY_NEXT_CONSTRAINT (dfa->nodes[cur_node_idx].constraint,
+ context))
+ continue;
+ }
+
+ /* How many bytes the node can accept? */
+ naccepted = check_node_accept_bytes (dfa, cur_node_idx, &mctx->input,
+ re_string_cur_idx (&mctx->input));
+ if (naccepted == 0)
+ continue;
+
+ /* The node can accepts 'naccepted' bytes. */
+ dest_idx = re_string_cur_idx (&mctx->input) + naccepted;
+ mctx->max_mb_elem_len = ((mctx->max_mb_elem_len < naccepted) ? naccepted
+ : mctx->max_mb_elem_len);
+ err = clean_state_log_if_needed (mctx, dest_idx);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+#ifdef DEBUG
+ assert (dfa->nexts[cur_node_idx] != -1);
+#endif
+ new_nodes = dfa->eclosures + dfa->nexts[cur_node_idx];
+
+ dest_state = mctx->state_log[dest_idx];
+ if (dest_state == NULL)
+ dest_nodes = *new_nodes;
+ else
+ {
+ err = re_node_set_init_union (&dest_nodes,
+ dest_state->entrance_nodes, new_nodes);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ context = re_string_context_at (&mctx->input, dest_idx - 1,
+ mctx->eflags);
+ mctx->state_log[dest_idx]
+ = re_acquire_state_context (&err, dfa, &dest_nodes, context);
+ if (dest_state != NULL)
+ re_node_set_free (&dest_nodes);
+ if (BE (mctx->state_log[dest_idx] == NULL && err != REG_NOERROR, 0))
+ return err;
+ }
+ return REG_NOERROR;
+}
+#endif /* RE_ENABLE_I18N */
+
+static reg_errcode_t
+transit_state_bkref (re_match_context_t *mctx, const re_node_set *nodes)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ reg_errcode_t err;
+ Idx i;
+ Idx cur_str_idx = re_string_cur_idx (&mctx->input);
+
+ for (i = 0; i < nodes->nelem; ++i)
+ {
+ Idx dest_str_idx, prev_nelem, bkc_idx;
+ Idx node_idx = nodes->elems[i];
+ unsigned int context;
+ const re_token_t *node = dfa->nodes + node_idx;
+ re_node_set *new_dest_nodes;
+
+ /* Check whether 'node' is a backreference or not. */
+ if (node->type != OP_BACK_REF)
+ continue;
+
+ if (node->constraint)
+ {
+ context = re_string_context_at (&mctx->input, cur_str_idx,
+ mctx->eflags);
+ if (NOT_SATISFY_NEXT_CONSTRAINT (node->constraint, context))
+ continue;
+ }
+
+ /* 'node' is a backreference.
+ Check the substring which the substring matched. */
+ bkc_idx = mctx->nbkref_ents;
+ err = get_subexp (mctx, node_idx, cur_str_idx);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+
+ /* And add the epsilon closures (which is 'new_dest_nodes') of
+ the backreference to appropriate state_log. */
+#ifdef DEBUG
+ assert (dfa->nexts[node_idx] != -1);
+#endif
+ for (; bkc_idx < mctx->nbkref_ents; ++bkc_idx)
+ {
+ Idx subexp_len;
+ re_dfastate_t *dest_state;
+ struct re_backref_cache_entry *bkref_ent;
+ bkref_ent = mctx->bkref_ents + bkc_idx;
+ if (bkref_ent->node != node_idx || bkref_ent->str_idx != cur_str_idx)
+ continue;
+ subexp_len = bkref_ent->subexp_to - bkref_ent->subexp_from;
+ new_dest_nodes = (subexp_len == 0
+ ? dfa->eclosures + dfa->edests[node_idx].elems[0]
+ : dfa->eclosures + dfa->nexts[node_idx]);
+ dest_str_idx = (cur_str_idx + bkref_ent->subexp_to
+ - bkref_ent->subexp_from);
+ context = re_string_context_at (&mctx->input, dest_str_idx - 1,
+ mctx->eflags);
+ dest_state = mctx->state_log[dest_str_idx];
+ prev_nelem = ((mctx->state_log[cur_str_idx] == NULL) ? 0
+ : mctx->state_log[cur_str_idx]->nodes.nelem);
+ /* Add 'new_dest_node' to state_log. */
+ if (dest_state == NULL)
+ {
+ mctx->state_log[dest_str_idx]
+ = re_acquire_state_context (&err, dfa, new_dest_nodes,
+ context);
+ if (BE (mctx->state_log[dest_str_idx] == NULL
+ && err != REG_NOERROR, 0))
+ goto free_return;
+ }
+ else
+ {
+ re_node_set dest_nodes;
+ err = re_node_set_init_union (&dest_nodes,
+ dest_state->entrance_nodes,
+ new_dest_nodes);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&dest_nodes);
+ goto free_return;
+ }
+ mctx->state_log[dest_str_idx]
+ = re_acquire_state_context (&err, dfa, &dest_nodes, context);
+ re_node_set_free (&dest_nodes);
+ if (BE (mctx->state_log[dest_str_idx] == NULL
+ && err != REG_NOERROR, 0))
+ goto free_return;
+ }
+ /* We need to check recursively if the backreference can epsilon
+ transit. */
+ if (subexp_len == 0
+ && mctx->state_log[cur_str_idx]->nodes.nelem > prev_nelem)
+ {
+ err = check_subexp_matching_top (mctx, new_dest_nodes,
+ cur_str_idx);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+ err = transit_state_bkref (mctx, new_dest_nodes);
+ if (BE (err != REG_NOERROR, 0))
+ goto free_return;
+ }
+ }
+ }
+ err = REG_NOERROR;
+ free_return:
+ return err;
+}
+
+/* Enumerate all the candidates which the backreference BKREF_NODE can match
+ at BKREF_STR_IDX, and register them by match_ctx_add_entry().
+ Note that we might collect inappropriate candidates here.
+ However, the cost of checking them strictly here is too high, then we
+ delay these checking for prune_impossible_nodes(). */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+get_subexp (re_match_context_t *mctx, Idx bkref_node, Idx bkref_str_idx)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ Idx subexp_num, sub_top_idx;
+ const char *buf = (const char *) re_string_get_buffer (&mctx->input);
+ /* Return if we have already checked BKREF_NODE at BKREF_STR_IDX. */
+ Idx cache_idx = search_cur_bkref_entry (mctx, bkref_str_idx);
+ if (cache_idx != -1)
+ {
+ const struct re_backref_cache_entry *entry
+ = mctx->bkref_ents + cache_idx;
+ do
+ if (entry->node == bkref_node)
+ return REG_NOERROR; /* We already checked it. */
+ while (entry++->more);
+ }
+
+ subexp_num = dfa->nodes[bkref_node].opr.idx;
+
+ /* For each sub expression */
+ for (sub_top_idx = 0; sub_top_idx < mctx->nsub_tops; ++sub_top_idx)
+ {
+ reg_errcode_t err;
+ re_sub_match_top_t *sub_top = mctx->sub_tops[sub_top_idx];
+ re_sub_match_last_t *sub_last;
+ Idx sub_last_idx, sl_str, bkref_str_off;
+
+ if (dfa->nodes[sub_top->node].opr.idx != subexp_num)
+ continue; /* It isn't related. */
+
+ sl_str = sub_top->str_idx;
+ bkref_str_off = bkref_str_idx;
+ /* At first, check the last node of sub expressions we already
+ evaluated. */
+ for (sub_last_idx = 0; sub_last_idx < sub_top->nlasts; ++sub_last_idx)
+ {
+ regoff_t sl_str_diff;
+ sub_last = sub_top->lasts[sub_last_idx];
+ sl_str_diff = sub_last->str_idx - sl_str;
+ /* The matched string by the sub expression match with the substring
+ at the back reference? */
+ if (sl_str_diff > 0)
+ {
+ if (BE (bkref_str_off + sl_str_diff > mctx->input.valid_len, 0))
+ {
+ /* Not enough chars for a successful match. */
+ if (bkref_str_off + sl_str_diff > mctx->input.len)
+ break;
+
+ err = clean_state_log_if_needed (mctx,
+ bkref_str_off
+ + sl_str_diff);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ buf = (const char *) re_string_get_buffer (&mctx->input);
+ }
+ if (memcmp (buf + bkref_str_off, buf + sl_str, sl_str_diff) != 0)
+ /* We don't need to search this sub expression any more. */
+ break;
+ }
+ bkref_str_off += sl_str_diff;
+ sl_str += sl_str_diff;
+ err = get_subexp_sub (mctx, sub_top, sub_last, bkref_node,
+ bkref_str_idx);
+
+ /* Reload buf, since the preceding call might have reallocated
+ the buffer. */
+ buf = (const char *) re_string_get_buffer (&mctx->input);
+
+ if (err == REG_NOMATCH)
+ continue;
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+
+ if (sub_last_idx < sub_top->nlasts)
+ continue;
+ if (sub_last_idx > 0)
+ ++sl_str;
+ /* Then, search for the other last nodes of the sub expression. */
+ for (; sl_str <= bkref_str_idx; ++sl_str)
+ {
+ Idx cls_node;
+ regoff_t sl_str_off;
+ const re_node_set *nodes;
+ sl_str_off = sl_str - sub_top->str_idx;
+ /* The matched string by the sub expression match with the substring
+ at the back reference? */
+ if (sl_str_off > 0)
+ {
+ if (BE (bkref_str_off >= mctx->input.valid_len, 0))
+ {
+ /* If we are at the end of the input, we cannot match. */
+ if (bkref_str_off >= mctx->input.len)
+ break;
+
+ err = extend_buffers (mctx, bkref_str_off + 1);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+
+ buf = (const char *) re_string_get_buffer (&mctx->input);
+ }
+ if (buf [bkref_str_off++] != buf[sl_str - 1])
+ break; /* We don't need to search this sub expression
+ any more. */
+ }
+ if (mctx->state_log[sl_str] == NULL)
+ continue;
+ /* Does this state have a ')' of the sub expression? */
+ nodes = &mctx->state_log[sl_str]->nodes;
+ cls_node = find_subexp_node (dfa, nodes, subexp_num,
+ OP_CLOSE_SUBEXP);
+ if (cls_node == -1)
+ continue; /* No. */
+ if (sub_top->path == NULL)
+ {
+ sub_top->path = calloc (sizeof (state_array_t),
+ sl_str - sub_top->str_idx + 1);
+ if (sub_top->path == NULL)
+ return REG_ESPACE;
+ }
+ /* Can the OP_OPEN_SUBEXP node arrive the OP_CLOSE_SUBEXP node
+ in the current context? */
+ err = check_arrival (mctx, sub_top->path, sub_top->node,
+ sub_top->str_idx, cls_node, sl_str,
+ OP_CLOSE_SUBEXP);
+ if (err == REG_NOMATCH)
+ continue;
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ sub_last = match_ctx_add_sublast (sub_top, cls_node, sl_str);
+ if (BE (sub_last == NULL, 0))
+ return REG_ESPACE;
+ err = get_subexp_sub (mctx, sub_top, sub_last, bkref_node,
+ bkref_str_idx);
+ if (err == REG_NOMATCH)
+ continue;
+ }
+ }
+ return REG_NOERROR;
+}
+
+/* Helper functions for get_subexp(). */
+
+/* Check SUB_LAST can arrive to the back reference BKREF_NODE at BKREF_STR.
+ If it can arrive, register the sub expression expressed with SUB_TOP
+ and SUB_LAST. */
+
+static reg_errcode_t
+get_subexp_sub (re_match_context_t *mctx, const re_sub_match_top_t *sub_top,
+ re_sub_match_last_t *sub_last, Idx bkref_node, Idx bkref_str)
+{
+ reg_errcode_t err;
+ Idx to_idx;
+ /* Can the subexpression arrive the back reference? */
+ err = check_arrival (mctx, &sub_last->path, sub_last->node,
+ sub_last->str_idx, bkref_node, bkref_str,
+ OP_OPEN_SUBEXP);
+ if (err != REG_NOERROR)
+ return err;
+ err = match_ctx_add_entry (mctx, bkref_node, bkref_str, sub_top->str_idx,
+ sub_last->str_idx);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ to_idx = bkref_str + sub_last->str_idx - sub_top->str_idx;
+ return clean_state_log_if_needed (mctx, to_idx);
+}
+
+/* Find the first node which is '(' or ')' and whose index is SUBEXP_IDX.
+ Search '(' if FL_OPEN, or search ')' otherwise.
+ TODO: This function isn't efficient...
+ Because there might be more than one nodes whose types are
+ OP_OPEN_SUBEXP and whose index is SUBEXP_IDX, we must check all
+ nodes.
+ E.g. RE: (a){2} */
+
+static Idx
+find_subexp_node (const re_dfa_t *dfa, const re_node_set *nodes,
+ Idx subexp_idx, int type)
+{
+ Idx cls_idx;
+ for (cls_idx = 0; cls_idx < nodes->nelem; ++cls_idx)
+ {
+ Idx cls_node = nodes->elems[cls_idx];
+ const re_token_t *node = dfa->nodes + cls_node;
+ if (node->type == type
+ && node->opr.idx == subexp_idx)
+ return cls_node;
+ }
+ return -1;
+}
+
+/* Check whether the node TOP_NODE at TOP_STR can arrive to the node
+ LAST_NODE at LAST_STR. We record the path onto PATH since it will be
+ heavily reused.
+ Return REG_NOERROR if it can arrive, or REG_NOMATCH otherwise. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+check_arrival (re_match_context_t *mctx, state_array_t *path, Idx top_node,
+ Idx top_str, Idx last_node, Idx last_str, int type)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ reg_errcode_t err = REG_NOERROR;
+ Idx subexp_num, backup_cur_idx, str_idx, null_cnt;
+ re_dfastate_t *cur_state = NULL;
+ re_node_set *cur_nodes, next_nodes;
+ re_dfastate_t **backup_state_log;
+ unsigned int context;
+
+ subexp_num = dfa->nodes[top_node].opr.idx;
+ /* Extend the buffer if we need. */
+ if (BE (path->alloc < last_str + mctx->max_mb_elem_len + 1, 0))
+ {
+ re_dfastate_t **new_array;
+ Idx old_alloc = path->alloc;
+ Idx incr_alloc = last_str + mctx->max_mb_elem_len + 1;
+ Idx new_alloc;
+ if (BE (IDX_MAX - old_alloc < incr_alloc, 0))
+ return REG_ESPACE;
+ new_alloc = old_alloc + incr_alloc;
+ if (BE (SIZE_MAX / sizeof (re_dfastate_t *) < new_alloc, 0))
+ return REG_ESPACE;
+ new_array = re_realloc (path->array, re_dfastate_t *, new_alloc);
+ if (BE (new_array == NULL, 0))
+ return REG_ESPACE;
+ path->array = new_array;
+ path->alloc = new_alloc;
+ memset (new_array + old_alloc, '\0',
+ sizeof (re_dfastate_t *) * (path->alloc - old_alloc));
+ }
+
+ str_idx = path->next_idx ? path->next_idx : top_str;
+
+ /* Temporary modify MCTX. */
+ backup_state_log = mctx->state_log;
+ backup_cur_idx = mctx->input.cur_idx;
+ mctx->state_log = path->array;
+ mctx->input.cur_idx = str_idx;
+
+ /* Setup initial node set. */
+ context = re_string_context_at (&mctx->input, str_idx - 1, mctx->eflags);
+ if (str_idx == top_str)
+ {
+ err = re_node_set_init_1 (&next_nodes, top_node);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ err = check_arrival_expand_ecl (dfa, &next_nodes, subexp_num, type);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&next_nodes);
+ return err;
+ }
+ }
+ else
+ {
+ cur_state = mctx->state_log[str_idx];
+ if (cur_state && cur_state->has_backref)
+ {
+ err = re_node_set_init_copy (&next_nodes, &cur_state->nodes);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ else
+ re_node_set_init_empty (&next_nodes);
+ }
+ if (str_idx == top_str || (cur_state && cur_state->has_backref))
+ {
+ if (next_nodes.nelem)
+ {
+ err = expand_bkref_cache (mctx, &next_nodes, str_idx,
+ subexp_num, type);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&next_nodes);
+ return err;
+ }
+ }
+ cur_state = re_acquire_state_context (&err, dfa, &next_nodes, context);
+ if (BE (cur_state == NULL && err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&next_nodes);
+ return err;
+ }
+ mctx->state_log[str_idx] = cur_state;
+ }
+
+ for (null_cnt = 0; str_idx < last_str && null_cnt <= mctx->max_mb_elem_len;)
+ {
+ re_node_set_empty (&next_nodes);
+ if (mctx->state_log[str_idx + 1])
+ {
+ err = re_node_set_merge (&next_nodes,
+ &mctx->state_log[str_idx + 1]->nodes);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&next_nodes);
+ return err;
+ }
+ }
+ if (cur_state)
+ {
+ err = check_arrival_add_next_nodes (mctx, str_idx,
+ &cur_state->non_eps_nodes,
+ &next_nodes);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&next_nodes);
+ return err;
+ }
+ }
+ ++str_idx;
+ if (next_nodes.nelem)
+ {
+ err = check_arrival_expand_ecl (dfa, &next_nodes, subexp_num, type);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&next_nodes);
+ return err;
+ }
+ err = expand_bkref_cache (mctx, &next_nodes, str_idx,
+ subexp_num, type);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&next_nodes);
+ return err;
+ }
+ }
+ context = re_string_context_at (&mctx->input, str_idx - 1, mctx->eflags);
+ cur_state = re_acquire_state_context (&err, dfa, &next_nodes, context);
+ if (BE (cur_state == NULL && err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&next_nodes);
+ return err;
+ }
+ mctx->state_log[str_idx] = cur_state;
+ null_cnt = cur_state == NULL ? null_cnt + 1 : 0;
+ }
+ re_node_set_free (&next_nodes);
+ cur_nodes = (mctx->state_log[last_str] == NULL ? NULL
+ : &mctx->state_log[last_str]->nodes);
+ path->next_idx = str_idx;
+
+ /* Fix MCTX. */
+ mctx->state_log = backup_state_log;
+ mctx->input.cur_idx = backup_cur_idx;
+
+ /* Then check the current node set has the node LAST_NODE. */
+ if (cur_nodes != NULL && re_node_set_contains (cur_nodes, last_node))
+ return REG_NOERROR;
+
+ return REG_NOMATCH;
+}
+
+/* Helper functions for check_arrival. */
+
+/* Calculate the destination nodes of CUR_NODES at STR_IDX, and append them
+ to NEXT_NODES.
+ TODO: This function is similar to the functions transit_state*(),
+ however this function has many additional works.
+ Can't we unify them? */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+check_arrival_add_next_nodes (re_match_context_t *mctx, Idx str_idx,
+ re_node_set *cur_nodes, re_node_set *next_nodes)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ bool ok;
+ Idx cur_idx;
+#ifdef RE_ENABLE_I18N
+ reg_errcode_t err = REG_NOERROR;
+#endif
+ re_node_set union_set;
+ re_node_set_init_empty (&union_set);
+ for (cur_idx = 0; cur_idx < cur_nodes->nelem; ++cur_idx)
+ {
+ int naccepted = 0;
+ Idx cur_node = cur_nodes->elems[cur_idx];
+#ifdef DEBUG
+ re_token_type_t type = dfa->nodes[cur_node].type;
+ assert (!IS_EPSILON_NODE (type));
+#endif
+#ifdef RE_ENABLE_I18N
+ /* If the node may accept "multi byte". */
+ if (dfa->nodes[cur_node].accept_mb)
+ {
+ naccepted = check_node_accept_bytes (dfa, cur_node, &mctx->input,
+ str_idx);
+ if (naccepted > 1)
+ {
+ re_dfastate_t *dest_state;
+ Idx next_node = dfa->nexts[cur_node];
+ Idx next_idx = str_idx + naccepted;
+ dest_state = mctx->state_log[next_idx];
+ re_node_set_empty (&union_set);
+ if (dest_state)
+ {
+ err = re_node_set_merge (&union_set, &dest_state->nodes);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&union_set);
+ return err;
+ }
+ }
+ ok = re_node_set_insert (&union_set, next_node);
+ if (BE (! ok, 0))
+ {
+ re_node_set_free (&union_set);
+ return REG_ESPACE;
+ }
+ mctx->state_log[next_idx] = re_acquire_state (&err, dfa,
+ &union_set);
+ if (BE (mctx->state_log[next_idx] == NULL
+ && err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&union_set);
+ return err;
+ }
+ }
+ }
+#endif /* RE_ENABLE_I18N */
+ if (naccepted
+ || check_node_accept (mctx, dfa->nodes + cur_node, str_idx))
+ {
+ ok = re_node_set_insert (next_nodes, dfa->nexts[cur_node]);
+ if (BE (! ok, 0))
+ {
+ re_node_set_free (&union_set);
+ return REG_ESPACE;
+ }
+ }
+ }
+ re_node_set_free (&union_set);
+ return REG_NOERROR;
+}
+
+/* For all the nodes in CUR_NODES, add the epsilon closures of them to
+ CUR_NODES, however exclude the nodes which are:
+ - inside the sub expression whose number is EX_SUBEXP, if FL_OPEN.
+ - out of the sub expression whose number is EX_SUBEXP, if !FL_OPEN.
+*/
+
+static reg_errcode_t
+check_arrival_expand_ecl (const re_dfa_t *dfa, re_node_set *cur_nodes,
+ Idx ex_subexp, int type)
+{
+ reg_errcode_t err;
+ Idx idx, outside_node;
+ re_node_set new_nodes;
+#ifdef DEBUG
+ assert (cur_nodes->nelem);
+#endif
+ err = re_node_set_alloc (&new_nodes, cur_nodes->nelem);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ /* Create a new node set NEW_NODES with the nodes which are epsilon
+ closures of the node in CUR_NODES. */
+
+ for (idx = 0; idx < cur_nodes->nelem; ++idx)
+ {
+ Idx cur_node = cur_nodes->elems[idx];
+ const re_node_set *eclosure = dfa->eclosures + cur_node;
+ outside_node = find_subexp_node (dfa, eclosure, ex_subexp, type);
+ if (outside_node == -1)
+ {
+ /* There are no problematic nodes, just merge them. */
+ err = re_node_set_merge (&new_nodes, eclosure);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&new_nodes);
+ return err;
+ }
+ }
+ else
+ {
+ /* There are problematic nodes, re-calculate incrementally. */
+ err = check_arrival_expand_ecl_sub (dfa, &new_nodes, cur_node,
+ ex_subexp, type);
+ if (BE (err != REG_NOERROR, 0))
+ {
+ re_node_set_free (&new_nodes);
+ return err;
+ }
+ }
+ }
+ re_node_set_free (cur_nodes);
+ *cur_nodes = new_nodes;
+ return REG_NOERROR;
+}
+
+/* Helper function for check_arrival_expand_ecl.
+ Check incrementally the epsilon closure of TARGET, and if it isn't
+ problematic append it to DST_NODES. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+check_arrival_expand_ecl_sub (const re_dfa_t *dfa, re_node_set *dst_nodes,
+ Idx target, Idx ex_subexp, int type)
+{
+ Idx cur_node;
+ for (cur_node = target; !re_node_set_contains (dst_nodes, cur_node);)
+ {
+ bool ok;
+
+ if (dfa->nodes[cur_node].type == type
+ && dfa->nodes[cur_node].opr.idx == ex_subexp)
+ {
+ if (type == OP_CLOSE_SUBEXP)
+ {
+ ok = re_node_set_insert (dst_nodes, cur_node);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ }
+ break;
+ }
+ ok = re_node_set_insert (dst_nodes, cur_node);
+ if (BE (! ok, 0))
+ return REG_ESPACE;
+ if (dfa->edests[cur_node].nelem == 0)
+ break;
+ if (dfa->edests[cur_node].nelem == 2)
+ {
+ reg_errcode_t err;
+ err = check_arrival_expand_ecl_sub (dfa, dst_nodes,
+ dfa->edests[cur_node].elems[1],
+ ex_subexp, type);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ cur_node = dfa->edests[cur_node].elems[0];
+ }
+ return REG_NOERROR;
+}
+
+
+/* For all the back references in the current state, calculate the
+ destination of the back references by the appropriate entry
+ in MCTX->BKREF_ENTS. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+expand_bkref_cache (re_match_context_t *mctx, re_node_set *cur_nodes,
+ Idx cur_str, Idx subexp_num, int type)
+{
+ const re_dfa_t *const dfa = mctx->dfa;
+ reg_errcode_t err;
+ Idx cache_idx_start = search_cur_bkref_entry (mctx, cur_str);
+ struct re_backref_cache_entry *ent;
+
+ if (cache_idx_start == -1)
+ return REG_NOERROR;
+
+ restart:
+ ent = mctx->bkref_ents + cache_idx_start;
+ do
+ {
+ Idx to_idx, next_node;
+
+ /* Is this entry ENT is appropriate? */
+ if (!re_node_set_contains (cur_nodes, ent->node))
+ continue; /* No. */
+
+ to_idx = cur_str + ent->subexp_to - ent->subexp_from;
+ /* Calculate the destination of the back reference, and append it
+ to MCTX->STATE_LOG. */
+ if (to_idx == cur_str)
+ {
+ /* The backreference did epsilon transit, we must re-check all the
+ node in the current state. */
+ re_node_set new_dests;
+ reg_errcode_t err2, err3;
+ next_node = dfa->edests[ent->node].elems[0];
+ if (re_node_set_contains (cur_nodes, next_node))
+ continue;
+ err = re_node_set_init_1 (&new_dests, next_node);
+ err2 = check_arrival_expand_ecl (dfa, &new_dests, subexp_num, type);
+ err3 = re_node_set_merge (cur_nodes, &new_dests);
+ re_node_set_free (&new_dests);
+ if (BE (err != REG_NOERROR || err2 != REG_NOERROR
+ || err3 != REG_NOERROR, 0))
+ {
+ err = (err != REG_NOERROR ? err
+ : (err2 != REG_NOERROR ? err2 : err3));
+ return err;
+ }
+ /* TODO: It is still inefficient... */
+ goto restart;
+ }
+ else
+ {
+ re_node_set union_set;
+ next_node = dfa->nexts[ent->node];
+ if (mctx->state_log[to_idx])
+ {
+ bool ok;
+ if (re_node_set_contains (&mctx->state_log[to_idx]->nodes,
+ next_node))
+ continue;
+ err = re_node_set_init_copy (&union_set,
+ &mctx->state_log[to_idx]->nodes);
+ ok = re_node_set_insert (&union_set, next_node);
+ if (BE (err != REG_NOERROR || ! ok, 0))
+ {
+ re_node_set_free (&union_set);
+ err = err != REG_NOERROR ? err : REG_ESPACE;
+ return err;
+ }
+ }
+ else
+ {
+ err = re_node_set_init_1 (&union_set, next_node);
+ if (BE (err != REG_NOERROR, 0))
+ return err;
+ }
+ mctx->state_log[to_idx] = re_acquire_state (&err, dfa, &union_set);
+ re_node_set_free (&union_set);
+ if (BE (mctx->state_log[to_idx] == NULL
+ && err != REG_NOERROR, 0))
+ return err;
+ }
+ }
+ while (ent++->more);
+ return REG_NOERROR;
+}
+
+/* Build transition table for the state.
+ Return true if successful. */
+
+static bool
+build_trtable (const re_dfa_t *dfa, re_dfastate_t *state)
+{
+ reg_errcode_t err;
+ Idx i, j;
+ int ch;
+ bool need_word_trtable = false;
+ bitset_word_t elem, mask;
+ bool dests_node_malloced = false;
+ bool dest_states_malloced = false;
+ Idx ndests; /* Number of the destination states from 'state'. */
+ re_dfastate_t **trtable;
+ re_dfastate_t **dest_states = NULL, **dest_states_word, **dest_states_nl;
+ re_node_set follows, *dests_node;
+ bitset_t *dests_ch;
+ bitset_t acceptable;
+
+ struct dests_alloc
+ {
+ re_node_set dests_node[SBC_MAX];
+ bitset_t dests_ch[SBC_MAX];
+ } *dests_alloc;
+
+ /* We build DFA states which corresponds to the destination nodes
+ from 'state'. 'dests_node[i]' represents the nodes which i-th
+ destination state contains, and 'dests_ch[i]' represents the
+ characters which i-th destination state accepts. */
+ if (__libc_use_alloca (sizeof (struct dests_alloc)))
+ dests_alloc = (struct dests_alloc *) alloca (sizeof (struct dests_alloc));
+ else
+ {
+ dests_alloc = re_malloc (struct dests_alloc, 1);
+ if (BE (dests_alloc == NULL, 0))
+ return false;
+ dests_node_malloced = true;
+ }
+ dests_node = dests_alloc->dests_node;
+ dests_ch = dests_alloc->dests_ch;
+
+ /* Initialize transition table. */
+ state->word_trtable = state->trtable = NULL;
+
+ /* At first, group all nodes belonging to 'state' into several
+ destinations. */
+ ndests = group_nodes_into_DFAstates (dfa, state, dests_node, dests_ch);
+ if (BE (ndests <= 0, 0))
+ {
+ if (dests_node_malloced)
+ re_free (dests_alloc);
+ /* Return false in case of an error, true otherwise. */
+ if (ndests == 0)
+ {
+ state->trtable = (re_dfastate_t **)
+ calloc (sizeof (re_dfastate_t *), SBC_MAX);
+ if (BE (state->trtable == NULL, 0))
+ return false;
+ return true;
+ }
+ return false;
+ }
+
+ err = re_node_set_alloc (&follows, ndests + 1);
+ if (BE (err != REG_NOERROR, 0))
+ goto out_free;
+
+ /* Avoid arithmetic overflow in size calculation. */
+ if (BE ((((SIZE_MAX - (sizeof (re_node_set) + sizeof (bitset_t)) * SBC_MAX)
+ / (3 * sizeof (re_dfastate_t *)))
+ < ndests),
+ 0))
+ goto out_free;
+
+ if (__libc_use_alloca ((sizeof (re_node_set) + sizeof (bitset_t)) * SBC_MAX
+ + ndests * 3 * sizeof (re_dfastate_t *)))
+ dest_states = (re_dfastate_t **)
+ alloca (ndests * 3 * sizeof (re_dfastate_t *));
+ else
+ {
+ dest_states = re_malloc (re_dfastate_t *, ndests * 3);
+ if (BE (dest_states == NULL, 0))
+ {
+out_free:
+ if (dest_states_malloced)
+ re_free (dest_states);
+ re_node_set_free (&follows);
+ for (i = 0; i < ndests; ++i)
+ re_node_set_free (dests_node + i);
+ if (dests_node_malloced)
+ re_free (dests_alloc);
+ return false;
+ }
+ dest_states_malloced = true;
+ }
+ dest_states_word = dest_states + ndests;
+ dest_states_nl = dest_states_word + ndests;
+ bitset_empty (acceptable);
+
+ /* Then build the states for all destinations. */
+ for (i = 0; i < ndests; ++i)
+ {
+ Idx next_node;
+ re_node_set_empty (&follows);
+ /* Merge the follows of this destination states. */
+ for (j = 0; j < dests_node[i].nelem; ++j)
+ {
+ next_node = dfa->nexts[dests_node[i].elems[j]];
+ if (next_node != -1)
+ {
+ err = re_node_set_merge (&follows, dfa->eclosures + next_node);
+ if (BE (err != REG_NOERROR, 0))
+ goto out_free;
+ }
+ }
+ dest_states[i] = re_acquire_state_context (&err, dfa, &follows, 0);
+ if (BE (dest_states[i] == NULL && err != REG_NOERROR, 0))
+ goto out_free;
+ /* If the new state has context constraint,
+ build appropriate states for these contexts. */
+ if (dest_states[i]->has_constraint)
+ {
+ dest_states_word[i] = re_acquire_state_context (&err, dfa, &follows,
+ CONTEXT_WORD);
+ if (BE (dest_states_word[i] == NULL && err != REG_NOERROR, 0))
+ goto out_free;
+
+ if (dest_states[i] != dest_states_word[i] && dfa->mb_cur_max > 1)
+ need_word_trtable = true;
+
+ dest_states_nl[i] = re_acquire_state_context (&err, dfa, &follows,
+ CONTEXT_NEWLINE);
+ if (BE (dest_states_nl[i] == NULL && err != REG_NOERROR, 0))
+ goto out_free;
+ }
+ else
+ {
+ dest_states_word[i] = dest_states[i];
+ dest_states_nl[i] = dest_states[i];
+ }
+ bitset_merge (acceptable, dests_ch[i]);
+ }
+
+ if (!BE (need_word_trtable, 0))
+ {
+ /* We don't care about whether the following character is a word
+ character, or we are in a single-byte character set so we can
+ discern by looking at the character code: allocate a
+ 256-entry transition table. */
+ trtable = state->trtable =
+ (re_dfastate_t **) calloc (sizeof (re_dfastate_t *), SBC_MAX);
+ if (BE (trtable == NULL, 0))
+ goto out_free;
+
+ /* For all characters ch...: */
+ for (i = 0; i < BITSET_WORDS; ++i)
+ for (ch = i * BITSET_WORD_BITS, elem = acceptable[i], mask = 1;
+ elem;
+ mask <<= 1, elem >>= 1, ++ch)
+ if (BE (elem & 1, 0))
+ {
+ /* There must be exactly one destination which accepts
+ character ch. See group_nodes_into_DFAstates. */
+ for (j = 0; (dests_ch[j][i] & mask) == 0; ++j)
+ ;
+
+ /* j-th destination accepts the word character ch. */
+ if (dfa->word_char[i] & mask)
+ trtable[ch] = dest_states_word[j];
+ else
+ trtable[ch] = dest_states[j];
+ }
+ }
+ else
+ {
+ /* We care about whether the following character is a word
+ character, and we are in a multi-byte character set: discern
+ by looking at the character code: build two 256-entry
+ transition tables, one starting at trtable[0] and one
+ starting at trtable[SBC_MAX]. */
+ trtable = state->word_trtable =
+ (re_dfastate_t **) calloc (sizeof (re_dfastate_t *), 2 * SBC_MAX);
+ if (BE (trtable == NULL, 0))
+ goto out_free;
+
+ /* For all characters ch...: */
+ for (i = 0; i < BITSET_WORDS; ++i)
+ for (ch = i * BITSET_WORD_BITS, elem = acceptable[i], mask = 1;
+ elem;
+ mask <<= 1, elem >>= 1, ++ch)
+ if (BE (elem & 1, 0))
+ {
+ /* There must be exactly one destination which accepts
+ character ch. See group_nodes_into_DFAstates. */
+ for (j = 0; (dests_ch[j][i] & mask) == 0; ++j)
+ ;
+
+ /* j-th destination accepts the word character ch. */
+ trtable[ch] = dest_states[j];
+ trtable[ch + SBC_MAX] = dest_states_word[j];
+ }
+ }
+
+ /* new line */
+ if (bitset_contain (acceptable, NEWLINE_CHAR))
+ {
+ /* The current state accepts newline character. */
+ for (j = 0; j < ndests; ++j)
+ if (bitset_contain (dests_ch[j], NEWLINE_CHAR))
+ {
+ /* k-th destination accepts newline character. */
+ trtable[NEWLINE_CHAR] = dest_states_nl[j];
+ if (need_word_trtable)
+ trtable[NEWLINE_CHAR + SBC_MAX] = dest_states_nl[j];
+ /* There must be only one destination which accepts
+ newline. See group_nodes_into_DFAstates. */
+ break;
+ }
+ }
+
+ if (dest_states_malloced)
+ re_free (dest_states);
+
+ re_node_set_free (&follows);
+ for (i = 0; i < ndests; ++i)
+ re_node_set_free (dests_node + i);
+
+ if (dests_node_malloced)
+ re_free (dests_alloc);
+
+ return true;
+}
+
+/* Group all nodes belonging to STATE into several destinations.
+ Then for all destinations, set the nodes belonging to the destination
+ to DESTS_NODE[i] and set the characters accepted by the destination
+ to DEST_CH[i]. This function return the number of destinations. */
+
+static Idx
+group_nodes_into_DFAstates (const re_dfa_t *dfa, const re_dfastate_t *state,
+ re_node_set *dests_node, bitset_t *dests_ch)
+{
+ reg_errcode_t err;
+ bool ok;
+ Idx i, j, k;
+ Idx ndests; /* Number of the destinations from 'state'. */
+ bitset_t accepts; /* Characters a node can accept. */
+ const re_node_set *cur_nodes = &state->nodes;
+ bitset_empty (accepts);
+ ndests = 0;
+
+ /* For all the nodes belonging to 'state', */
+ for (i = 0; i < cur_nodes->nelem; ++i)
+ {
+ re_token_t *node = &dfa->nodes[cur_nodes->elems[i]];
+ re_token_type_t type = node->type;
+ unsigned int constraint = node->constraint;
+
+ /* Enumerate all single byte character this node can accept. */
+ if (type == CHARACTER)
+ bitset_set (accepts, node->opr.c);
+ else if (type == SIMPLE_BRACKET)
+ {
+ bitset_merge (accepts, node->opr.sbcset);
+ }
+ else if (type == OP_PERIOD)
+ {
+#ifdef RE_ENABLE_I18N
+ if (dfa->mb_cur_max > 1)
+ bitset_merge (accepts, dfa->sb_char);
+ else
+#endif
+ bitset_set_all (accepts);
+ if (!(dfa->syntax & RE_DOT_NEWLINE))
+ bitset_clear (accepts, '\n');
+ if (dfa->syntax & RE_DOT_NOT_NULL)
+ bitset_clear (accepts, '\0');
+ }
+#ifdef RE_ENABLE_I18N
+ else if (type == OP_UTF8_PERIOD)
+ {
+ if (ASCII_CHARS % BITSET_WORD_BITS == 0)
+ memset (accepts, -1, ASCII_CHARS / CHAR_BIT);
+ else
+ bitset_merge (accepts, utf8_sb_map);
+ if (!(dfa->syntax & RE_DOT_NEWLINE))
+ bitset_clear (accepts, '\n');
+ if (dfa->syntax & RE_DOT_NOT_NULL)
+ bitset_clear (accepts, '\0');
+ }
+#endif
+ else
+ continue;
+
+ /* Check the 'accepts' and sift the characters which are not
+ match it the context. */
+ if (constraint)
+ {
+ if (constraint & NEXT_NEWLINE_CONSTRAINT)
+ {
+ bool accepts_newline = bitset_contain (accepts, NEWLINE_CHAR);
+ bitset_empty (accepts);
+ if (accepts_newline)
+ bitset_set (accepts, NEWLINE_CHAR);
+ else
+ continue;
+ }
+ if (constraint & NEXT_ENDBUF_CONSTRAINT)
+ {
+ bitset_empty (accepts);
+ continue;
+ }
+
+ if (constraint & NEXT_WORD_CONSTRAINT)
+ {
+ bitset_word_t any_set = 0;
+ if (type == CHARACTER && !node->word_char)
+ {
+ bitset_empty (accepts);
+ continue;
+ }
+#ifdef RE_ENABLE_I18N
+ if (dfa->mb_cur_max > 1)
+ for (j = 0; j < BITSET_WORDS; ++j)
+ any_set |= (accepts[j] &= (dfa->word_char[j] |
~dfa->sb_char[j]));
+ else
+#endif
+ for (j = 0; j < BITSET_WORDS; ++j)
+ any_set |= (accepts[j] &= dfa->word_char[j]);
+ if (!any_set)
+ continue;
+ }
+ if (constraint & NEXT_NOTWORD_CONSTRAINT)
+ {
+ bitset_word_t any_set = 0;
+ if (type == CHARACTER && node->word_char)
+ {
+ bitset_empty (accepts);
+ continue;
+ }
+#ifdef RE_ENABLE_I18N
+ if (dfa->mb_cur_max > 1)
+ for (j = 0; j < BITSET_WORDS; ++j)
+ any_set |= (accepts[j] &= ~(dfa->word_char[j] &
dfa->sb_char[j]));
+ else
+#endif
+ for (j = 0; j < BITSET_WORDS; ++j)
+ any_set |= (accepts[j] &= ~dfa->word_char[j]);
+ if (!any_set)
+ continue;
+ }
+ }
+
+ /* Then divide 'accepts' into DFA states, or create a new
+ state. Above, we make sure that accepts is not empty. */
+ for (j = 0; j < ndests; ++j)
+ {
+ bitset_t intersec; /* Intersection sets, see below. */
+ bitset_t remains;
+ /* Flags, see below. */
+ bitset_word_t has_intersec, not_subset, not_consumed;
+
+ /* Optimization, skip if this state doesn't accept the character. */
+ if (type == CHARACTER && !bitset_contain (dests_ch[j], node->opr.c))
+ continue;
+
+ /* Enumerate the intersection set of this state and 'accepts'. */
+ has_intersec = 0;
+ for (k = 0; k < BITSET_WORDS; ++k)
+ has_intersec |= intersec[k] = accepts[k] & dests_ch[j][k];
+ /* And skip if the intersection set is empty. */
+ if (!has_intersec)
+ continue;
+
+ /* Then check if this state is a subset of 'accepts'. */
+ not_subset = not_consumed = 0;
+ for (k = 0; k < BITSET_WORDS; ++k)
+ {
+ not_subset |= remains[k] = ~accepts[k] & dests_ch[j][k];
+ not_consumed |= accepts[k] = accepts[k] & ~dests_ch[j][k];
+ }
+
+ /* If this state isn't a subset of 'accepts', create a
+ new group state, which has the 'remains'. */
+ if (not_subset)
+ {
+ bitset_copy (dests_ch[ndests], remains);
+ bitset_copy (dests_ch[j], intersec);
+ err = re_node_set_init_copy (dests_node + ndests, &dests_node[j]);
+ if (BE (err != REG_NOERROR, 0))
+ goto error_return;
+ ++ndests;
+ }
+
+ /* Put the position in the current group. */
+ ok = re_node_set_insert (&dests_node[j], cur_nodes->elems[i]);
+ if (BE (! ok, 0))
+ goto error_return;
+
+ /* If all characters are consumed, go to next node. */
+ if (!not_consumed)
+ break;
+ }
+ /* Some characters remain, create a new group. */
+ if (j == ndests)
+ {
+ bitset_copy (dests_ch[ndests], accepts);
+ err = re_node_set_init_1 (dests_node + ndests, cur_nodes->elems[i]);
+ if (BE (err != REG_NOERROR, 0))
+ goto error_return;
+ ++ndests;
+ bitset_empty (accepts);
+ }
+ }
+ return ndests;
+ error_return:
+ for (j = 0; j < ndests; ++j)
+ re_node_set_free (dests_node + j);
+ return -1;
+}
+
+#ifdef RE_ENABLE_I18N
+/* Check how many bytes the node 'dfa->nodes[node_idx]' accepts.
+ Return the number of the bytes the node accepts.
+ STR_IDX is the current index of the input string.
+
+ This function handles the nodes which can accept one character, or
+ one collating element like '.', '[a-z]', opposite to the other nodes
+ can only accept one byte. */
+
+# ifdef _LIBC
+# include <locale/weight.h>
+# endif
+
+static int
+check_node_accept_bytes (const re_dfa_t *dfa, Idx node_idx,
+ const re_string_t *input, Idx str_idx)
+{
+ const re_token_t *node = dfa->nodes + node_idx;
+ int char_len, elem_len;
+ Idx i;
+
+ if (BE (node->type == OP_UTF8_PERIOD, 0))
+ {
+ unsigned char c = re_string_byte_at (input, str_idx), d;
+ if (BE (c < 0xc2, 1))
+ return 0;
+
+ if (str_idx + 2 > input->len)
+ return 0;
+
+ d = re_string_byte_at (input, str_idx + 1);
+ if (c < 0xe0)
+ return (d < 0x80 || d > 0xbf) ? 0 : 2;
+ else if (c < 0xf0)
+ {
+ char_len = 3;
+ if (c == 0xe0 && d < 0xa0)
+ return 0;
+ }
+ else if (c < 0xf8)
+ {
+ char_len = 4;
+ if (c == 0xf0 && d < 0x90)
+ return 0;
+ }
+ else if (c < 0xfc)
+ {
+ char_len = 5;
+ if (c == 0xf8 && d < 0x88)
+ return 0;
+ }
+ else if (c < 0xfe)
+ {
+ char_len = 6;
+ if (c == 0xfc && d < 0x84)
+ return 0;
+ }
+ else
+ return 0;
+
+ if (str_idx + char_len > input->len)
+ return 0;
+
+ for (i = 1; i < char_len; ++i)
+ {
+ d = re_string_byte_at (input, str_idx + i);
+ if (d < 0x80 || d > 0xbf)
+ return 0;
+ }
+ return char_len;
+ }
+
+ char_len = re_string_char_size_at (input, str_idx);
+ if (node->type == OP_PERIOD)
+ {
+ if (char_len <= 1)
+ return 0;
+ /* FIXME: I don't think this if is needed, as both '\n'
+ and '\0' are char_len == 1. */
+ /* '.' accepts any one character except the following two cases. */
+ if ((!(dfa->syntax & RE_DOT_NEWLINE) &&
+ re_string_byte_at (input, str_idx) == '\n') ||
+ ((dfa->syntax & RE_DOT_NOT_NULL) &&
+ re_string_byte_at (input, str_idx) == '\0'))
+ return 0;
+ return char_len;
+ }
+
+ elem_len = re_string_elem_size_at (input, str_idx);
+ if ((elem_len <= 1 && char_len <= 1) || char_len == 0)
+ return 0;
+
+ if (node->type == COMPLEX_BRACKET)
+ {
+ const re_charset_t *cset = node->opr.mbcset;
+# ifdef _LIBC
+ const unsigned char *pin
+ = ((const unsigned char *) re_string_get_buffer (input) + str_idx);
+ Idx j;
+ uint32_t nrules;
+# endif /* _LIBC */
+ int match_len = 0;
+ wchar_t wc = ((cset->nranges || cset->nchar_classes || cset->nmbchars)
+ ? re_string_wchar_at (input, str_idx) : 0);
+
+ /* match with multibyte character? */
+ for (i = 0; i < cset->nmbchars; ++i)
+ if (wc == cset->mbchars[i])
+ {
+ match_len = char_len;
+ goto check_node_accept_bytes_match;
+ }
+ /* match with character_class? */
+ for (i = 0; i < cset->nchar_classes; ++i)
+ {
+ wctype_t wt = cset->char_classes[i];
+ if (__iswctype (wc, wt))
+ {
+ match_len = char_len;
+ goto check_node_accept_bytes_match;
+ }
+ }
+
+# ifdef _LIBC
+ nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
+ if (nrules != 0)
+ {
+ unsigned int in_collseq = 0;
+ const int32_t *table, *indirect;
+ const unsigned char *weights, *extra;
+ const char *collseqwc;
+
+ /* match with collating_symbol? */
+ if (cset->ncoll_syms)
+ extra = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_SYMB_EXTRAMB);
+ for (i = 0; i < cset->ncoll_syms; ++i)
+ {
+ const unsigned char *coll_sym = extra + cset->coll_syms[i];
+ /* Compare the length of input collating element and
+ the length of current collating element. */
+ if (*coll_sym != elem_len)
+ continue;
+ /* Compare each bytes. */
+ for (j = 0; j < *coll_sym; j++)
+ if (pin[j] != coll_sym[1 + j])
+ break;
+ if (j == *coll_sym)
+ {
+ /* Match if every bytes is equal. */
+ match_len = j;
+ goto check_node_accept_bytes_match;
+ }
+ }
+
+ if (cset->nranges)
+ {
+ if (elem_len <= char_len)
+ {
+ collseqwc = _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQWC);
+ in_collseq = __collseq_table_lookup (collseqwc, wc);
+ }
+ else
+ in_collseq = find_collation_sequence_value (pin, elem_len);
+ }
+ /* match with range expression? */
+ /* FIXME: Implement rational ranges here, too. */
+ for (i = 0; i < cset->nranges; ++i)
+ if (cset->range_starts[i] <= in_collseq
+ && in_collseq <= cset->range_ends[i])
+ {
+ match_len = elem_len;
+ goto check_node_accept_bytes_match;
+ }
+
+ /* match with equivalence_class? */
+ if (cset->nequiv_classes)
+ {
+ const unsigned char *cp = pin;
+ table = (const int32_t *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB);
+ weights = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_WEIGHTMB);
+ extra = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAMB);
+ indirect = (const int32_t *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_INDIRECTMB);
+ int32_t idx = findidx (table, indirect, extra, &cp, elem_len);
+ int32_t rule = idx >> 24;
+ idx &= 0xffffff;
+ if (idx > 0)
+ {
+ size_t weight_len = weights[idx];
+ for (i = 0; i < cset->nequiv_classes; ++i)
+ {
+ int32_t equiv_class_idx = cset->equiv_classes[i];
+ int32_t equiv_class_rule = equiv_class_idx >> 24;
+ equiv_class_idx &= 0xffffff;
+ if (weights[equiv_class_idx] == weight_len
+ && equiv_class_rule == rule
+ && memcmp (weights + idx + 1,
+ weights + equiv_class_idx + 1,
+ weight_len) == 0)
+ {
+ match_len = elem_len;
+ goto check_node_accept_bytes_match;
+ }
+ }
+ }
+ }
+ }
+ else
+# endif /* _LIBC */
+ {
+ /* match with range expression? */
+ for (i = 0; i < cset->nranges; ++i)
+ {
+ if (cset->range_starts[i] <= wc && wc <= cset->range_ends[i])
+ {
+ match_len = char_len;
+ goto check_node_accept_bytes_match;
+ }
+ }
+ }
+ check_node_accept_bytes_match:
+ if (!cset->non_match)
+ return match_len;
+ else
+ {
+ if (match_len > 0)
+ return 0;
+ else
+ return (elem_len > char_len) ? elem_len : char_len;
+ }
+ }
+ return 0;
+}
+
+# ifdef _LIBC
+static unsigned int
+find_collation_sequence_value (const unsigned char *mbs, size_t mbs_len)
+{
+ uint32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
+ if (nrules == 0)
+ {
+ if (mbs_len == 1)
+ {
+ /* No valid character. Match it as a single byte character. */
+ const unsigned char *collseq = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQMB);
+ return collseq[mbs[0]];
+ }
+ return UINT_MAX;
+ }
+ else
+ {
+ int32_t idx;
+ const unsigned char *extra = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_SYMB_EXTRAMB);
+ int32_t extrasize = (const unsigned char *)
+ _NL_CURRENT (LC_COLLATE, _NL_COLLATE_SYMB_EXTRAMB + 1) - extra;
+
+ for (idx = 0; idx < extrasize;)
+ {
+ int mbs_cnt;
+ bool found = false;
+ int32_t elem_mbs_len;
+ /* Skip the name of collating element name. */
+ idx = idx + extra[idx] + 1;
+ elem_mbs_len = extra[idx++];
+ if (mbs_len == elem_mbs_len)
+ {
+ for (mbs_cnt = 0; mbs_cnt < elem_mbs_len; ++mbs_cnt)
+ if (extra[idx + mbs_cnt] != mbs[mbs_cnt])
+ break;
+ if (mbs_cnt == elem_mbs_len)
+ /* Found the entry. */
+ found = true;
+ }
+ /* Skip the byte sequence of the collating element. */
+ idx += elem_mbs_len;
+ /* Adjust for the alignment. */
+ idx = (idx + 3) & ~3;
+ /* Skip the collation sequence value. */
+ idx += sizeof (uint32_t);
+ /* Skip the wide char sequence of the collating element. */
+ idx = idx + sizeof (uint32_t) * (*(int32_t *) (extra + idx) + 1);
+ /* If we found the entry, return the sequence value. */
+ if (found)
+ return *(uint32_t *) (extra + idx);
+ /* Skip the collation sequence value. */
+ idx += sizeof (uint32_t);
+ }
+ return UINT_MAX;
+ }
+}
+# endif /* _LIBC */
+#endif /* RE_ENABLE_I18N */
+
+/* Check whether the node accepts the byte which is IDX-th
+ byte of the INPUT. */
+
+static bool
+check_node_accept (const re_match_context_t *mctx, const re_token_t *node,
+ Idx idx)
+{
+ unsigned char ch;
+ ch = re_string_byte_at (&mctx->input, idx);
+ switch (node->type)
+ {
+ case CHARACTER:
+ if (node->opr.c != ch)
+ return false;
+ break;
+
+ case SIMPLE_BRACKET:
+ if (!bitset_contain (node->opr.sbcset, ch))
+ return false;
+ break;
+
+#ifdef RE_ENABLE_I18N
+ case OP_UTF8_PERIOD:
+ if (ch >= ASCII_CHARS)
+ return false;
+ FALLTHROUGH;
+#endif
+ case OP_PERIOD:
+ if ((ch == '\n' && !(mctx->dfa->syntax & RE_DOT_NEWLINE))
+ || (ch == '\0' && (mctx->dfa->syntax & RE_DOT_NOT_NULL)))
+ return false;
+ break;
+
+ default:
+ return false;
+ }
+
+ if (node->constraint)
+ {
+ /* The node has constraints. Check whether the current context
+ satisfies the constraints. */
+ unsigned int context = re_string_context_at (&mctx->input, idx,
+ mctx->eflags);
+ if (NOT_SATISFY_NEXT_CONSTRAINT (node->constraint, context))
+ return false;
+ }
+
+ return true;
+}
+
+/* Extend the buffers, if the buffers have run out. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+extend_buffers (re_match_context_t *mctx, int min_len)
+{
+ reg_errcode_t ret;
+ re_string_t *pstr = &mctx->input;
+
+ /* Avoid overflow. */
+ if (BE (MIN (IDX_MAX, SIZE_MAX / sizeof (re_dfastate_t *)) / 2
+ <= pstr->bufs_len, 0))
+ return REG_ESPACE;
+
+ /* Double the lengths of the buffers, but allocate at least MIN_LEN. */
+ ret = re_string_realloc_buffers (pstr,
+ MAX (min_len,
+ MIN (pstr->len, pstr->bufs_len * 2)));
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+
+ if (mctx->state_log != NULL)
+ {
+ /* And double the length of state_log. */
+ /* XXX We have no indication of the size of this buffer. If this
+ allocation fail we have no indication that the state_log array
+ does not have the right size. */
+ re_dfastate_t **new_array = re_realloc (mctx->state_log, re_dfastate_t *,
+ pstr->bufs_len + 1);
+ if (BE (new_array == NULL, 0))
+ return REG_ESPACE;
+ mctx->state_log = new_array;
+ }
+
+ /* Then reconstruct the buffers. */
+ if (pstr->icase)
+ {
+#ifdef RE_ENABLE_I18N
+ if (pstr->mb_cur_max > 1)
+ {
+ ret = build_wcs_upper_buffer (pstr);
+ if (BE (ret != REG_NOERROR, 0))
+ return ret;
+ }
+ else
+#endif /* RE_ENABLE_I18N */
+ build_upper_buffer (pstr);
+ }
+ else
+ {
+#ifdef RE_ENABLE_I18N
+ if (pstr->mb_cur_max > 1)
+ build_wcs_buffer (pstr);
+ else
+#endif /* RE_ENABLE_I18N */
+ {
+ if (pstr->trans != NULL)
+ re_string_translate_buffer (pstr);
+ }
+ }
+ return REG_NOERROR;
+}
+
+
+/* Functions for matching context. */
+
+/* Initialize MCTX. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+match_ctx_init (re_match_context_t *mctx, int eflags, Idx n)
+{
+ mctx->eflags = eflags;
+ mctx->match_last = -1;
+ if (n > 0)
+ {
+ /* Avoid overflow. */
+ size_t max_object_size =
+ MAX (sizeof (struct re_backref_cache_entry),
+ sizeof (re_sub_match_top_t *));
+ if (BE (MIN (IDX_MAX, SIZE_MAX / max_object_size) < n, 0))
+ return REG_ESPACE;
+
+ mctx->bkref_ents = re_malloc (struct re_backref_cache_entry, n);
+ mctx->sub_tops = re_malloc (re_sub_match_top_t *, n);
+ if (BE (mctx->bkref_ents == NULL || mctx->sub_tops == NULL, 0))
+ return REG_ESPACE;
+ }
+ /* Already zero-ed by the caller.
+ else
+ mctx->bkref_ents = NULL;
+ mctx->nbkref_ents = 0;
+ mctx->nsub_tops = 0; */
+ mctx->abkref_ents = n;
+ mctx->max_mb_elem_len = 1;
+ mctx->asub_tops = n;
+ return REG_NOERROR;
+}
+
+/* Clean the entries which depend on the current input in MCTX.
+ This function must be invoked when the matcher changes the start index
+ of the input, or changes the input string. */
+
+static void
+match_ctx_clean (re_match_context_t *mctx)
+{
+ Idx st_idx;
+ for (st_idx = 0; st_idx < mctx->nsub_tops; ++st_idx)
+ {
+ Idx sl_idx;
+ re_sub_match_top_t *top = mctx->sub_tops[st_idx];
+ for (sl_idx = 0; sl_idx < top->nlasts; ++sl_idx)
+ {
+ re_sub_match_last_t *last = top->lasts[sl_idx];
+ re_free (last->path.array);
+ re_free (last);
+ }
+ re_free (top->lasts);
+ if (top->path)
+ {
+ re_free (top->path->array);
+ re_free (top->path);
+ }
+ re_free (top);
+ }
+
+ mctx->nsub_tops = 0;
+ mctx->nbkref_ents = 0;
+}
+
+/* Free all the memory associated with MCTX. */
+
+static void
+match_ctx_free (re_match_context_t *mctx)
+{
+ /* First, free all the memory associated with MCTX->SUB_TOPS. */
+ match_ctx_clean (mctx);
+ re_free (mctx->sub_tops);
+ re_free (mctx->bkref_ents);
+}
+
+/* Add a new backreference entry to MCTX.
+ Note that we assume that caller never call this function with duplicate
+ entry, and call with STR_IDX which isn't smaller than any existing entry.
+*/
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+match_ctx_add_entry (re_match_context_t *mctx, Idx node, Idx str_idx, Idx from,
+ Idx to)
+{
+ if (mctx->nbkref_ents >= mctx->abkref_ents)
+ {
+ struct re_backref_cache_entry* new_entry;
+ new_entry = re_realloc (mctx->bkref_ents, struct re_backref_cache_entry,
+ mctx->abkref_ents * 2);
+ if (BE (new_entry == NULL, 0))
+ {
+ re_free (mctx->bkref_ents);
+ return REG_ESPACE;
+ }
+ mctx->bkref_ents = new_entry;
+ memset (mctx->bkref_ents + mctx->nbkref_ents, '\0',
+ sizeof (struct re_backref_cache_entry) * mctx->abkref_ents);
+ mctx->abkref_ents *= 2;
+ }
+ if (mctx->nbkref_ents > 0
+ && mctx->bkref_ents[mctx->nbkref_ents - 1].str_idx == str_idx)
+ mctx->bkref_ents[mctx->nbkref_ents - 1].more = 1;
+
+ mctx->bkref_ents[mctx->nbkref_ents].node = node;
+ mctx->bkref_ents[mctx->nbkref_ents].str_idx = str_idx;
+ mctx->bkref_ents[mctx->nbkref_ents].subexp_from = from;
+ mctx->bkref_ents[mctx->nbkref_ents].subexp_to = to;
+
+ /* This is a cache that saves negative results of check_dst_limits_calc_pos.
+ If bit N is clear, means that this entry won't epsilon-transition to
+ an OP_OPEN_SUBEXP or OP_CLOSE_SUBEXP for the N+1-th subexpression. If
+ it is set, check_dst_limits_calc_pos_1 will recurse and try to find one
+ such node.
+
+ A backreference does not epsilon-transition unless it is empty, so set
+ to all zeros if FROM != TO. */
+ mctx->bkref_ents[mctx->nbkref_ents].eps_reachable_subexps_map
+ = (from == to ? -1 : 0);
+
+ mctx->bkref_ents[mctx->nbkref_ents++].more = 0;
+ if (mctx->max_mb_elem_len < to - from)
+ mctx->max_mb_elem_len = to - from;
+ return REG_NOERROR;
+}
+
+/* Return the first entry with the same str_idx, or -1 if none is
+ found. Note that MCTX->BKREF_ENTS is already sorted by MCTX->STR_IDX. */
+
+static Idx
+search_cur_bkref_entry (const re_match_context_t *mctx, Idx str_idx)
+{
+ Idx left, right, mid, last;
+ last = right = mctx->nbkref_ents;
+ for (left = 0; left < right;)
+ {
+ mid = (left + right) / 2;
+ if (mctx->bkref_ents[mid].str_idx < str_idx)
+ left = mid + 1;
+ else
+ right = mid;
+ }
+ if (left < last && mctx->bkref_ents[left].str_idx == str_idx)
+ return left;
+ else
+ return -1;
+}
+
+/* Register the node NODE, whose type is OP_OPEN_SUBEXP, and which matches
+ at STR_IDX. */
+
+static reg_errcode_t
+__attribute_warn_unused_result__
+match_ctx_add_subtop (re_match_context_t *mctx, Idx node, Idx str_idx)
+{
+#ifdef DEBUG
+ assert (mctx->sub_tops != NULL);
+ assert (mctx->asub_tops > 0);
+#endif
+ if (BE (mctx->nsub_tops == mctx->asub_tops, 0))
+ {
+ Idx new_asub_tops = mctx->asub_tops * 2;
+ re_sub_match_top_t **new_array = re_realloc (mctx->sub_tops,
+ re_sub_match_top_t *,
+ new_asub_tops);
+ if (BE (new_array == NULL, 0))
+ return REG_ESPACE;
+ mctx->sub_tops = new_array;
+ mctx->asub_tops = new_asub_tops;
+ }
+ mctx->sub_tops[mctx->nsub_tops] = calloc (1, sizeof (re_sub_match_top_t));
+ if (BE (mctx->sub_tops[mctx->nsub_tops] == NULL, 0))
+ return REG_ESPACE;
+ mctx->sub_tops[mctx->nsub_tops]->node = node;
+ mctx->sub_tops[mctx->nsub_tops++]->str_idx = str_idx;
+ return REG_NOERROR;
+}
+
+/* Register the node NODE, whose type is OP_CLOSE_SUBEXP, and which matches
+ at STR_IDX, whose corresponding OP_OPEN_SUBEXP is SUB_TOP. */
+
+static re_sub_match_last_t *
+match_ctx_add_sublast (re_sub_match_top_t *subtop, Idx node, Idx str_idx)
+{
+ re_sub_match_last_t *new_entry;
+ if (BE (subtop->nlasts == subtop->alasts, 0))
+ {
+ Idx new_alasts = 2 * subtop->alasts + 1;
+ re_sub_match_last_t **new_array = re_realloc (subtop->lasts,
+ re_sub_match_last_t *,
+ new_alasts);
+ if (BE (new_array == NULL, 0))
+ return NULL;
+ subtop->lasts = new_array;
+ subtop->alasts = new_alasts;
+ }
+ new_entry = calloc (1, sizeof (re_sub_match_last_t));
+ if (BE (new_entry != NULL, 1))
+ {
+ subtop->lasts[subtop->nlasts] = new_entry;
+ new_entry->node = node;
+ new_entry->str_idx = str_idx;
+ ++subtop->nlasts;
+ }
+ return new_entry;
+}
+
+static void
+sift_ctx_init (re_sift_context_t *sctx, re_dfastate_t **sifted_sts,
+ re_dfastate_t **limited_sts, Idx last_node, Idx last_str_idx)
+{
+ sctx->sifted_states = sifted_sts;
+ sctx->limited_states = limited_sts;
+ sctx->last_node = last_node;
+ sctx->last_str_idx = last_str_idx;
+ re_node_set_init_empty (&sctx->limits);
+}
diff --git a/lib/unistd.in.h b/lib/unistd.in.h
index b6a348f..55bbb6c 100644
--- a/lib/unistd.in.h
+++ b/lib/unistd.in.h
@@ -432,12 +432,12 @@ extern char **environ;
#elif defined GNULIB_POSIXCHECK
# if HAVE_RAW_DECL_ENVIRON
_GL_UNISTD_INLINE char ***
+_GL_WARN_ON_USE_ATTRIBUTE ("environ is unportable - "
+ "use gnulib module environ for portability")
rpl_environ (void)
{
return &environ;
}
-_GL_WARN_ON_USE (rpl_environ, "environ is unportable - "
- "use gnulib module environ for portability");
# undef environ
# define environ (*rpl_environ ())
# endif
diff --git a/lib/warn-on-use.h b/lib/warn-on-use.h
index e76c384..72d67cc 100644
--- a/lib/warn-on-use.h
+++ b/lib/warn-on-use.h
@@ -20,23 +20,32 @@
supported by the compiler. If the compiler does not support this
feature, the macro expands to an unused extern declaration.
- This macro is useful for marking a function as a potential
+ _GL_WARN_ON_USE_ATTRIBUTE ("literal string") expands to the
+ attribute used in _GL_WARN_ON_USE. If the compiler does not support
+ this feature, it expands to empty.
+
+ These macros are useful for marking a function as a potential
portability trap, with the intent that "literal string" include
instructions on the replacement function that should be used
- instead. However, one of the reasons that a function is a
- portability trap is if it has the wrong signature. Declaring
- FUNCTION with a different signature in C is a compilation error, so
- this macro must use the same type as any existing declaration so
- that programs that avoid the problematic FUNCTION do not fail to
- compile merely because they included a header that poisoned the
- function. But this implies that _GL_WARN_ON_USE is only safe to
- use if FUNCTION is known to already have a declaration. Use of
- this macro implies that there must not be any other macro hiding
- the declaration of FUNCTION; but undefining FUNCTION first is part
- of the poisoning process anyway (although for symbols that are
- provided only via a macro, the result is a compilation error rather
- than a warning containing "literal string"). Also note that in
- C++, it is only safe to use if FUNCTION has no overloads.
+ instead.
+ _GL_WARN_ON_USE is for functions with 'extern' linkage.
+ _GL_WARN_ON_USE_ATTRIBUTE is for functions with 'static' or 'inline'
+ linkage.
+
+ However, one of the reasons that a function is a portability trap is
+ if it has the wrong signature. Declaring FUNCTION with a different
+ signature in C is a compilation error, so this macro must use the
+ same type as any existing declaration so that programs that avoid
+ the problematic FUNCTION do not fail to compile merely because they
+ included a header that poisoned the function. But this implies that
+ _GL_WARN_ON_USE is only safe to use if FUNCTION is known to already
+ have a declaration. Use of this macro implies that there must not
+ be any other macro hiding the declaration of FUNCTION; but
+ undefining FUNCTION first is part of the poisoning process anyway
+ (although for symbols that are provided only via a macro, the result
+ is a compilation error rather than a warning containing
+ "literal string"). Also note that in C++, it is only safe to use if
+ FUNCTION has no overloads.
For an example, it is possible to poison 'getline' by:
- adding a call to gl_WARN_ON_USE_PREPARE([[#include <stdio.h>]],
@@ -54,12 +63,21 @@
(less common usage, like &environ, will cause a compilation error
rather than issue the nice warning, but the end result of informing
the developer about their portability problem is still achieved):
- #if HAVE_RAW_DECL_ENVIRON
- static char ***rpl_environ (void) { return &environ; }
- _GL_WARN_ON_USE (rpl_environ, "environ is not always properly declared");
- # undef environ
- # define environ (*rpl_environ ())
- #endif
+ #if HAVE_RAW_DECL_ENVIRON
+ static char ***
+ rpl_environ (void) { return &environ; }
+ _GL_WARN_ON_USE (rpl_environ, "environ is not always properly declared");
+ # undef environ
+ # define environ (*rpl_environ ())
+ #endif
+ or better (avoiding contradictory use of 'static' and 'extern'):
+ #if HAVE_RAW_DECL_ENVIRON
+ static char ***
+ _GL_WARN_ON_USE_ATTRIBUTE ("environ is not always properly declared")
+ rpl_environ (void) { return &environ; }
+ # undef environ
+ # define environ (*rpl_environ ())
+ #endif
*/
#ifndef _GL_WARN_ON_USE
@@ -67,13 +85,17 @@
/* A compiler attribute is available in gcc versions 4.3.0 and later. */
# define _GL_WARN_ON_USE(function, message) \
extern __typeof__ (function) function __attribute__ ((__warning__ (message)))
+# define _GL_WARN_ON_USE_ATTRIBUTE(message) \
+ __attribute__ ((__warning__ (message)))
# elif __GNUC__ >= 3 && GNULIB_STRICT_CHECKING
/* Verify the existence of the function. */
# define _GL_WARN_ON_USE(function, message) \
extern __typeof__ (function) function
+# define _GL_WARN_ON_USE_ATTRIBUTE(message)
# else /* Unsupported. */
# define _GL_WARN_ON_USE(function, message) \
_GL_WARN_EXTERN_C int _gl_warn_on_use
+# define _GL_WARN_ON_USE_ATTRIBUTE(message)
# endif
#endif
diff --git a/lisp/auth-source.el b/lisp/auth-source.el
index abff0de..261e972 100644
--- a/lisp/auth-source.el
+++ b/lisp/auth-source.el
@@ -779,7 +779,7 @@ Calls `auth-source-search' with the :delete property in
SPEC set to t.
The backend may not actually delete the entries.
Returns the deleted entries."
- (auth-source-search (plist-put spec :delete t)))
+ (apply #'auth-source-search (plist-put spec :delete t)))
(defun auth-source-search-collection (collection value)
"Returns t is VALUE is t or COLLECTION is t or COLLECTION contains VALUE."
diff --git a/lisp/bookmark.el b/lisp/bookmark.el
index 78f3e32..58a2794 100644
--- a/lisp/bookmark.el
+++ b/lisp/bookmark.el
@@ -1102,7 +1102,7 @@ BOOKMARK is usually a bookmark name (a string). It can
also be a
bookmark record, but this is usually only done by programmatic callers.
If DISPLAY-FUNC is non-nil, it is a function to invoke to display the
-bookmark. It defaults to `switch-to-buffer'. A typical value for
+bookmark. It defaults to `pop-to-buffer-same-window'. A typical value for
DISPLAY-FUNC would be `switch-to-buffer-other-window'."
(interactive
(list (bookmark-completing-read "Jump to bookmark"
@@ -1110,7 +1110,10 @@ DISPLAY-FUNC would be `switch-to-buffer-other-window'."
(unless bookmark
(error "No bookmark specified"))
(bookmark-maybe-historicize-string bookmark)
- (bookmark--jump-via bookmark (or display-func 'switch-to-buffer)))
+ ;; Don't use `switch-to-buffer' because it would let the
+ ;; window-point override the bookmark's point when
+ ;; `switch-to-buffer-preserve-window-point' is non-nil.
+ (bookmark--jump-via bookmark (or display-func 'pop-to-buffer-same-window)))
;;;###autoload
diff --git a/lisp/calendar/todo-mode.el b/lisp/calendar/todo-mode.el
index 5161ae8..c1c2921 100644
--- a/lisp/calendar/todo-mode.el
+++ b/lisp/calendar/todo-mode.el
@@ -853,17 +853,18 @@ category. With non-nil argument BACK, visit the
numerically
previous category (the highest numbered one, if the current
category is the first)."
(interactive)
- (setq todo-category-number
- (1+ (mod (- todo-category-number (if back 2 0))
- (length todo-categories))))
- (when todo-skip-archived-categories
- (while (and (zerop (todo-get-count 'todo))
- (zerop (todo-get-count 'done))
- (not (zerop (todo-get-count 'archived))))
- (setq todo-category-number
- (funcall (if back #'1- #'1+) todo-category-number))))
- (todo-category-select)
- (goto-char (point-min)))
+ (let ((setcatnum (lambda () (1+ (mod (- todo-category-number
+ (if back 2 0))
+ (length todo-categories))))))
+ (setq todo-category-number (funcall setcatnum))
+ (when todo-skip-archived-categories
+ (while (and (zerop (todo-get-count 'todo))
+ (zerop (todo-get-count 'done))
+ (not (zerop (todo-get-count 'archived))))
+ (setq todo-category-number (funcall setcatnum))))
+ (todo-category-select)
+ (if transient-mark-mode (deactivate-mark))
+ (goto-char (point-min))))
(defun todo-backward-category ()
"Visit the numerically previous category in this todo file.
@@ -928,11 +929,13 @@ Categories mode."
(when goto-archive (todo-archive-mode))
(set-window-buffer (selected-window)
(set-buffer (find-buffer-visiting file0)))
+ (if transient-mark-mode (deactivate-mark))
(unless todo-global-current-todo-file
(setq todo-global-current-todo-file todo-current-todo-file))
(todo-category-number category)
(todo-category-select)
(goto-char (point-min))
+ (if (bound-and-true-p hl-line-mode) (hl-line-highlight))
(when add-item (todo-insert-item--basic))))))
(defun todo-next-item (&optional count)
@@ -1018,15 +1021,17 @@ empty line above the done items separator."
(setq shown (progn
(goto-char (point-min))
(re-search-forward todo-done-string-start nil t)))
- (if (not (pos-visible-in-window-p shown))
- (recenter)
- (goto-char opoint)))))))
+ (if (pos-visible-in-window-p shown)
+ (goto-char opoint)
+ (recenter)
+ (if transient-mark-mode (deactivate-mark))))))))
(defun todo-toggle-view-done-only ()
"Switch between displaying only done or only todo items."
(interactive)
(setq todo-show-done-only (not todo-show-done-only))
- (todo-category-select))
+ (todo-category-select)
+ (if transient-mark-mode (deactivate-mark)))
(defun todo-toggle-item-highlighting ()
"Highlight or unhighlight the todo item the cursor is on."
@@ -1860,15 +1865,18 @@ their associated keys and their effects."
(region (eq where 'region))
(here (eq where 'here))
diary-item)
- (when copy
- (cond
- ((not (eq major-mode 'todo-mode))
- (user-error "You must be in Todo mode to copy a todo item"))
- ((todo-done-item-p)
- (user-error "You cannot copy a done item as a new todo item"))
- ((looking-at "^$")
- (user-error "Point must be on a todo item to copy it")))
- (setq diary-item (todo-diary-item-p)))
+ (when (and arg here)
+ (user-error "Here insertion only valid in current category"))
+ (when (and (or copy here)
+ (or (not (eq major-mode 'todo-mode)) (todo-done-item-p)
+ (when copy (looking-at "^$"))
+ (save-excursion
+ (beginning-of-line)
+ ;; Point is on done items separator.
+ (looking-at todo-category-done))))
+ (user-error (concat "Item " (if copy "copying" "insertion")
+ " is not valid here")))
+ (when copy (setq diary-item (todo-diary-item-p)))
(when region
(let (use-empty-active-region)
(unless (and todo-use-only-highlighted-region (use-region-p))
@@ -1876,7 +1884,6 @@ their associated keys and their effects."
(let* ((obuf (current-buffer))
(ocat (todo-current-category))
(opoint (point))
- (todo-mm (eq major-mode 'todo-mode))
(cat+file (cond ((equal arg '(4))
(todo-read-category "Insert in category: "))
((equal arg '(16))
@@ -1894,7 +1901,10 @@ their associated keys and their effects."
(new-item (cond (copy (todo-item-string))
(region (buffer-substring-no-properties
(region-beginning) (region-end)))
- (t (read-from-minibuffer "Todo item: "))))
+ (t (if (eq major-mode 'todo-archive-mode)
+ (user-error (concat "Cannot insert a new
Todo"
+ " item in an archive"))
+ (read-from-minibuffer "Todo item: ")))))
(date-string (cond
((eq date-type 'date)
(todo-read-date))
@@ -1931,7 +1941,6 @@ their associated keys and their effects."
(unless todo-global-current-todo-file
(setq todo-global-current-todo-file todo-current-todo-file))
(let ((buffer-read-only nil)
- (called-from-outside (not (and todo-mm (equal cat ocat))))
done-only item-added)
(unless copy
(setq new-item
@@ -1955,14 +1964,8 @@ their associated keys and their effects."
"\n\t" new-item nil nil
1)))
(unwind-protect
(progn
- ;; Make sure the correct category is selected. There
- ;; are two cases: (i) we just visited the file, so no
- ;; category is selected yet, or (ii) we invoked
- ;; insertion "here" from outside the category we want
- ;; to insert in (with priority insertion, category
- ;; selection is done by todo-set-item-priority).
- (when (or (= (- (point-max) (point-min)) (buffer-size))
- (and here called-from-outside))
+ ;; If we just visited the file, no category is selected yet.
+ (when (= (- (point-max) (point-min)) (buffer-size))
(todo-category-number cat)
(todo-category-select))
;; If only done items are displayed in category,
@@ -1973,16 +1976,7 @@ their associated keys and their effects."
(setq done-only t)
(todo-toggle-view-done-only))
(if here
- (progn
- ;; If command was invoked with point in done
- ;; items section or outside of the current
- ;; category, can't insert "here", so to be
- ;; useful give new item top priority.
- (when (or (todo-done-item-section-p)
- called-from-outside
- done-only)
- (goto-char (point-min)))
- (todo-insert-with-overlays new-item))
+ (todo-insert-with-overlays new-item)
(todo-set-item-priority new-item cat t))
(setq item-added t))
;; If user cancels before setting priority, restore
@@ -2097,7 +2091,14 @@ the item at point."
(setq todo-categories-with-marks
(assq-delete-all cat todo-categories-with-marks)))
(todo-update-categories-sexp)
- (todo-prefix-overlays)))
+ (todo-prefix-overlays)
+ (when (and (zerop (todo-get-count 'diary))
+ (save-excursion
+ (goto-char (point-min))
+ (re-search-forward
+ (concat "^" (regexp-quote todo-category-done))
+ nil t)))
+ (let (todo-show-with-done) (todo-category-select)))))
(if ov (delete-overlay ov)))))
(defvar todo-edit-item--param-key-alist)
@@ -2233,7 +2234,8 @@ made in the number or names of categories."
(insert item))
(kill-buffer)
(unless (eq (current-buffer) buf)
- (set-window-buffer (selected-window) (set-buffer buf))))
+ (set-window-buffer (selected-window) (set-buffer buf)))
+ (if transient-mark-mode (deactivate-mark)))
;; We got here via `F e'.
(when (todo-check-format)
;; FIXME: separate out sexp check?
@@ -2340,7 +2342,7 @@ made in the number or names of categories."
((or (string= omonth "*") (= mm 13))
(user-error "Cannot increment *"))
(t
- (let ((mminc (+ mm inc)))
+ (let ((mminc (+ mm inc (if (< inc 0) 12 0))))
;; Increment or decrement month by INC
;; modulo 12.
(setq mm (% mminc 12))
@@ -2549,7 +2551,11 @@ whose value can be either of the symbols `raise' or
`lower',
meaning to raise or lower the item's priority by one."
(interactive)
(unless (and (or (called-interactively-p 'any) (memq arg '(raise lower)))
- (or (todo-done-item-p) (looking-at "^$")))
+ ;; Noop if point is not on a todo (i.e. not done) item.
+ (or (todo-done-item-p) (looking-at "^$")
+ ;; On done items separator.
+ (save-excursion (beginning-of-line)
+ (looking-at todo-category-done))))
(let* ((item (or item (todo-item-string)))
(marked (todo-marked-item-p))
(cat (or cat (cond ((eq major-mode 'todo-mode)
@@ -2697,9 +2703,13 @@ section in the category moved to."
(interactive "P")
(let* ((cat1 (todo-current-category))
(marked (assoc cat1 todo-categories-with-marks)))
- ;; Noop if point is not on an item and there are no marked items.
- (unless (and (looking-at "^$")
- (not marked))
+ (unless
+ ;; Noop if point is not on an item and there are no marked items.
+ (and (or (looking-at "^$")
+ ;; On done items separator.
+ (save-excursion (beginning-of-line)
+ (looking-at todo-category-done)))
+ (not marked))
(let* ((buffer-read-only)
(file1 todo-current-todo-file)
(item (todo-item-string))
@@ -2856,10 +2866,14 @@ visible."
(let* ((cat (todo-current-category))
(marked (assoc cat todo-categories-with-marks)))
(when marked (todo--user-error-if-marked-done-item))
- (unless (and (not marked)
- (or (todo-done-item-p)
- ;; Point is between todo and done items.
- (looking-at "^$")))
+ (unless
+ ;; Noop if point is not on a todo (i.e. not done) item and
+ ;; there are no marked items.
+ (and (or (todo-done-item-p) (looking-at "^$")
+ ;; On done items separator.
+ (save-excursion (beginning-of-line)
+ (looking-at todo-category-done)))
+ (not marked))
(let* ((date-string (calendar-date-string (calendar-current-date) t t))
(time-string (if todo-always-add-time-string
(concat " " (substring (current-time-string)
@@ -3830,6 +3844,7 @@ face."
(goto-char (point-min))
(while (not (eobp))
(setq match (re-search-forward regex nil t))
+ (if (and match transient-mark-mode) (deactivate-mark))
(goto-char (line-beginning-position))
(unless (or (equal (point) 1)
(looking-at (concat "^" (regexp-quote todo-category-beg))))
@@ -4028,19 +4043,22 @@ regexp items."
(interactive "P")
(todo-filter-items 'regexp arg t))
+(defvar todo--fifiles-history nil
+ "List of short file names used by todo-find-filtered-items-file.")
+
(defun todo-find-filtered-items-file ()
"Choose a filtered items file and visit it."
(interactive)
(let ((files (directory-files todo-directory t "\\.tod[rty]$" t))
falist file)
(dolist (f files)
- (let ((type (cond ((equal (file-name-extension f) "todr") "regexp")
+ (let ((sf-name (todo-short-file-name f))
+ (type (cond ((equal (file-name-extension f) "todr") "regexp")
((equal (file-name-extension f) "todt") "top")
((equal (file-name-extension f) "tody") "diary"))))
- (push (cons (concat (todo-short-file-name f) " (" type ")") f)
- falist)))
- (setq file (completing-read "Choose a filtered items file: "
- falist nil t nil nil (car falist)))
+ (push (cons (concat sf-name " (" type ")") f) falist)))
+ (setq file (completing-read "Choose a filtered items file: " falist nil t
nil
+ 'todo--fifiles-history (caar falist)))
(setq file (cdr (assoc-string file falist)))
(find-file file)
(unless (derived-mode-p 'todo-filtered-items-mode)
@@ -4050,25 +4068,27 @@ regexp items."
(defun todo-go-to-source-item ()
"Display the file and category of the filtered item at point."
(interactive)
- (let* ((str (todo-item-string))
- (buf (current-buffer))
- (res (todo-find-item str))
- (found (nth 0 res))
- (file (nth 1 res))
- (cat (nth 2 res)))
- (if (not found)
- (message "Category %s does not contain this item." cat)
- (kill-buffer buf)
- (set-window-buffer (selected-window)
- (set-buffer (find-buffer-visiting file)))
- (setq todo-current-todo-file file)
- (setq todo-category-number (todo-category-number cat))
- (let ((todo-show-with-done (if (or todo-filter-done-items
- (eq (cdr found) 'done))
- t
- todo-show-with-done)))
- (todo-category-select))
- (goto-char (car found)))))
+ (unless (looking-at "^$") ; Empty line at EOB.
+ (let* ((str (todo-item-string))
+ (buf (current-buffer))
+ (res (todo-find-item str))
+ (found (nth 0 res))
+ (file (nth 1 res))
+ (cat (nth 2 res)))
+ (if (not found)
+ (message "Category %s does not contain this item." cat)
+ (kill-buffer buf)
+ (set-window-buffer (selected-window)
+ (set-buffer (find-buffer-visiting file)))
+ (setq todo-current-todo-file file)
+ (setq todo-category-number (todo-category-number cat))
+ (let ((todo-show-with-done (if (or todo-filter-done-items
+ (eq (cdr found) 'done))
+ t
+ todo-show-with-done)))
+ (todo-category-select))
+ (if transient-mark-mode (deactivate-mark))
+ (goto-char (car found))))))
(defvar todo-multiple-filter-files nil
"List of files selected from `todo-multiple-filter-files' widget.")
@@ -4520,8 +4540,11 @@ its priority has changed, and `same' otherwise."
(defun todo-save-filtered-items-buffer ()
"Save current Filtered Items buffer to a file.
If the file already exists, overwrite it only on confirmation."
- (let ((filename (or (buffer-file-name) (todo-filter-items-filename))))
- (write-file filename t)))
+ (let ((filename (or (buffer-file-name) (todo-filter-items-filename)))
+ (bufname (buffer-name)))
+ (write-file filename t)
+ (setq buffer-read-only t)
+ (rename-buffer bufname)))
;;
-----------------------------------------------------------------------------
;;; Printing Todo mode buffers
@@ -5132,6 +5155,8 @@ but the categories sexp differs from the current value of
(forward-line)
(looking-at (concat "^"
(regexp-quote todo-category-done))))))
+ ;; Point is on done items separator.
+ (save-excursion (beginning-of-line) (looking-at todo-category-done))
;; Buffer is widened.
(looking-at (regexp-quote todo-category-beg)))
(goto-char (line-beginning-position))
@@ -5141,8 +5166,11 @@ but the categories sexp differs from the current value of
(defun todo-item-end ()
"Move to end of current todo item and return its position."
- ;; Items cannot end with a blank line.
- (unless (looking-at "^$")
+ (unless (or
+ ;; Items cannot end with a blank line.
+ (looking-at "^$")
+ ;; Point is on done items separator.
+ (save-excursion (beginning-of-line) (looking-at
todo-category-done)))
(let* ((done (todo-done-item-p))
(to-lim nil)
;; For todo items, end is before the done items section, for done
@@ -5293,6 +5321,7 @@ Overrides `diary-goto-entry'."
nil t)
(todo-category-number (match-string 1))
(todo-category-select)
+ (if transient-mark-mode (deactivate-mark))
(goto-char opoint))))))
(add-function :override diary-goto-entry-function #'todo-diary-goto-entry)
@@ -6419,9 +6448,6 @@ Filtered Items mode following todo (not done) items."
("N" todo-toggle-prefix-numbers)
("PB" todo-print-buffer)
("PF" todo-print-buffer-to-file)
- ("b" todo-backward-category)
- ("d" todo-item-done)
- ("f" todo-forward-category)
("j" todo-jump-to-category)
("n" todo-next-item)
("p" todo-previous-item)
@@ -6436,6 +6462,8 @@ Filtered Items mode following todo (not done) items."
("Fc" todo-show-categories-table)
("S" todo-search)
("X" todo-clear-matches)
+ ("b" todo-backward-category)
+ ("f" todo-forward-category)
("*" todo-toggle-mark-item)
)
"List of key bindings for Todo and Todo Archive modes.")
diff --git a/lisp/char-fold.el b/lisp/char-fold.el
index 9c05e36..86bd603 100644
--- a/lisp/char-fold.el
+++ b/lisp/char-fold.el
@@ -214,7 +214,7 @@ from which to start."
(when (> spaces 0)
(push (char-fold--make-space-string spaces) out))
(let ((regexp (apply #'concat (nreverse out))))
- ;; Limited by `MAX_BUF_SIZE' in `regex.c'.
+ ;; Limited by `MAX_BUF_SIZE' in `regex-emacs.c'.
(if (> (length regexp) 5000)
(regexp-quote string)
regexp))))
diff --git a/lisp/comint.el b/lisp/comint.el
index 71a2b5e..a9c3e47 100644
--- a/lisp/comint.el
+++ b/lisp/comint.el
@@ -78,7 +78,7 @@
;;
;; Not bound by default in comint-mode (some are in shell mode)
;; comint-run Run a program under comint-mode
-;; send-invisible Read a line w/o echo, and send to proc
+;; comint-send-invisible Read a line w/o echo, and send to proc
;; comint-dynamic-complete-filename Complete filename at point.
;; comint-dynamic-list-filename-completions List completions in help buffer.
;; comint-replace-by-expanded-filename Expand and complete filename at point;
@@ -632,7 +632,7 @@ Input ring history expansion can be achieved with the
commands
Input ring expansion is controlled by the variable `comint-input-autoexpand',
and addition is controlled by the variable `comint-input-ignoredups'.
-Commands with no default key bindings include `send-invisible',
+Commands with no default key bindings include `comint-send-invisible',
`completion-at-point', `comint-dynamic-list-filename-completions', and
`comint-magic-space'.
@@ -2247,7 +2247,7 @@ This function could be on
`comint-output-filter-functions' or bound to a key."
(error nil))
(while (re-search-forward "\r+$" pmark t)
(replace-match "" t t)))))
-(defalias 'shell-strip-ctrl-m 'comint-strip-ctrl-m)
+(define-obsolete-function-alias 'shell-strip-ctrl-m #'comint-strip-ctrl-m
"27.1")
(defun comint-show-maximum-output ()
"Put the end of the buffer at the bottom of the window."
@@ -2357,9 +2357,9 @@ a buffer local variable."
;; These three functions are for entering text you don't want echoed or
;; saved -- typically passwords to ftp, telnet, or somesuch.
-;; Just enter m-x send-invisible and type in your line.
+;; Just enter m-x comint-send-invisible and type in your line.
-(defun send-invisible (&optional prompt)
+(defun comint-send-invisible (&optional prompt)
"Read a string without echoing.
Then send it to the process running in the current buffer.
The string is sent using `comint-input-sender'.
@@ -2382,18 +2382,19 @@ Security bug: your string can still be temporarily
recovered with
(message "Warning: text will be echoed")))
(error "Buffer %s has no process" (current-buffer)))))
+(define-obsolete-function-alias 'send-invisible #'comint-send-invisible "27.1")
+
(defun comint-watch-for-password-prompt (string)
"Prompt in the minibuffer for password and send without echoing.
-This function uses `send-invisible' to read and send a password to the buffer's
-process if STRING contains a password prompt defined by
-`comint-password-prompt-regexp'.
+Looks for a match to `comint-password-prompt-regexp' in order
+to detect the need to (prompt and) send a password.
This function could be in the list `comint-output-filter-functions'."
(when (let ((case-fold-search t))
(string-match comint-password-prompt-regexp string))
(when (string-match "^[ \n\r\t\v\f\b\a]+" string)
(setq string (replace-match "" t t string)))
- (send-invisible string)))
+ (comint-send-invisible string)))
;; Low-level process communication
diff --git a/lisp/cus-theme.el b/lisp/cus-theme.el
index e5e7877..995c55b 100644
--- a/lisp/cus-theme.el
+++ b/lisp/cus-theme.el
@@ -1,4 +1,4 @@
-;;; cus-theme.el -- custom theme creation user interface
+;;; cus-theme.el -- custom theme creation user interface -*- lexical-binding:
t -*-
;;
;; Copyright (C) 2001-2018 Free Software Foundation, Inc.
;;
@@ -47,7 +47,7 @@
Do not call this mode function yourself. It is meant for internal use."
(use-local-map custom-new-theme-mode-map)
(custom--initialize-widget-variables)
- (set (make-local-variable 'revert-buffer-function) 'custom-theme-revert))
+ (setq-local revert-buffer-function #'custom-theme-revert))
(put 'custom-new-theme-mode 'mode-class 'special)
(defvar custom-theme-name nil)
@@ -93,15 +93,14 @@ named *Custom Theme*."
(switch-to-buffer (get-buffer-create (or buffer "*Custom Theme*")))
(let ((inhibit-read-only t))
(erase-buffer)
- (dolist (ov (overlays-in (point-min) (point-max)))
- (delete-overlay ov)))
+ (delete-all-overlays))
(custom-new-theme-mode)
(make-local-variable 'custom-theme-name)
- (set (make-local-variable 'custom-theme--save-name) theme)
- (set (make-local-variable 'custom-theme-faces) nil)
- (set (make-local-variable 'custom-theme-variables) nil)
- (set (make-local-variable 'custom-theme-description) "")
- (set (make-local-variable 'custom-theme--migrate-settings) nil)
+ (setq-local custom-theme--save-name theme)
+ (setq-local custom-theme-faces nil)
+ (setq-local custom-theme-variables nil)
+ (setq-local custom-theme-description "")
+ (setq-local custom-theme--migrate-settings nil)
(make-local-variable 'custom-theme-insert-face-marker)
(make-local-variable 'custom-theme-insert-variable-marker)
(make-local-variable 'custom-theme--listed-faces)
@@ -118,13 +117,13 @@ remove them from your saved Custom file.\n\n"))
:tag " Visit Theme "
:help-echo "Insert the settings of a pre-defined theme."
:action (lambda (_widget &optional _event)
- (call-interactively 'custom-theme-visit-theme)))
+ (call-interactively #'custom-theme-visit-theme)))
(widget-insert " ")
(widget-create 'push-button
:tag " Merge Theme "
:help-echo "Merge in the settings of a pre-defined theme."
:action (lambda (_widget &optional _event)
- (call-interactively 'custom-theme-merge-theme)))
+ (call-interactively #'custom-theme-merge-theme)))
(widget-insert " ")
(widget-create 'push-button
:tag " Revert "
@@ -142,7 +141,7 @@ remove them from your saved Custom file.\n\n"))
(widget-create 'text
:value (format-time-string "Created %Y-%m-%d.")))
(widget-create 'push-button
- :notify (function custom-theme-write)
+ :notify #'custom-theme-write
" Save Theme ")
(when (eq theme 'user)
(setq custom-theme--migrate-settings t)
@@ -188,7 +187,7 @@ remove them from your saved Custom file.\n\n"))
:mouse-face 'highlight
:pressed-face 'highlight
:action (lambda (_widget &optional _event)
- (call-interactively 'custom-theme-add-face)))
+ (call-interactively #'custom-theme-add-face)))
;; If THEME is non-nil, insert all of that theme's variables.
(widget-insert "\n\n Theme variables:\n ")
@@ -207,7 +206,7 @@ remove them from your saved Custom file.\n\n"))
:mouse-face 'highlight
:pressed-face 'highlight
:action (lambda (_widget &optional _event)
- (call-interactively 'custom-theme-add-variable)))
+ (call-interactively #'custom-theme-add-variable)))
(widget-insert ?\n)
(widget-setup)
(goto-char (point-min))
@@ -254,7 +253,7 @@ interactively, this defaults to the current value of VAR."
:tag (custom-unlispify-tag-name symbol)
:value symbol
:shown-value (list val)
- :notify 'ignore
+ :notify #'ignore
:custom-level 0
:custom-state 'hidden
:custom-style 'simple))
@@ -313,7 +312,7 @@ SPEC, if non-nil, should be a face spec to which to set the
widget."
(interactive
(list
(intern (completing-read "Find custom theme: "
- (mapcar 'symbol-name
+ (mapcar #'symbol-name
(custom-available-themes))))))
(unless (custom-theme-name-valid-p theme)
(error "No valid theme named `%s'" theme))
@@ -328,7 +327,7 @@ SPEC, if non-nil, should be a face spec to which to set the
widget."
(interactive
(list
(intern (completing-read "Merge custom theme: "
- (mapcar 'symbol-name
+ (mapcar #'symbol-name
(custom-available-themes))))))
(unless (eq theme 'user)
(unless (custom-theme-name-valid-p theme)
@@ -343,8 +342,8 @@ SPEC, if non-nil, should be a face spec to which to set the
widget."
(memq name '(custom-enabled-themes
custom-safe-themes)))
(funcall (if option
- 'custom-theme-add-variable
- 'custom-theme-add-face)
+ #'custom-theme-add-variable
+ #'custom-theme-add-face)
name value)))))
theme)
@@ -475,7 +474,7 @@ It includes all faces in list FACES."
(interactive
(list
(intern (completing-read "Describe custom theme: "
- (mapcar 'symbol-name
+ (mapcar #'symbol-name
(custom-available-themes))))))
(unless (custom-theme-name-valid-p theme)
(error "Invalid theme name `%s'" theme))
@@ -513,8 +512,7 @@ It includes all faces in list FACES."
(condition-case nil
(read (current-buffer))
(end-of-file nil)))))
- (and sexp (listp sexp)
- (eq (car sexp) 'deftheme)
+ (and (eq (car-safe sexp) 'deftheme)
(setq doc (nth 2 sexp)))))))
(princ "\n\nDocumentation:\n")
(princ (if (stringp doc)
@@ -552,10 +550,10 @@ It includes all faces in list FACES."
Do not call this mode function yourself. It is meant for internal use."
(use-local-map custom-theme-choose-mode-map)
(custom--initialize-widget-variables)
- (set (make-local-variable 'revert-buffer-function)
- (lambda (_ignore-auto noconfirm)
- (when (or noconfirm (y-or-n-p "Discard current choices? "))
- (customize-themes (current-buffer))))))
+ (setq-local revert-buffer-function
+ (lambda (_ignore-auto noconfirm)
+ (when (or noconfirm (y-or-n-p "Discard current choices? "))
+ (customize-themes (current-buffer))))))
(put 'custom-theme-choose-mode 'mode-class 'special)
;;;###autoload
@@ -568,7 +566,7 @@ omitted, a buffer named *Custom Themes* is used."
(let ((inhibit-read-only t))
(erase-buffer))
(custom-theme-choose-mode)
- (set (make-local-variable 'custom--listed-themes) nil)
+ (setq-local custom--listed-themes nil)
(make-local-variable 'custom-theme-allow-multiple-selections)
(and (null custom-theme-allow-multiple-selections)
(> (length custom-enabled-themes) 1)
@@ -616,11 +614,11 @@ Theme files are named *-theme.el in `"))
(widget-create 'push-button
:tag " Save Theme Settings "
:help-echo "Save the selected themes for future sessions."
- :action 'custom-theme-save)
+ :action #'custom-theme-save)
(widget-insert ?\n)
(widget-create 'checkbox
:value custom-theme-allow-multiple-selections
- :action 'custom-theme-selections-toggle)
+ :action #'custom-theme-selections-toggle)
(widget-insert (propertize " Select more than one theme at a time"
'face '(variable-pitch (:height 0.9))))
@@ -632,13 +630,13 @@ Theme files are named *-theme.el in `"))
:value (custom-theme-enabled-p theme)
:theme-name theme
:help-echo help-echo
- :action 'custom-theme-checkbox-toggle))
+ :action #'custom-theme-checkbox-toggle))
(push (cons theme widget) custom--listed-themes)
(widget-create-child-and-convert widget 'push-button
:button-face-get 'ignore
:mouse-face-get 'ignore
:value (format " %s" theme)
- :action 'widget-parent-action
+ :action #'widget-parent-action
:help-echo help-echo)
(widget-insert " -- "
(propertize (custom-theme-summary theme)
@@ -662,8 +660,7 @@ Theme files are named *-theme.el in `"))
(condition-case nil
(read (current-buffer))
(end-of-file nil)))))
- (and sexp (listp sexp)
- (eq (car sexp) 'deftheme)
+ (and (eq (car-safe sexp) 'deftheme)
(setq doc (nth 2 sexp))))))))
(cond ((null doc)
"(no documentation available)")
diff --git a/lisp/custom.el b/lisp/custom.el
index 4a778a0..a08f7fd 100644
--- a/lisp/custom.el
+++ b/lisp/custom.el
@@ -1,4 +1,4 @@
-;;; custom.el --- tools for declaring and initializing options
+;;; custom.el --- tools for declaring and initializing options -*-
lexical-binding: t -*-
;;
;; Copyright (C) 1996-1997, 1999, 2001-2018 Free Software Foundation,
;; Inc.
@@ -150,7 +150,7 @@ set to nil, as the value is no longer rogue."
(put symbol 'force-value nil))
(if (keywordp doc)
(error "Doc string is missing"))
- (let ((initialize 'custom-initialize-reset)
+ (let ((initialize #'custom-initialize-reset)
(requests nil))
(unless (memq :group args)
(custom-add-to-group (custom-current-group) symbol 'custom-variable))
@@ -426,7 +426,7 @@ information."
(defun custom-declare-group (symbol members doc &rest args)
"Like `defgroup', but SYMBOL is evaluated as a normal argument."
(while members
- (apply 'custom-add-to-group symbol (car members))
+ (apply #'custom-add-to-group symbol (car members))
(setq members (cdr members)))
(when doc
;; This text doesn't get into DOC.
@@ -618,11 +618,8 @@ VARIABLE is a symbol that names a user option.
The result is that the change is treated as having been made through Custom."
(put variable 'customized-value (list (custom-quote (eval variable)))))
-
-;;; Custom Themes
-
-;;; Loading files needed to customize a symbol.
-;;; This is in custom.el because menu-bar.el needs it for toggle cmds.
+;; Loading files needed to customize a symbol.
+;; This is in custom.el because menu-bar.el needs it for toggle cmds.
(defvar custom-load-recursion nil
"Hack to avoid recursive dependencies.")
@@ -633,14 +630,12 @@ The result is that the change is treated as having been
made through Custom."
(let ((custom-load-recursion t))
;; Load these files if not already done,
;; to make sure we know all the dependencies of SYMBOL.
- (condition-case nil
- (require 'cus-load)
- (error nil))
- (condition-case nil
- (require 'cus-start)
- (error nil))
+ (ignore-errors
+ (require 'cus-load))
+ (ignore-errors
+ (require 'cus-start))
(dolist (load (get symbol 'custom-loads))
- (cond ((symbolp load) (condition-case nil (require load) (error nil)))
+ (cond ((symbolp load) (ignore-errors (require load)))
;; This is subsumed by the test below, but it's much faster.
((assoc load load-history))
;; This was just (assoc (locate-library load) load-history)
@@ -658,7 +653,7 @@ The result is that the change is treated as having been
made through Custom."
;; We are still loading it when we call this,
;; and it is not in load-history yet.
((equal load "cus-edit"))
- (t (condition-case nil (load load) (error nil))))))))
+ (t (ignore-errors (load load))))))))
(defvar custom-local-buffer nil
"Non-nil, in a Customization buffer, means customize a specific buffer.
@@ -691,16 +686,12 @@ this sets the local binding in that buffer instead."
(defun custom-quote (sexp)
"Quote SEXP if it is not self quoting."
- (if (or (memq sexp '(t nil))
- (keywordp sexp)
- (and (listp sexp)
- (memq (car sexp) '(lambda)))
- (stringp sexp)
- (numberp sexp)
- (vectorp sexp)
-;;; (and (fboundp 'characterp)
-;;; (characterp sexp))
- )
+ ;; Can't use `macroexp-quote' because it is loaded after `custom.el'
+ ;; during bootstrap. See `loadup.el'.
+ (if (and (not (consp sexp))
+ (or (keywordp sexp)
+ (not (symbolp sexp))
+ (booleanp sexp)))
sexp
(list 'quote sexp)))
@@ -715,18 +706,16 @@ To actually save the value, call `custom-save-all'.
Return non-nil if the `saved-value' property actually changed."
(custom-load-symbol symbol)
- (let* ((get (or (get symbol 'custom-get) 'default-value))
+ (let* ((get (or (get symbol 'custom-get) #'default-value))
(value (funcall get symbol))
(saved (get symbol 'saved-value))
(standard (get symbol 'standard-value))
(comment (get symbol 'customized-variable-comment)))
;; Save default value if different from standard value.
- (if (or (null standard)
- (not (equal value (condition-case nil
- (eval (car standard))
- (error nil)))))
- (put symbol 'saved-value (list (custom-quote value)))
- (put symbol 'saved-value nil))
+ (put symbol 'saved-value
+ (unless (and standard
+ (equal value (ignore-errors (eval (car standard)))))
+ (list (custom-quote value))))
;; Clear customized information (set, but not saved).
(put symbol 'customized-value nil)
;; Save any comment that might have been set.
@@ -744,15 +733,14 @@ default value. Otherwise, set it to nil.
Return non-nil if the `customized-value' property actually changed."
(custom-load-symbol symbol)
- (let* ((get (or (get symbol 'custom-get) 'default-value))
+ (let* ((get (or (get symbol 'custom-get) #'default-value))
(value (funcall get symbol))
(customized (get symbol 'customized-value))
(old (or (get symbol 'saved-value) (get symbol 'standard-value))))
;; Mark default value as set if different from old value.
(if (not (and old
- (equal value (condition-case nil
- (eval (car old))
- (error nil)))))
+ (equal value (ignore-errors
+ (eval (car old))))))
(progn (put symbol 'customized-value (list (custom-quote value)))
(custom-push-theme 'theme-value symbol 'user 'set
(custom-quote value)))
@@ -776,7 +764,7 @@ E.g. dumped variables whose default depends on run-time
information."
;; always do the funcall step, even if symbol was not bound before.
(or (default-boundp symbol)
(eval `(defvar ,symbol nil))) ; reset below, so any value is fine
- (funcall (or (get symbol 'custom-set) 'set-default)
+ (funcall (or (get symbol 'custom-set) #'set-default)
symbol
(eval (car (or (get symbol 'saved-value) (get symbol
'standard-value))))))
@@ -946,7 +934,7 @@ the default value for the SYMBOL to the value of EXP.
REQUEST is a list of features we must require in order to
handle SYMBOL properly.
COMMENT is a comment string about SYMBOL."
- (apply 'custom-theme-set-variables 'user args))
+ (apply #'custom-theme-set-variables 'user args))
(defun custom-theme-set-variables (theme &rest args)
"Initialize variables for theme THEME according to settings in ARGS.
@@ -994,8 +982,8 @@ COMMENT is a comment string about SYMBOL."
set)
(when requests
(put symbol 'custom-requests requests)
- (mapc 'require requests))
- (setq set (or (get symbol 'custom-set) 'custom-set-default))
+ (mapc #'require requests))
+ (setq set (or (get symbol 'custom-set) #'custom-set-default))
(put symbol 'saved-value (list value))
(put symbol 'saved-variable-comment comment)
;; Allow for errors in the case where the setter has
@@ -1091,26 +1079,29 @@ list, in which A occurs before B if B was defined with a
;; they were used to supply keyword-value pairs like `:immediate',
;; `:variable-reset-string', etc. We don't use any of these, so ignore them.
-(defmacro deftheme (theme &optional doc &rest ignored)
+(defmacro deftheme (theme &optional doc &rest _ignored)
"Declare THEME to be a Custom theme.
The optional argument DOC is a doc string describing the theme.
Any theme `foo' should be defined in a file called `foo-theme.el';
see `custom-make-theme-feature' for more information."
- (declare (doc-string 2))
+ (declare (doc-string 2)
+ (advertised-calling-convention (theme &optional doc) "22.1"))
(let ((feature (custom-make-theme-feature theme)))
;; It is better not to use backquote in this file,
;; because that makes a bootstrapping problem
;; if you need to recompile all the Lisp files using interpreted code.
(list 'custom-declare-theme (list 'quote theme) (list 'quote feature)
doc)))
-(defun custom-declare-theme (theme feature &optional doc &rest ignored)
+(defun custom-declare-theme (theme feature &optional doc &rest _ignored)
"Like `deftheme', but THEME is evaluated as a normal argument.
FEATURE is the feature this theme provides. Normally, this is a symbol
created from THEME by `custom-make-theme-feature'."
+ (declare (advertised-calling-convention (theme feature &optional doc)
"22.1"))
(unless (custom-theme-name-valid-p theme)
(error "Custom theme cannot be named %S" theme))
- (add-to-list 'custom-known-themes theme)
+ (unless (memq theme custom-known-themes)
+ (push theme custom-known-themes))
(put theme 'theme-feature feature)
(when doc (put theme 'theme-documentation doc)))
@@ -1218,7 +1209,7 @@ Return t if THEME was successfully loaded, nil otherwise."
(interactive
(list
(intern (completing-read "Load custom theme: "
- (mapcar 'symbol-name
+ (mapcar #'symbol-name
(custom-available-themes))))
nil nil))
(unless (custom-theme-name-valid-p theme)
@@ -1233,43 +1224,47 @@ Return t if THEME was successfully loaded, nil
otherwise."
(put theme 'theme-settings nil)
(put theme 'theme-feature nil)
(put theme 'theme-documentation nil))
- (let ((fn (locate-file (concat (symbol-name theme) "-theme.el")
- (custom-theme--load-path)
- '("" "c"))))
- (unless fn
- (error "Unable to find theme file for `%s'" theme))
- (with-temp-buffer
- (insert-file-contents fn)
- ;; Check file safety with `custom-safe-themes', prompting the
- ;; user if necessary.
- (when (or no-confirm
- (eq custom-safe-themes t)
- (and (memq 'default custom-safe-themes)
- (equal (file-name-directory fn)
- (expand-file-name "themes/" data-directory)))
- (let ((hash (secure-hash 'sha256 (current-buffer))))
- (or (member hash custom-safe-themes)
- (custom-theme-load-confirm hash))))
- (let ((custom--inhibit-theme-enable t)
- (buffer-file-name fn)) ;For load-history.
- (eval-buffer))
- ;; Optimization: if the theme changes the `default' face, put that
- ;; entry first. This avoids some `frame-set-background-mode' rigmarole
- ;; by assigning the new background immediately.
- (let* ((settings (get theme 'theme-settings))
- (tail settings)
- found)
- (while (and tail (not found))
- (and (eq (nth 0 (car tail)) 'theme-face)
- (eq (nth 1 (car tail)) 'default)
- (setq found (car tail)))
- (setq tail (cdr tail)))
- (if found
- (put theme 'theme-settings (cons found (delq found settings)))))
- ;; Finally, enable the theme.
- (unless no-enable
- (enable-theme theme))
- t))))
+ (let ((file (locate-file (concat (symbol-name theme) "-theme.el")
+ (custom-theme--load-path)
+ '("" "c")))
+ (custom--inhibit-theme-enable t))
+ ;; Check file safety with `custom-safe-themes', prompting the
+ ;; user if necessary.
+ (cond ((not file)
+ (error "Unable to find theme file for `%s'" theme))
+ ((or no-confirm
+ (eq custom-safe-themes t)
+ (and (memq 'default custom-safe-themes)
+ (equal (file-name-directory file)
+ (expand-file-name "themes/" data-directory))))
+ ;; Theme is safe; load byte-compiled version if available.
+ (load (file-name-sans-extension file) nil t nil t))
+ ((with-temp-buffer
+ (insert-file-contents file)
+ (let ((hash (secure-hash 'sha256 (current-buffer))))
+ (when (or (member hash custom-safe-themes)
+ (custom-theme-load-confirm hash))
+ (eval-buffer nil nil file)
+ t))))
+ (t
+ (error "Unable to load theme `%s'" theme))))
+ ;; Optimization: if the theme changes the `default' face, put that
+ ;; entry first. This avoids some `frame-set-background-mode' rigmarole
+ ;; by assigning the new background immediately.
+ (let* ((settings (get theme 'theme-settings))
+ (tail settings)
+ found)
+ (while (and tail (not found))
+ (and (eq (nth 0 (car tail)) 'theme-face)
+ (eq (nth 1 (car tail)) 'default)
+ (setq found (car tail)))
+ (setq tail (cdr tail)))
+ (when found
+ (put theme 'theme-settings (cons found (delq found settings)))))
+ ;; Finally, enable the theme.
+ (unless no-enable
+ (enable-theme theme))
+ t)
(defun custom-theme-load-confirm (hash)
"Query the user about loading a Custom theme that may not be safe.
@@ -1292,11 +1287,9 @@ query also about adding HASH to `custom-safe-themes'."
(defun custom-theme-name-valid-p (name)
"Return t if NAME is a valid name for a Custom theme, nil otherwise.
NAME should be a symbol."
- (and (symbolp name)
- name
- (not (or (zerop (length (symbol-name name)))
- (eq name 'user)
- (eq name 'changed)))))
+ (and (not (memq name '(nil user changed)))
+ (symbolp name)
+ (not (string= "" (symbol-name name)))))
(defun custom-available-themes ()
"Return a list of Custom themes available for loading.
@@ -1307,19 +1300,25 @@ The returned symbols may not correspond to themes that
have been
loaded, and no effort is made to check that the files contain
valid Custom themes. For a list of loaded themes, check the
variable `custom-known-themes'."
- (let (sym themes)
+ (let ((suffix "-theme\\.el\\'")
+ themes)
(dolist (dir (custom-theme--load-path))
- (when (file-directory-p dir)
- (dolist (file (file-expand-wildcards
- (expand-file-name "*-theme.el" dir) t))
- (setq file (file-name-nondirectory file))
- (and (string-match "\\`\\(.+\\)-theme.el\\'" file)
- (setq sym (intern (match-string 1 file)))
- (custom-theme-name-valid-p sym)
- (push sym themes)))))
- (nreverse (delete-dups themes))))
+ ;; `custom-theme--load-path' promises DIR exists and is a
+ ;; directory, but `custom.el' is loaded too early during
+ ;; bootstrap to use `cl-lib' macros, so guard with
+ ;; `file-directory-p' instead of calling `cl-assert'.
+ (dolist (file (and (file-directory-p dir)
+ (directory-files dir nil suffix)))
+ (let ((theme (intern (substring file 0 (string-match-p suffix file)))))
+ (and (custom-theme-name-valid-p theme)
+ (not (memq theme themes))
+ (push theme themes)))))
+ (nreverse themes)))
(defun custom-theme--load-path ()
+ "Expand `custom-theme-load-path' into a list of directories.
+Members of `custom-theme-load-path' that either don't exist or
+are not directories are omitted from the expansion."
(let (lpath)
(dolist (f custom-theme-load-path)
(cond ((eq f 'custom-theme-directory)
@@ -1346,8 +1345,8 @@ function runs. To disable other themes, use
`disable-theme'."
(completing-read
"Enable custom theme: "
obarray (lambda (sym) (get sym 'theme-settings)) t))))
- (if (not (custom-theme-p theme))
- (error "Undefined Custom theme %s" theme))
+ (unless (custom-theme-p theme)
+ (error "Undefined Custom theme %s" theme))
(let ((settings (get theme 'theme-settings)))
;; Loop through theme settings, recalculating vars/faces.
(dolist (s settings)
@@ -1387,23 +1386,23 @@ Setting this variable through Customize calls
`enable-theme' or
(let (failures)
(setq themes (delq 'user (delete-dups themes)))
;; Disable all themes not in THEMES.
- (if (boundp symbol)
- (dolist (theme (symbol-value symbol))
- (if (not (memq theme themes))
- (disable-theme theme))))
+ (dolist (theme (and (boundp symbol)
+ (symbol-value symbol)))
+ (unless (memq theme themes)
+ (disable-theme theme)))
;; Call `enable-theme' or `load-theme' on each of THEMES.
(dolist (theme (reverse themes))
(condition-case nil
(if (custom-theme-p theme)
(enable-theme theme)
(load-theme theme))
- (error (setq failures (cons theme failures)
- themes (delq theme themes)))))
+ (error (push theme failures)
+ (setq themes (delq theme themes)))))
(enable-theme 'user)
(custom-set-default symbol themes)
- (if failures
- (message "Failed to enable theme: %s"
- (mapconcat 'symbol-name failures ", "))))))
+ (when failures
+ (message "Failed to enable theme(s): %s"
+ (mapconcat #'symbol-name failures ", "))))))
(defsubst custom-theme-enabled-p (theme)
"Return non-nil if THEME is enabled."
@@ -1415,7 +1414,7 @@ See `custom-enabled-themes' for a list of enabled themes."
(interactive (list (intern
(completing-read
"Disable custom theme: "
- (mapcar 'symbol-name custom-enabled-themes)
+ (mapcar #'symbol-name custom-enabled-themes)
nil t))))
(when (custom-theme-enabled-p theme)
(let ((settings (get theme 'theme-settings)))
@@ -1431,23 +1430,23 @@ See `custom-enabled-themes' for a list of enabled
themes."
;; If the face spec specified by this theme is in the
;; saved-face property, reset that property.
(when (equal (nth 3 s) (get symbol 'saved-face))
- (put symbol 'saved-face (and val (cadr (car val)))))))))
- ;; Recompute faces on all frames.
- (dolist (frame (frame-list))
- ;; We must reset the fg and bg color frame parameters, or
- ;; `face-set-after-frame-default' will use the existing
- ;; parameters, which could be from the disabled theme.
- (set-frame-parameter frame 'background-color
- (custom--frame-color-default
- frame :background "background" "Background"
- "unspecified-bg" "white"))
- (set-frame-parameter frame 'foreground-color
- (custom--frame-color-default
- frame :foreground "foreground" "Foreground"
- "unspecified-fg" "black"))
- (face-set-after-frame-default frame))
- (setq custom-enabled-themes
- (delq theme custom-enabled-themes)))))
+ (put symbol 'saved-face (cadar val))))))))
+ ;; Recompute faces on all frames.
+ (dolist (frame (frame-list))
+ ;; We must reset the fg and bg color frame parameters, or
+ ;; `face-set-after-frame-default' will use the existing
+ ;; parameters, which could be from the disabled theme.
+ (set-frame-parameter frame 'background-color
+ (custom--frame-color-default
+ frame :background "background" "Background"
+ "unspecified-bg" "white"))
+ (set-frame-parameter frame 'foreground-color
+ (custom--frame-color-default
+ frame :foreground "foreground" "Foreground"
+ "unspecified-fg" "black"))
+ (face-set-after-frame-default frame))
+ (setq custom-enabled-themes
+ (delq theme custom-enabled-themes))))
;; Only used if window-system not null.
(declare-function x-get-resource "frame.c"
@@ -1481,7 +1480,7 @@ This function returns nil if no custom theme specifies a
value for VARIABLE."
(if (and valspec
(or (get variable 'force-value)
(default-boundp variable)))
- (funcall (or (get variable 'custom-set) 'set-default) variable
+ (funcall (or (get variable 'custom-set) #'set-default) variable
(eval (car valspec))))))
(defun custom-theme-recalc-face (face)
@@ -1522,7 +1521,7 @@ Each of the arguments ARGS has this form:
(VARIABLE IGNORED)
This means reset VARIABLE. (The argument IGNORED is ignored)."
- (apply 'custom-theme-reset-variables 'user args))
+ (apply #'custom-theme-reset-variables 'user args))
;;; The End.
diff --git a/lisp/dired-aux.el b/lisp/dired-aux.el
index 925a7d5..21ee50c 100644
--- a/lisp/dired-aux.el
+++ b/lisp/dired-aux.el
@@ -1989,6 +1989,19 @@ Optional arg HOW-TO determines how to treat the target.
dired-dirs)))
+
+;; We use this function in `dired-create-directory' and
+;; `dired-create-empty-file'; the return value is the new entry
+;; in the updated Dired buffer.
+(defun dired--find-topmost-parent-dir (filename)
+ "Return the topmost nonexistent parent dir of FILENAME.
+FILENAME is a full file name."
+ (let ((try filename) new)
+ (while (and try (not (file-exists-p try)) (not (equal new try)))
+ (setq new try
+ try (directory-file-name (file-name-directory try))))
+ new))
+
;;;###autoload
(defun dired-create-directory (directory)
"Create a directory called DIRECTORY.
@@ -1997,18 +2010,32 @@ If DIRECTORY already exists, signal an error."
(interactive
(list (read-file-name "Create directory: " (dired-current-directory))))
(let* ((expanded (directory-file-name (expand-file-name directory)))
- (try expanded) new)
+ new)
(if (file-exists-p expanded)
(error "Cannot create directory %s: file exists" expanded))
- ;; Find the topmost nonexistent parent dir (variable `new')
- (while (and try (not (file-exists-p try)) (not (equal new try)))
- (setq new try
- try (directory-file-name (file-name-directory try))))
+ (setq new (dired--find-topmost-parent-dir expanded))
(make-directory expanded t)
(when new
(dired-add-file new)
(dired-move-to-filename))))
+;;;###autoload
+(defun dired-create-empty-file (file)
+ "Create an empty file called FILE.
+ Add a new entry for the new file in the Dired buffer.
+ Parent directories of FILE are created as needed.
+ If FILE already exists, signal an error."
+ (interactive (list (read-file-name "Create empty file: ")))
+ (let* ((expanded (expand-file-name file))
+ new)
+ (if (file-exists-p expanded)
+ (error "Cannot create file %s: file exists" expanded))
+ (setq new (dired--find-topmost-parent-dir expanded))
+ (make-empty-file file 'parents)
+ (when new
+ (dired-add-file new)
+ (dired-move-to-filename))))
+
(defun dired-into-dir-with-symlinks (target)
(and (file-directory-p target)
(not (file-symlink-p target))))
diff --git a/lisp/dired.el b/lisp/dired.el
index 1348df6..26a7449 100644
--- a/lisp/dired.el
+++ b/lisp/dired.el
@@ -1802,6 +1802,9 @@ Do so according to the former subdir alist
OLD-SUBDIR-ALIST."
(define-key map [menu-bar immediate create-directory]
'(menu-item "Create Directory..." dired-create-directory
:help "Create a directory"))
+ (define-key map [menu-bar immediate create-empty-file]
+ '(menu-item "Create Empty file..." dired-create-empty-file
+ :help "Create an empty file"))
(define-key map [menu-bar immediate wdired-mode]
'(menu-item "Edit File Names" wdired-change-to-wdired-mode
:help "Put a Dired buffer in a mode in which filenames are
editable"
diff --git a/lisp/emacs-lisp/backtrace.el b/lisp/emacs-lisp/backtrace.el
new file mode 100644
index 0000000..f13b43b
--- /dev/null
+++ b/lisp/emacs-lisp/backtrace.el
@@ -0,0 +1,916 @@
+;;; backtrace.el --- generic major mode for Elisp backtraces -*-
lexical-binding: t -*-
+
+;; Copyright (C) 2018 Free Software Foundation, Inc.
+
+;; Author: Gemini Lasswell
+;; Keywords: lisp, tools, maint
+;; Version: 1.0
+
+;; This file is part of GNU Emacs.
+
+;; GNU Emacs is free software: you can redistribute it and/or modify
+;; it under the terms of the GNU General Public License as published by
+;; the Free Software Foundation, either version 3 of the License, or
+;; (at your option) any later version.
+
+;; GNU Emacs is distributed in the hope that it will be useful,
+;; but WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+;; GNU General Public License for more details.
+
+;; You should have received a copy of the GNU General Public License
+;; along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>.
+
+;;; Commentary:
+
+;; This file defines Backtrace mode, a generic major mode for displaying
+;; Elisp stack backtraces, which can be used as is or inherited from
+;; by another mode.
+
+;; For usage information, see the documentation of `backtrace-mode'.
+
+;;; Code:
+
+(eval-when-compile (require 'cl-lib))
+(eval-when-compile (require 'pcase))
+(eval-when-compile (require 'subr-x)) ; if-let
+(require 'help-mode) ; Define `help-function-def' button type.
+(require 'lisp-mode)
+
+;;; Options
+
+(defgroup backtrace nil
+ "Viewing of Elisp backtraces."
+ :group 'lisp)
+
+(defcustom backtrace-fontify t
+ "If non-nil, fontify Backtrace buffers.
+Set to nil to disable fontification, which may be necessary in
+order to debug the code that does fontification."
+ :type 'boolean
+ :group 'backtrace
+ :version "27.1")
+
+(defcustom backtrace-line-length 5000
+ "Target length for lines in Backtrace buffers.
+Backtrace mode will attempt to abbreviate printing of backtrace
+frames to make them shorter than this, but success is not
+guaranteed. If set to nil or zero, Backtrace mode will not
+abbreviate the forms it prints."
+ :type 'integer
+ :group 'backtrace
+ :version "27.1")
+
+;;; Backtrace frame data structure
+
+(cl-defstruct
+ (backtrace-frame
+ (:constructor backtrace-make-frame))
+ evald ; Non-nil if argument evaluation is complete.
+ fun ; The function called/to call in this frame.
+ args ; Either evaluated or unevaluated arguments to the function.
+ flags ; A plist, possible properties are :debug-on-exit and
:source-available.
+ locals ; An alist containing variable names and values.
+ buffer ; If non-nil, the buffer in use by eval-buffer or eval-region.
+ pos ; The position in the buffer.
+ )
+
+(cl-defun backtrace-get-frames
+ (&optional base &key (constructor #'backtrace-make-frame))
+ "Collect all frames of current backtrace into a list.
+The list will contain objects made by CONSTRUCTOR, which
+defaults to `backtrace-make-frame' and which, if provided, should
+be the constructor of a structure which includes
+`backtrace-frame'. If non-nil, BASE should be a function, and
+frames before its nearest activation frame are discarded."
+ (let ((frames nil)
+ (eval-buffers eval-buffer-list))
+ (mapbacktrace (lambda (evald fun args flags)
+ (push (funcall constructor
+ :evald evald :fun fun
+ :args args :flags flags)
+ frames))
+ (or base 'backtrace-get-frames))
+ (setq frames (nreverse frames))
+ ;; Add local variables to each frame, and the buffer position
+ ;; to frames containing eval-buffer or eval-region.
+ (dotimes (idx (length frames))
+ (let ((frame (nth idx frames)))
+ ;; `backtrace--locals' gives an error when idx is 0. But the
+ ;; locals for frame 0 are not needed, because when we get here
+ ;; from debug-on-entry, the locals aren't bound yet, and when
+ ;; coming from Edebug or ERT there is an Edebug or ERT
+ ;; function at frame 0.
+ (when (> idx 0)
+ (setf (backtrace-frame-locals frame)
+ (backtrace--locals idx (or base 'backtrace-get-frames))))
+ (when (and eval-buffers (memq (backtrace-frame-fun frame)
+ '(eval-buffer eval-region)))
+ ;; This will get the wrong result if there are two nested
+ ;; eval-region calls for the same buffer. That's not a very
+ ;; useful case.
+ (with-current-buffer (pop eval-buffers)
+ (setf (backtrace-frame-buffer frame) (current-buffer))
+ (setf (backtrace-frame-pos frame) (point))))))
+ frames))
+
+;; Button definition for jumping to a buffer position.
+
+(define-button-type 'backtrace-buffer-pos
+ 'action #'backtrace--pop-to-buffer-pos
+ 'help-echo "mouse-2, RET: Show reading position")
+
+(defun backtrace--pop-to-buffer-pos (button)
+ "Pop to the buffer and position for the BUTTON at point."
+ (let* ((buffer (button-get button 'backtrace-buffer))
+ (pos (button-get button 'backtrace-pos)))
+ (if (buffer-live-p buffer)
+ (progn
+ (pop-to-buffer buffer)
+ (goto-char (max (point-min) (min (point-max) pos))))
+ (message "Buffer has been killed"))))
+
+;; Font Locking support
+
+(defconst backtrace--font-lock-keywords
+ '((backtrace--match-ellipsis-in-string
+ (1 'button prepend)))
+ "Expressions to fontify in Backtrace mode.
+Fontify these in addition to the expressions Emacs Lisp mode
+fontifies.")
+
+(defconst backtrace-font-lock-keywords
+ (append lisp-el-font-lock-keywords-for-backtraces
+ backtrace--font-lock-keywords)
+ "Default expressions to highlight in Backtrace mode.")
+(defconst backtrace-font-lock-keywords-1
+ (append lisp-el-font-lock-keywords-for-backtraces-1
+ backtrace--font-lock-keywords)
+ "Subdued level highlighting for Backtrace mode.")
+(defconst backtrace-font-lock-keywords-2
+ (append lisp-el-font-lock-keywords-for-backtraces-2
+ backtrace--font-lock-keywords)
+ "Gaudy level highlighting for Backtrace mode.")
+
+(defun backtrace--match-ellipsis-in-string (bound)
+ ;; Fontify ellipses within strings as buttons.
+ ;; This is necessary because ellipses are text property buttons
+ ;; instead of overlay buttons, which is done because there could
+ ;; be a large number of them.
+ (when (re-search-forward "\\(\\.\\.\\.\\)\"" bound t)
+ (and (get-text-property (- (point) 2) 'cl-print-ellipsis)
+ (get-text-property (- (point) 3) 'cl-print-ellipsis)
+ (get-text-property (- (point) 4) 'cl-print-ellipsis))))
+
+;;; Xref support
+
+(defun backtrace--xref-backend () 'elisp)
+
+;;; Backtrace mode variables
+
+(defvar-local backtrace-frames nil
+ "Stack frames displayed in the current Backtrace buffer.
+This should be a list of `backtrace-frame' objects.")
+
+(defvar-local backtrace-view nil
+ "A plist describing how to render backtrace frames.
+Possible entries are :show-flags, :show-locals and :print-circle.")
+
+(defvar-local backtrace-insert-header-function nil
+ "Function for inserting a header for the current Backtrace buffer.
+If nil, no header will be created. Note that Backtrace buffers
+are fontified as in Emacs Lisp Mode, the header text included.")
+
+(defvar backtrace-revert-hook nil
+ "Hook run before reverting a Backtrace buffer.
+This is commonly used to recompute `backtrace-frames'.")
+
+(defvar-local backtrace-print-function #'cl-prin1
+ "Function used to print values in the current Backtrace buffer.")
+
+(defvar-local backtrace-goto-source-functions nil
+ "Abnormal hook used to jump to the source code for the current frame.
+Each hook function is called with no argument, and should return
+non-nil if it is able to switch to the buffer containing the
+source code. Execution of the hook will stop if one of the
+functions returns non-nil. When adding a function to this hook,
+you should also set the :source-available flag for the backtrace
+frames where the source code location is known.")
+
+(defvar backtrace-mode-map
+ (let ((map (copy-keymap special-mode-map)))
+ (set-keymap-parent map button-buffer-map)
+ (define-key map "n" 'backtrace-forward-frame)
+ (define-key map "p" 'backtrace-backward-frame)
+ (define-key map "v" 'backtrace-toggle-locals)
+ (define-key map "#" 'backtrace-toggle-print-circle)
+ (define-key map "s" 'backtrace-goto-source)
+ (define-key map "\C-m" 'backtrace-help-follow-symbol)
+ (define-key map "+" 'backtrace-multi-line)
+ (define-key map "-" 'backtrace-single-line)
+ (define-key map "." 'backtrace-expand-ellipses)
+ (define-key map [follow-link] 'mouse-face)
+ (define-key map [mouse-2] 'mouse-select-window)
+ (easy-menu-define nil map ""
+ '("Backtrace"
+ ["Next Frame" backtrace-forward-frame
+ :help "Move cursor forwards to the start of a backtrace frame"]
+ ["Previous Frame" backtrace-backward-frame
+ :help "Move cursor backwards to the start of a backtrace frame"]
+ "--"
+ ["Show Variables" backtrace-toggle-locals
+ :style toggle
+ :active (backtrace-get-index)
+ :selected (plist-get (backtrace-get-view) :show-locals)
+ :help "Show or hide the local variables for the frame at point"]
+ ["Expand \"...\"s" backtrace-expand-ellipses
+ :help "Expand all the abbreviated forms in the current frame"]
+ ["Show on Multiple Lines" backtrace-multi-line
+ :help "Use line breaks and indentation to make a form more readable"]
+ ["Show on Single Line" backtrace-single-line]
+ "--"
+ ["Go to Source" backtrace-goto-source
+ :active (and (backtrace-get-index)
+ (plist-get (backtrace-frame-flags
+ (nth (backtrace-get-index) backtrace-frames))
+ :source-available))
+ :help "Show the source code for the current frame"]
+ ["Help for Symbol" backtrace-help-follow-symbol
+ :help "Show help for symbol at point"]
+ ["Describe Backtrace Mode" describe-mode
+ :help "Display documentation for backtrace-mode"]))
+ map)
+ "Local keymap for `backtrace-mode' buffers.")
+
+(defconst backtrace--flags-width 2
+ "Width in characters of the flags for a backtrace frame.")
+
+;;; Navigation and Text Properties
+
+;; This mode uses the following text properties:
+;; backtrace-index: The index into the buffer-local variable
+;; `backtrace-frames' for the frame at point, or nil if outside of a
+;; frame (in the buffer header).
+;; backtrace-view: A plist describing how the frame is printed. See
+;; the docstring for the buffer-local variable `backtrace-view.
+;; backtrace-section: The part of a frame which point is in. Either
+;; `func' or `locals'. At the moment just used to show and hide the
+;; local variables. Derived modes which do additional printing
+;; could define their own frame sections.
+;; backtrace-form: A value applied to each printed representation of a
+;; top-level s-expression, which needs to be different for sexps
+;; printed adjacent to each other, so the limits can be quickly
+;; found for pretty-printing.
+
+(defsubst backtrace-get-index (&optional pos)
+ "Return the index of the backtrace frame at POS.
+The value is an index into `backtrace-frames', or nil.
+POS, if omitted or nil, defaults to point."
+ (get-text-property (or pos (point)) 'backtrace-index))
+
+(defsubst backtrace-get-section (&optional pos)
+ "Return the section of a backtrace frame at POS.
+POS, if omitted or nil, defaults to point."
+ (get-text-property (or pos (point)) 'backtrace-section))
+
+(defsubst backtrace-get-view (&optional pos)
+ "Return the view plist of the backtrace frame at POS.
+POS, if omitted or nil, defaults to point."
+ (get-text-property (or pos (point)) 'backtrace-view))
+
+(defsubst backtrace-get-form (&optional pos)
+ "Return the backtrace form data for the form printed at POS.
+POS, if omitted or nil, defaults to point."
+ (get-text-property (or pos (point)) 'backtrace-form))
+
+(defun backtrace-get-frame-start (&optional pos)
+ "Return the beginning position of the frame at POS in the buffer.
+POS, if omitted or nil, defaults to point."
+ (let ((posn (or pos (point))))
+ (if (or (= (point-min) posn)
+ (not (eq (backtrace-get-index posn)
+ (backtrace-get-index (1- posn)))))
+ posn
+ (previous-single-property-change posn 'backtrace-index nil
(point-min)))))
+
+(defun backtrace-get-frame-end (&optional pos)
+ "Return the position of the end of the frame at POS in the buffer.
+POS, if omitted or nil, defaults to point."
+ (next-single-property-change (or pos (point))
+ 'backtrace-index nil (point-max)))
+
+(defun backtrace-forward-frame ()
+ "Move forward to the beginning of the next frame."
+ (interactive)
+ (let ((max (backtrace-get-frame-end)))
+ (when (= max (point-max))
+ (user-error "No next stack frame"))
+ (goto-char max)))
+
+(defun backtrace-backward-frame ()
+ "Move backward to the start of a stack frame."
+ (interactive)
+ (let ((current-index (backtrace-get-index))
+ (min (backtrace-get-frame-start)))
+ (if (or (and (/= (point) (point-max)) (null current-index))
+ (= min (point-min))
+ (and (= min (point))
+ (null (backtrace-get-index (1- min)))))
+ (user-error "No previous stack frame"))
+ (if (= min (point))
+ (goto-char (backtrace-get-frame-start (1- min)))
+ (goto-char min))))
+
+;; Other Backtrace mode commands
+
+(defun backtrace-revert (&rest _ignored)
+ "The `revert-buffer-function' for `backtrace-mode'.
+It runs `backtrace-revert-hook', then calls `backtrace-print'."
+ (interactive)
+ (unless (derived-mode-p 'backtrace-mode)
+ (error "The current buffer is not in Backtrace mode"))
+ (run-hooks 'backtrace-revert-hook)
+ (backtrace-print t))
+
+(defmacro backtrace--with-output-variables (view &rest body)
+ "Bind output variables according to VIEW and execute BODY."
+ (declare (indent 1))
+ `(let ((print-escape-control-characters t)
+ (print-escape-newlines t)
+ (print-circle (plist-get ,view :print-circle))
+ (standard-output (current-buffer)))
+ ,@body))
+
+(defun backtrace-toggle-locals (&optional all)
+ "Toggle the display of local variables for the backtrace frame at point.
+With prefix argument ALL, toggle the value of :show-locals in
+`backtrace-view', which affects all of the backtrace frames in
+the buffer."
+ (interactive "P")
+ (if all
+ (let ((pos (make-marker))
+ (visible (not (plist-get backtrace-view :show-locals))))
+ (setq backtrace-view (plist-put backtrace-view :show-locals visible))
+ (set-marker-insertion-type pos t)
+ (set-marker pos (point))
+ (goto-char (point-min))
+ ;; Skip the header.
+ (unless (backtrace-get-index)
+ (goto-char (backtrace-get-frame-end)))
+ (while (< (point) (point-max))
+ (backtrace--set-frame-locals-visible visible)
+ (goto-char (backtrace-get-frame-end)))
+ (goto-char pos)
+ (when (invisible-p pos)
+ (goto-char (backtrace-get-frame-start))))
+ (let ((index (backtrace-get-index)))
+ (unless index
+ (user-error "Not in a stack frame"))
+ (backtrace--set-frame-locals-visible
+ (not (plist-get (backtrace-get-view) :show-locals))))))
+
+(defun backtrace--set-frame-locals-visible (visible)
+ "Set the visibility of the local vars for the frame at point to VISIBLE."
+ (let ((pos (point))
+ (index (backtrace-get-index))
+ (start (backtrace-get-frame-start))
+ (end (backtrace-get-frame-end))
+ (view (copy-sequence (backtrace-get-view)))
+ (inhibit-read-only t))
+ (setq view (plist-put view :show-locals visible))
+ (goto-char (backtrace-get-frame-start))
+ (while (not (or (= (point) end)
+ (eq (backtrace-get-section) 'locals)))
+ (goto-char (next-single-property-change (point)
+ 'backtrace-section nil end)))
+ (cond
+ ((and (= (point) end) visible)
+ ;; The locals section doesn't exist so create it.
+ (let ((standard-output (current-buffer)))
+ (backtrace--with-output-variables view
+ (backtrace--print-locals
+ (nth index backtrace-frames) view))
+ (add-text-properties end (point) `(backtrace-index ,index))
+ (goto-char pos)))
+ ((/= (point) end)
+ ;; The locals section does exist, so add or remove the overlay.
+ (backtrace--set-locals-visible-overlay (point) end visible)
+ (goto-char (if (invisible-p pos) start pos))))
+ (add-text-properties start (backtrace-get-frame-end)
+ `(backtrace-view ,view))))
+
+(defun backtrace--set-locals-visible-overlay (beg end visible)
+ (backtrace--change-button-skip beg end (not visible))
+ (if visible
+ (remove-overlays beg end 'invisible t)
+ (let ((o (make-overlay beg end)))
+ (overlay-put o 'invisible t)
+ (overlay-put o 'evaporate t))))
+
+(defun backtrace--change-button-skip (beg end value)
+ "Change the skip property on all buttons between BEG and END.
+Set it to VALUE unless the button is a `backtrace-ellipsis' button."
+ (let ((inhibit-read-only t))
+ (setq beg (next-button beg))
+ (while (and beg (< beg end))
+ (unless (eq (button-type beg) 'backtrace-ellipsis)
+ (button-put beg 'skip value))
+ (setq beg (next-button beg)))))
+
+(defun backtrace-toggle-print-circle (&optional all)
+ "Toggle `print-circle' for the backtrace frame at point.
+With prefix argument ALL, toggle the value of :print-circle in
+`backtrace-view', which affects all of the backtrace frames in
+the buffer."
+ (interactive "P")
+ (backtrace--toggle-feature :print-circle all))
+
+(defun backtrace--toggle-feature (feature all)
+ "Toggle FEATURE for the current backtrace frame or for the buffer.
+FEATURE should be one of the options in `backtrace-view'. If ALL
+is non-nil, toggle FEATURE for all frames in the buffer. After
+toggling the feature, reprint the affected frame(s). Afterwards
+position point at the start of the frame it was in before."
+ (if all
+ (let ((index (backtrace-get-index))
+ (pos (point))
+ (at-end (= (point) (point-max)))
+ (value (not (plist-get backtrace-view feature))))
+ (setq backtrace-view (plist-put backtrace-view feature value))
+ (goto-char (point-min))
+ ;; Skip the header.
+ (unless (backtrace-get-index)
+ (goto-char (backtrace-get-frame-end)))
+ (while (< (point) (point-max))
+ (backtrace--set-feature feature value)
+ (goto-char (backtrace-get-frame-end)))
+ (if (not index)
+ (goto-char (if at-end (point-max) pos))
+ (goto-char (point-min))
+ (while (and (not (eql index (backtrace-get-index)))
+ (< (point) (point-max)))
+ (goto-char (backtrace-get-frame-end)))))
+ (let ((index (backtrace-get-index)))
+ (unless index
+ (user-error "Not in a stack frame"))
+ (backtrace--set-feature feature
+ (not (plist-get (backtrace-get-view)
feature))))))
+
+(defun backtrace--set-feature (feature value)
+ "Set FEATURE in the view plist of the frame at point to VALUE.
+Reprint the frame with the new view plist."
+ (let ((inhibit-read-only t)
+ (view (copy-sequence (backtrace-get-view)))
+ (index (backtrace-get-index))
+ (min (backtrace-get-frame-start))
+ (max (backtrace-get-frame-end)))
+ (setq view (plist-put view feature value))
+ (delete-region min max)
+ (goto-char min)
+ (backtrace-print-frame (nth index backtrace-frames) view)
+ (add-text-properties min (point)
+ `(backtrace-index ,index backtrace-view ,view))
+ (goto-char min)))
+
+(defun backtrace-expand-ellipsis (button)
+ "Expand display of the elided form at BUTTON."
+ (interactive)
+ (goto-char (button-start button))
+ (unless (get-text-property (point) 'cl-print-ellipsis)
+ (if (and (> (point) (point-min))
+ (get-text-property (1- (point)) 'cl-print-ellipsis))
+ (backward-char)
+ (user-error "No ellipsis to expand here")))
+ (let* ((end (next-single-property-change (point) 'cl-print-ellipsis))
+ (begin (previous-single-property-change end 'cl-print-ellipsis))
+ (value (get-text-property begin 'cl-print-ellipsis))
+ (props (backtrace-get-text-properties begin))
+ (inhibit-read-only t))
+ (backtrace--with-output-variables (backtrace-get-view)
+ (delete-region begin end)
+ (insert (cl-print-to-string-with-limit #'cl-print-expand-ellipsis value
+ backtrace-line-length))
+ (setq end (point))
+ (goto-char begin)
+ (while (< (point) end)
+ (let ((next (next-single-property-change (point) 'cl-print-ellipsis
+ nil end)))
+ (when (get-text-property (point) 'cl-print-ellipsis)
+ (make-text-button (point) next :type 'backtrace-ellipsis))
+ (goto-char next)))
+ (goto-char begin)
+ (add-text-properties begin end props))))
+
+(defun backtrace-expand-ellipses (&optional no-limit)
+ "Expand display of all \"...\"s in the backtrace frame at point.
+\\<backtrace-mode-map>
+Each ellipsis will be limited to `backtrace-line-length'
+characters in its expansion. With optional prefix argument
+NO-LIMIT, do not limit the number of characters. Note that with
+or without the argument, using this command can result in very
+long lines and very poor display performance. If this happens
+and is a problem, use `\\[revert-buffer]' to return to the
+initial state of the Backtrace buffer."
+ (interactive "P")
+ (save-excursion
+ (let ((start (backtrace-get-frame-start))
+ (end (backtrace-get-frame-end))
+ (backtrace-line-length (unless no-limit backtrace-line-length)))
+ (goto-char end)
+ (while (> (point) start)
+ (let ((next (previous-single-property-change (point) 'cl-print-ellipsis
+ nil start)))
+ (when (get-text-property (point) 'cl-print-ellipsis)
+ (push-button (point)))
+ (goto-char next))))))
+
+(defun backtrace-multi-line ()
+ "Show the top level s-expression at point on multiple lines with
indentation."
+ (interactive)
+ (backtrace--reformat-sexp #'backtrace--multi-line))
+
+(defun backtrace--multi-line ()
+ "Pretty print the current buffer, then remove the trailing newline."
+ (set-syntax-table emacs-lisp-mode-syntax-table)
+ (pp-buffer)
+ (goto-char (1- (point-max)))
+ (delete-char 1))
+
+(defun backtrace-single-line ()
+ "Show the top level s-expression at point on one line."
+ (interactive)
+ (backtrace--reformat-sexp #'backtrace--single-line))
+
+(defun backtrace--single-line ()
+ "Replace line breaks and following indentation with spaces.
+Works on the current buffer."
+ (goto-char (point-min))
+ (while (re-search-forward "\n[[:blank:]]*" nil t)
+ (replace-match " ")))
+
+(defun backtrace--reformat-sexp (format-function)
+ "Reformat the top level sexp at point.
+Locate the top level sexp at or following point on the same line,
+and reformat it with FORMAT-FUNCTION, preserving the location of
+point within the sexp. If no sexp is found before the end of
+the line or buffer, signal an error.
+
+FORMAT-FUNCTION will be called without arguments, with the
+current buffer set to a temporary buffer containing only the
+content of the sexp."
+ (let* ((orig-pos (point))
+ (pos (point))
+ (tag (backtrace-get-form pos))
+ (end (next-single-property-change pos 'backtrace-form))
+ (begin (previous-single-property-change end 'backtrace-form
+ nil (point-min))))
+ (unless tag
+ (when (or (= end (point-max)) (> end (point-at-eol)))
+ (user-error "No form here to reformat"))
+ (goto-char end)
+ (setq pos end
+ end (next-single-property-change pos 'backtrace-form)
+ begin (previous-single-property-change end 'backtrace-form
+ nil (point-min))))
+ (let* ((offset (when (>= orig-pos begin) (- orig-pos begin)))
+ (offset-marker (when offset (make-marker)))
+ (content (buffer-substring begin end))
+ (props (backtrace-get-text-properties begin))
+ (inhibit-read-only t))
+ (delete-region begin end)
+ (insert (with-temp-buffer
+ (insert content)
+ (when offset
+ (set-marker-insertion-type offset-marker t)
+ (set-marker offset-marker (+ (point-min) offset)))
+ (funcall format-function)
+ (when offset
+ (setq offset (- (marker-position offset-marker)
(point-min))))
+ (buffer-string)))
+ (when offset
+ (set-marker offset-marker (+ begin offset)))
+ (save-excursion
+ (goto-char begin)
+ (indent-sexp))
+ (add-text-properties begin (point) props)
+ (if offset
+ (goto-char (marker-position offset-marker))
+ (goto-char orig-pos)))))
+
+(defun backtrace-get-text-properties (pos)
+ "Return a plist of backtrace-mode's text properties at POS."
+ (apply #'append
+ (mapcar (lambda (prop)
+ (list prop (get-text-property pos prop)))
+ '(backtrace-section backtrace-index backtrace-view
+ backtrace-form))))
+
+(defun backtrace-goto-source ()
+ "If its location is known, jump to the source code for the frame at point."
+ (interactive)
+ (let* ((index (or (backtrace-get-index) (user-error "Not in a stack frame")))
+ (frame (nth index backtrace-frames))
+ (source-available (plist-get (backtrace-frame-flags frame)
+ :source-available)))
+ (unless (and source-available
+ (catch 'done
+ (dolist (func backtrace-goto-source-functions)
+ (when (funcall func)
+ (throw 'done t)))))
+ (user-error "Source code location not known"))))
+
+(defun backtrace-help-follow-symbol (&optional pos)
+ "Follow cross-reference at POS, defaulting to point.
+For the cross-reference format, see `help-make-xrefs'."
+ (interactive "d")
+ (unless pos
+ (setq pos (point)))
+ (unless (push-button pos)
+ ;; Check if the symbol under point is a function or variable.
+ (let ((sym
+ (intern
+ (save-excursion
+ (goto-char pos) (skip-syntax-backward "w_")
+ (buffer-substring (point)
+ (progn (skip-syntax-forward "w_")
+ (point)))))))
+ (when (or (boundp sym) (fboundp sym) (facep sym))
+ (describe-symbol sym)))))
+
+;; Print backtrace frames
+
+(defun backtrace-print (&optional remember-pos)
+ "Populate the current Backtrace mode buffer.
+This erases the buffer and inserts printed representations of the
+frames. Optional argument REMEMBER-POS, if non-nil, means to
+move point to the entry with the same ID element as the current
+line and recenter window line accordingly."
+ (let ((inhibit-read-only t)
+ entry-index saved-pt window-line)
+ (and remember-pos
+ (setq entry-index (backtrace-get-index))
+ (when (eq (window-buffer) (current-buffer))
+ (setq window-line
+ (count-screen-lines (window-start) (point)))))
+ (erase-buffer)
+ (when backtrace-insert-header-function
+ (funcall backtrace-insert-header-function))
+ (dotimes (idx (length backtrace-frames))
+ (let ((beg (point))
+ (elt (nth idx backtrace-frames)))
+ (and entry-index
+ (equal entry-index idx)
+ (setq entry-index nil
+ saved-pt (point)))
+ (backtrace-print-frame elt backtrace-view)
+ (add-text-properties
+ beg (point)
+ `(backtrace-index ,idx backtrace-view ,backtrace-view))))
+ (set-buffer-modified-p nil)
+ ;; If REMEMBER-POS was specified, move to the "old" location.
+ (if saved-pt
+ (progn (goto-char saved-pt)
+ (when window-line
+ (recenter window-line)))
+ (goto-char (point-min)))))
+
+;; Define button type used for ...'s.
+;; Set skip property so you don't have to TAB through 100 of them to
+;; get to the next function name.
+(define-button-type 'backtrace-ellipsis
+ 'skip t 'action #'backtrace-expand-ellipsis
+ 'help-echo "mouse-2, RET: expand this ellipsis")
+
+(defun backtrace-print-to-string (obj &optional limit)
+ "Return a printed representation of OBJ formatted for backtraces.
+Attempt to get the length of the returned string under LIMIT
+charcters with appropriate settings of `print-level' and
+`print-length.' LIMIT defaults to `backtrace-line-length'."
+ (backtrace--with-output-variables backtrace-view
+ (backtrace--print-to-string obj limit)))
+
+(defun backtrace--print-to-string (sexp &optional limit)
+ ;; This is for use by callers who wrap the call with
+ ;; backtrace--with-output-variables.
+ (setq limit (or limit backtrace-line-length))
+ (with-temp-buffer
+ (insert (cl-print-to-string-with-limit #'backtrace--print sexp limit))
+ ;; Add a unique backtrace-form property.
+ (put-text-property (point-min) (point) 'backtrace-form (gensym))
+ ;; Make buttons from all the "..."s. Since there might be many of
+ ;; them, use text property buttons.
+ (goto-char (point-min))
+ (while (< (point) (point-max))
+ (let ((end (next-single-property-change (point) 'cl-print-ellipsis
+ nil (point-max))))
+ (when (get-text-property (point) 'cl-print-ellipsis)
+ (make-text-button (point) end :type 'backtrace-ellipsis))
+ (goto-char end)))
+ (buffer-string)))
+
+(defun backtrace-print-frame (frame view)
+ "Insert a backtrace FRAME at point formatted according to VIEW.
+Tag the sections of the frame with the `backtrace-section' text
+property for use by navigation."
+ (backtrace--with-output-variables view
+ (backtrace--print-flags frame view)
+ (backtrace--print-func-and-args frame view)
+ (backtrace--print-locals frame view)))
+
+(defun backtrace--print-flags (frame view)
+ "Print the flags of a backtrace FRAME if enabled in VIEW."
+ (let ((beg (point))
+ (flag (plist-get (backtrace-frame-flags frame) :debug-on-exit))
+ (source (plist-get (backtrace-frame-flags frame) :source-available)))
+ (when (plist-get view :show-flags)
+ (when source (insert ">"))
+ (when flag (insert "*")))
+ (insert (make-string (- backtrace--flags-width (- (point) beg)) ?\s))
+ (put-text-property beg (point) 'backtrace-section 'func)))
+
+(defun backtrace--print-func-and-args (frame _view)
+ "Print the function, arguments and buffer position of a backtrace FRAME.
+Format it according to VIEW."
+ (let* ((beg (point))
+ (evald (backtrace-frame-evald frame))
+ (fun (backtrace-frame-fun frame))
+ (args (backtrace-frame-args frame))
+ (def (and (symbolp fun) (fboundp fun) (symbol-function fun)))
+ (fun-file (or (symbol-file fun 'defun)
+ (and (subrp def)
+ (not (eq 'unevalled (cdr (subr-arity def))))
+ (find-lisp-object-file-name fun def))))
+ (fun-pt (point)))
+ (cond
+ ((and evald (not debugger-stack-frame-as-list))
+ (if (atom fun)
+ (funcall backtrace-print-function fun)
+ (insert
+ (backtrace--print-to-string fun (when args (/ backtrace-line-length
2)))))
+ (if args
+ (insert (backtrace--print-to-string
+ args (max (truncate (/ backtrace-line-length 5))
+ (- backtrace-line-length (- (point) beg)))))
+ ;; The backtrace-form property is so that backtrace-multi-line
+ ;; will find it. backtrace-multi-line doesn't do anything
+ ;; useful with it, just being consistent.
+ (let ((start (point)))
+ (insert "()")
+ (put-text-property start (point) 'backtrace-form t))))
+ (t
+ (let ((fun-and-args (cons fun args)))
+ (insert (backtrace--print-to-string fun-and-args)))
+ (cl-incf fun-pt)))
+ (when fun-file
+ (make-text-button fun-pt (+ fun-pt (length (symbol-name fun)))
+ :type 'help-function-def
+ 'help-args (list fun fun-file)))
+ ;; After any frame that uses eval-buffer, insert a comment that
+ ;; states the buffer position it's reading at.
+ (when (backtrace-frame-pos frame)
+ (insert " ; Reading at ")
+ (let ((pos (point)))
+ (insert (format "buffer position %d" (backtrace-frame-pos frame)))
+ (make-button pos (point) :type 'backtrace-buffer-pos
+ 'backtrace-buffer (backtrace-frame-buffer frame)
+ 'backtrace-pos (backtrace-frame-pos frame))))
+ (insert "\n")
+ (put-text-property beg (point) 'backtrace-section 'func)))
+
+(defun backtrace--print-locals (frame view)
+ "Print a backtrace FRAME's local variables according to VIEW.
+Print them only if :show-locals is non-nil in the VIEW plist."
+ (when (plist-get view :show-locals)
+ (let* ((beg (point))
+ (locals (backtrace-frame-locals frame)))
+ (if (null locals)
+ (insert " [no locals]\n")
+ (pcase-dolist (`(,symbol . ,value) locals)
+ (insert " ")
+ (backtrace--print symbol)
+ (insert " = ")
+ (insert (backtrace--print-to-string value))
+ (insert "\n")))
+ (put-text-property beg (point) 'backtrace-section 'locals))))
+
+(defun backtrace--print (obj &optional stream)
+ "Attempt to print OBJ to STREAM using `backtrace-print-function'.
+Fall back to `prin1' if there is an error."
+ (condition-case err
+ (funcall backtrace-print-function obj stream)
+ (error
+ (message "Error in backtrace printer: %S" err)
+ (prin1 obj stream))))
+
+(defun backtrace-update-flags ()
+ "Update the display of the flags in the backtrace frame at point."
+ (let ((view (backtrace-get-view))
+ (begin (backtrace-get-frame-start)))
+ (when (plist-get view :show-flags)
+ (save-excursion
+ (goto-char begin)
+ (let ((props (backtrace-get-text-properties begin))
+ (inhibit-read-only t)
+ (standard-output (current-buffer)))
+ (delete-char backtrace--flags-width)
+ (backtrace--print-flags (nth (backtrace-get-index) backtrace-frames)
+ view)
+ (add-text-properties begin (point) props))))))
+
+(defun backtrace--filter-visible (beg end &optional _delete)
+ "Return the visible text between BEG and END."
+ (let ((result ""))
+ (while (< beg end)
+ (let ((next (next-single-char-property-change beg 'invisible)))
+ (unless (get-char-property beg 'invisible)
+ (setq result (concat result (buffer-substring beg (min end next)))))
+ (setq beg next)))
+ result))
+
+;;; The mode definition
+
+(define-derived-mode backtrace-mode special-mode "Backtrace"
+ "Generic major mode for examining an Elisp stack backtrace.
+This mode can be used directly, or other major modes can be
+derived from it, using `define-derived-mode'.
+
+In this major mode, the buffer contains some optional lines of
+header text followed by backtrace frames, each consisting of one
+or more whole lines.
+
+Letters in this mode do not insert themselves; instead they are
+commands.
+\\<backtrace-mode-map>
+\\{backtrace-mode-map}
+
+A mode which inherits from Backtrace mode, or a command which
+creates a backtrace-mode buffer, should usually do the following:
+
+ - Set `backtrace-revert-hook', if the buffer contents need
+ to be specially recomputed prior to `revert-buffer'.
+ - Maybe set `backtrace-insert-header-function' to a function to create
+ header text for the buffer.
+ - Set `backtrace-frames' (see below).
+ - Maybe modify `backtrace-view' (see below).
+ - Maybe set `backtrace-print-function'.
+
+A command which creates or switches to a Backtrace mode buffer,
+such as `ert-results-pop-to-backtrace-for-test-at-point', should
+initialize `backtrace-frames' to a list of `backtrace-frame'
+objects (`backtrace-get-frames' is provided for that purpose, if
+desired), and may optionally modify `backtrace-view', which is a
+plist describing the appearance of the backtrace. Finally, it
+should call `backtrace-print'.
+
+`backtrace-print' calls `backtrace-insert-header-function'
+followed by `backtrace-print-frame', once for each stack frame."
+ :syntax-table emacs-lisp-mode-syntax-table
+ (when backtrace-fontify
+ (setq font-lock-defaults
+ '((backtrace-font-lock-keywords
+ backtrace-font-lock-keywords-1
+ backtrace-font-lock-keywords-2)
+ nil nil nil nil
+ (font-lock-syntactic-face-function
+ . lisp-font-lock-syntactic-face-function))))
+ (setq truncate-lines t)
+ (buffer-disable-undo)
+ ;; In debug.el, from 1998 to 2009 this was set to nil, reason stated
+ ;; was because of bytecode. Since 2009 it's been set to t, but the
+ ;; default is t so I think this isn't necessary.
+ ;; (set-buffer-multibyte t)
+ (setq-local revert-buffer-function #'backtrace-revert)
+ (setq-local filter-buffer-substring-function #'backtrace--filter-visible)
+ (setq-local indent-line-function 'lisp-indent-line)
+ (setq-local indent-region-function 'lisp-indent-region)
+ (add-hook 'xref-backend-functions #'backtrace--xref-backend nil t))
+
+(put 'backtrace-mode 'mode-class 'special)
+
+;;; Backtrace printing
+
+;;;###autoload
+(defun backtrace ()
+ "Print a trace of Lisp function calls currently active.
+Output stream used is value of `standard-output'."
+ (princ (backtrace-to-string (backtrace-get-frames 'backtrace)))
+ nil)
+
+(defun backtrace-to-string(&optional frames)
+ "Format FRAMES, a list of `backtrace-frame' objects, for output.
+Return the result as a string. If FRAMES is nil, use all
+function calls currently active."
+ (unless frames (setq frames (backtrace-get-frames 'backtrace-to-string)))
+ (let ((backtrace-fontify nil))
+ (with-temp-buffer
+ (backtrace-mode)
+ (setq backtrace-view '(:show-flags t)
+ backtrace-frames frames
+ backtrace-print-function #'cl-prin1)
+ (backtrace-print)
+ (substring-no-properties (filter-buffer-substring (point-min)
+ (point-max))))))
+
+(provide 'backtrace)
+
+;;; backtrace.el ends here
diff --git a/lisp/emacs-lisp/cl-macs.el b/lisp/emacs-lisp/cl-macs.el
index 011965a..d0d1c3b 100644
--- a/lisp/emacs-lisp/cl-macs.el
+++ b/lisp/emacs-lisp/cl-macs.el
@@ -2083,10 +2083,7 @@ This is like `cl-flet', but for macros instead of
functions.
\(fn ((NAME ARGLIST BODY...) ...) FORM...)"
(declare (indent 1)
- (debug
- ((&rest (&define name (&rest arg) cl-declarations-or-string
- def-body))
- cl-declarations body)))
+ (debug (cl-macrolet-expr)))
(if (cdr bindings)
`(cl-macrolet (,(car bindings)) (cl-macrolet ,(cdr bindings) ,@body))
(if (null bindings) (macroexp-progn body)
diff --git a/lisp/emacs-lisp/cl-print.el b/lisp/emacs-lisp/cl-print.el
index 1eae8fa..c63f5ac 100644
--- a/lisp/emacs-lisp/cl-print.el
+++ b/lisp/emacs-lisp/cl-print.el
@@ -55,10 +55,19 @@ call other entry points instead, such as `cl-prin1'."
;; we should only use it for objects which don't have nesting.
(prin1 object stream))
+(cl-defgeneric cl-print-object-contents (_object _start _stream)
+ "Dispatcher to print the contents of OBJECT on STREAM.
+Print the contents starting with the item at START, without
+delimiters."
+ ;; Every cl-print-object method which can print an ellipsis should
+ ;; have a matching cl-print-object-contents method to expand an
+ ;; ellipsis.
+ (error "Missing cl-print-object-contents method"))
+
(cl-defmethod cl-print-object ((object cons) stream)
(if (and cl-print--depth (natnump print-level)
(> cl-print--depth print-level))
- (princ "..." stream)
+ (cl-print-insert-ellipsis object 0 stream)
(let ((car (pop object))
(count 1))
(if (and print-quoted
@@ -84,23 +93,60 @@ call other entry points instead, such as `cl-prin1'."
(princ " " stream)
(if (or (not (natnump print-length)) (> print-length count))
(cl-print-object (pop object) stream)
- (princ "..." stream)
+ (cl-print-insert-ellipsis object print-length stream)
(setq object nil))
(cl-incf count))
(when object
(princ " . " stream) (cl-print-object object stream))
(princ ")" stream)))))
+(cl-defmethod cl-print-object-contents ((object cons) _start stream)
+ (let ((count 0))
+ (while (and (consp object)
+ (not (cond
+ (cl-print--number-table
+ (numberp (gethash object cl-print--number-table)))
+ ((memq object cl-print--currently-printing))
+ (t (push object cl-print--currently-printing)
+ nil))))
+ (unless (zerop count)
+ (princ " " stream))
+ (if (or (not (natnump print-length)) (> print-length count))
+ (cl-print-object (pop object) stream)
+ (cl-print-insert-ellipsis object print-length stream)
+ (setq object nil))
+ (cl-incf count))
+ (when object
+ (princ " . " stream) (cl-print-object object stream))))
+
(cl-defmethod cl-print-object ((object vector) stream)
- (princ "[" stream)
- (let ((count (length object)))
- (dotimes (i (if (natnump print-length)
- (min print-length count) count))
- (unless (zerop i) (princ " " stream))
- (cl-print-object (aref object i) stream))
- (when (and (natnump print-length) (< print-length count))
- (princ " ..." stream)))
- (princ "]" stream))
+ (if (and cl-print--depth (natnump print-level)
+ (> cl-print--depth print-level))
+ (cl-print-insert-ellipsis object 0 stream)
+ (princ "[" stream)
+ (let* ((len (length object))
+ (limit (if (natnump print-length)
+ (min print-length len) len)))
+ (dotimes (i limit)
+ (unless (zerop i) (princ " " stream))
+ (cl-print-object (aref object i) stream))
+ (when (< limit len)
+ (princ " " stream)
+ (cl-print-insert-ellipsis object limit stream)))
+ (princ "]" stream)))
+
+(cl-defmethod cl-print-object-contents ((object vector) start stream)
+ (let* ((len (length object))
+ (limit (if (natnump print-length)
+ (min (+ start print-length) len) len))
+ (i start))
+ (while (< i limit)
+ (unless (= i start) (princ " " stream))
+ (cl-print-object (aref object i) stream)
+ (cl-incf i))
+ (when (< limit len)
+ (princ " " stream)
+ (cl-print-insert-ellipsis object limit stream))))
(cl-defmethod cl-print-object ((object hash-table) stream)
(princ "#<hash-table " stream)
@@ -109,7 +155,7 @@ call other entry points instead, such as `cl-prin1'."
(princ (hash-table-count object) stream)
(princ "/" stream)
(princ (hash-table-size object) stream)
- (princ (format " 0x%x" (sxhash object)) stream)
+ (princ (format " %#x" (sxhash object)) stream)
(princ ">" stream))
(define-button-type 'help-byte-code
@@ -166,7 +212,7 @@ into a button whose action shows the function's
disassembly.")
(let ((button-start (and cl-print-compiled-button
(bufferp stream)
(with-current-buffer stream (point)))))
- (princ (format "#<bytecode 0x%x>" (sxhash object)) stream)
+ (princ (format "#<bytecode %#x>" (sxhash object)) stream)
(when (eq cl-print-compiled 'static)
(princ " " stream)
(cl-print-object (aref object 2) stream))
@@ -199,21 +245,135 @@ into a button whose action shows the function's
disassembly.")
(princ ")" stream)))
(cl-defmethod cl-print-object ((object cl-structure-object) stream)
- (princ "#s(" stream)
+ (if (and cl-print--depth (natnump print-level)
+ (> cl-print--depth print-level))
+ (cl-print-insert-ellipsis object 0 stream)
+ (princ "#s(" stream)
+ (let* ((class (cl-find-class (type-of object)))
+ (slots (cl--struct-class-slots class))
+ (len (length slots))
+ (limit (if (natnump print-length)
+ (min print-length len) len)))
+ (princ (cl--struct-class-name class) stream)
+ (dotimes (i limit)
+ (let ((slot (aref slots i)))
+ (princ " :" stream)
+ (princ (cl--slot-descriptor-name slot) stream)
+ (princ " " stream)
+ (cl-print-object (aref object (1+ i)) stream)))
+ (when (< limit len)
+ (princ " " stream)
+ (cl-print-insert-ellipsis object limit stream)))
+ (princ ")" stream)))
+
+(cl-defmethod cl-print-object-contents ((object cl-structure-object) start
stream)
(let* ((class (cl-find-class (type-of object)))
(slots (cl--struct-class-slots class))
- (count (length slots)))
- (princ (cl--struct-class-name class) stream)
- (dotimes (i (if (natnump print-length)
- (min print-length count) count))
+ (len (length slots))
+ (limit (if (natnump print-length)
+ (min (+ start print-length) len) len))
+ (i start))
+ (while (< i limit)
(let ((slot (aref slots i)))
- (princ " :" stream)
+ (unless (= i start) (princ " " stream))
+ (princ ":" stream)
(princ (cl--slot-descriptor-name slot) stream)
(princ " " stream)
- (cl-print-object (aref object (1+ i)) stream)))
- (when (and (natnump print-length) (< print-length count))
- (princ " ..." stream)))
- (princ ")" stream))
+ (cl-print-object (aref object (1+ i)) stream))
+ (cl-incf i))
+ (when (< limit len)
+ (princ " " stream)
+ (cl-print-insert-ellipsis object limit stream))))
+
+(cl-defmethod cl-print-object ((object string) stream)
+ (unless stream (setq stream standard-output))
+ (let* ((has-properties (or (text-properties-at 0 object)
+ (next-property-change 0 object)))
+ (len (length object))
+ (limit (if (natnump print-length) (min print-length len) len)))
+ (if (and has-properties
+ cl-print--depth
+ (natnump print-level)
+ (> cl-print--depth print-level))
+ (cl-print-insert-ellipsis object 0 stream)
+ ;; Print all or part of the string
+ (when has-properties
+ (princ "#(" stream))
+ (if (= limit len)
+ (prin1 (if has-properties (substring-no-properties object) object)
+ stream)
+ (let ((part (concat (substring-no-properties object 0 limit) "...")))
+ (prin1 part stream)
+ (when (bufferp stream)
+ (with-current-buffer stream
+ (cl-print-propertize-ellipsis object limit
+ (- (point) 4)
+ (- (point) 1) stream)))))
+ ;; Print the property list.
+ (when has-properties
+ (let* ((interval-limit (and (natnump print-length)
+ (max 1 (/ print-length 3))))
+ (interval-count 0)
+ (start-pos (if (text-properties-at 0 object)
+ 0 (next-property-change 0 object)))
+ (end-pos (next-property-change start-pos object len)))
+ (while (and (or (null interval-limit)
+ (< interval-count interval-limit))
+ (< start-pos len))
+ (let ((props (text-properties-at start-pos object)))
+ (when props
+ (princ " " stream) (princ start-pos stream)
+ (princ " " stream) (princ end-pos stream)
+ (princ " " stream) (cl-print-object props stream)
+ (cl-incf interval-count))
+ (setq start-pos end-pos
+ end-pos (next-property-change start-pos object len))))
+ (when (< start-pos len)
+ (princ " " stream)
+ (cl-print-insert-ellipsis object (list start-pos) stream)))
+ (princ ")" stream)))))
+
+(cl-defmethod cl-print-object-contents ((object string) start stream)
+ ;; If START is an integer, it is an index into the string, and the
+ ;; ellipsis that needs to be expanded is part of the string. If
+ ;; START is a cons, its car is an index into the string, and the
+ ;; ellipsis that needs to be expanded is in the property list.
+ (let* ((len (length object)))
+ (if (atom start)
+ ;; Print part of the string.
+ (let* ((limit (if (natnump print-length)
+ (min (+ start print-length) len) len))
+ (substr (substring-no-properties object start limit))
+ (printed (prin1-to-string substr))
+ (trimmed (substring printed 1 (1- (length printed)))))
+ (princ trimmed)
+ (when (< limit len)
+ (cl-print-insert-ellipsis object limit stream)))
+
+ ;; Print part of the property list.
+ (let* ((first t)
+ (interval-limit (and (natnump print-length)
+ (max 1 (/ print-length 3))))
+ (interval-count 0)
+ (start-pos (car start))
+ (end-pos (next-property-change start-pos object len)))
+ (while (and (or (null interval-limit)
+ (< interval-count interval-limit))
+ (< start-pos len))
+ (let ((props (text-properties-at start-pos object)))
+ (when props
+ (if first
+ (setq first nil)
+ (princ " " stream))
+ (princ start-pos stream)
+ (princ " " stream) (princ end-pos stream)
+ (princ " " stream) (cl-print-object props stream)
+ (cl-incf interval-count))
+ (setq start-pos end-pos
+ end-pos (next-property-change start-pos object len))))
+ (when (< start-pos len)
+ (princ " " stream)
+ (cl-print-insert-ellipsis object (list start-pos) stream))))))
;;; Circularity and sharing.
@@ -275,8 +435,17 @@ into a button whose action shows the function's
disassembly.")
(push cdr stack)
(push car stack))
((pred stringp)
- ;; We presumably won't print its text-properties.
- nil)
+ (let* ((len (length object))
+ (start (if (text-properties-at 0 object)
+ 0 (next-property-change 0 object)))
+ (end (and start
+ (next-property-change start object len))))
+ (while (and start (< start len))
+ (let ((props (text-properties-at start object)))
+ (when props
+ (push props stack))
+ (setq start end
+ end (next-property-change start object len))))))
((or (pred arrayp) (pred byte-code-function-p))
;; FIXME: Inefficient for char-tables!
(dotimes (i (length object))
@@ -291,6 +460,48 @@ into a button whose action shows the function's
disassembly.")
(cl-print--find-sharing object print-number-table)))
print-number-table))
+(defun cl-print-insert-ellipsis (object start stream)
+ "Print \"...\" to STREAM with the `cl-print-ellipsis' text property.
+Save state in the text property in order to print the elided part
+of OBJECT later. START should be 0 if the whole OBJECT is being
+elided, otherwise it should be an index or other pointer into the
+internals of OBJECT which can be passed to
+`cl-print-object-contents' at a future time."
+ (unless stream (setq stream standard-output))
+ (let ((ellipsis-start (and (bufferp stream)
+ (with-current-buffer stream (point)))))
+ (princ "..." stream)
+ (when ellipsis-start
+ (with-current-buffer stream
+ (cl-print-propertize-ellipsis object start ellipsis-start (point)
+ stream)))))
+
+(defun cl-print-propertize-ellipsis (object start beg end stream)
+ "Add the `cl-print-ellipsis' property between BEG and END.
+STREAM should be a buffer. OBJECT and START are as described in
+`cl-print-insert-ellipsis'."
+ (let ((value (list object start cl-print--number-table
+ cl-print--currently-printing)))
+ (with-current-buffer stream
+ (put-text-property beg end 'cl-print-ellipsis value stream))))
+
+;;;###autoload
+(defun cl-print-expand-ellipsis (value stream)
+ "Print the expansion of an ellipsis to STREAM.
+VALUE should be the value of the `cl-print-ellipsis' text property
+which was attached to the ellipsis by `cl-prin1'."
+ (let ((cl-print--depth 1)
+ (object (nth 0 value))
+ (start (nth 1 value))
+ (cl-print--number-table (nth 2 value))
+ (print-number-table (nth 2 value))
+ (cl-print--currently-printing (nth 3 value)))
+ (when (eq object (car cl-print--currently-printing))
+ (pop cl-print--currently-printing))
+ (if (equal start 0)
+ (cl-print-object object stream)
+ (cl-print-object-contents object start stream))))
+
;;;###autoload
(defun cl-prin1 (object &optional stream)
"Print OBJECT on STREAM according to its type.
@@ -313,5 +524,45 @@ node `(elisp)Output Variables'."
(cl-prin1 object (current-buffer))
(buffer-string)))
+;;;###autoload
+(defun cl-print-to-string-with-limit (print-function value limit)
+ "Return a string containing a printed representation of VALUE.
+Attempt to get the length of the returned string under LIMIT
+characters with appropriate settings of `print-level' and
+`print-length.' Use PRINT-FUNCTION to print, which should take
+the arguments VALUE and STREAM and which should respect
+`print-length' and `print-level'. LIMIT may be nil or zero in
+which case PRINT-FUNCTION will be called with `print-level' and
+`print-length' bound to nil.
+
+Use this function with `cl-prin1' to print an object,
+abbreviating it with ellipses to fit within a size limit. Use
+this function with `cl-prin1-expand-ellipsis' to expand an
+ellipsis, abbreviating the expansion to stay within a size
+limit."
+ (setq limit (and (natnump limit)
+ (not (zerop limit))
+ limit))
+ ;; Since this is used by the debugger when stack space may be
+ ;; limited, if you increase print-level here, add more depth in
+ ;; call_debugger (bug#31919).
+ (let* ((print-length (when limit (min limit 50)))
+ (print-level (when limit (min 8 (truncate (log limit)))))
+ (delta (when limit
+ (max 1 (truncate (/ print-length print-level))))))
+ (with-temp-buffer
+ (catch 'done
+ (while t
+ (erase-buffer)
+ (funcall print-function value (current-buffer))
+ ;; Stop when either print-level is too low or the value is
+ ;; successfully printed in the space allowed.
+ (when (or (not limit)
+ (< (- (point-max) (point-min)) limit)
+ (= print-level 2))
+ (throw 'done (buffer-string)))
+ (cl-decf print-level)
+ (cl-decf print-length delta))))))
+
(provide 'cl-print)
;;; cl-print.el ends here
diff --git a/lisp/emacs-lisp/debug.el b/lisp/emacs-lisp/debug.el
index 0efaa63..7fc2b41 100644
--- a/lisp/emacs-lisp/debug.el
+++ b/lisp/emacs-lisp/debug.el
@@ -28,6 +28,7 @@
;;; Code:
(require 'cl-lib)
+(require 'backtrace)
(require 'button)
(defgroup debugger nil
@@ -133,6 +134,25 @@ where CAUSE can be:
- exit: called because of exit of a flagged function.
- error: called because of `debug-on-error'.")
+(cl-defstruct (debugger--buffer-state
+ (:constructor debugger--save-buffer-state
+ (&aux (mode major-mode)
+ (header backtrace-insert-header-function)
+ (frames backtrace-frames)
+ (content (buffer-string))
+ (pos (point)))))
+ mode header frames content pos)
+
+(defun debugger--restore-buffer-state (state)
+ (unless (derived-mode-p (debugger--buffer-state-mode state))
+ (funcall (debugger--buffer-state-mode state)))
+ (setq backtrace-insert-header-function (debugger--buffer-state-header state)
+ backtrace-frames (debugger--buffer-state-frames state))
+ (let ((inhibit-read-only t))
+ (erase-buffer)
+ (insert (debugger--buffer-state-content state)))
+ (goto-char (debugger--buffer-state-pos state)))
+
;;;###autoload
(setq debugger 'debug)
;;;###autoload
@@ -174,7 +194,7 @@ first will be printed into the backtrace buffer."
(debugger-previous-state
(if (get-buffer "*Backtrace*")
(with-current-buffer (get-buffer "*Backtrace*")
- (list major-mode (buffer-string)))))
+ (debugger--save-buffer-state))))
(debugger-args args)
(debugger-buffer (get-buffer-create "*Backtrace*"))
(debugger-old-buffer (current-buffer))
@@ -236,7 +256,8 @@ first will be printed into the backtrace buffer."
(window-total-height debugger-window)))
(error nil)))
(setq debugger-previous-window debugger-window))
- (debugger-mode)
+ (unless (derived-mode-p 'debugger-mode)
+ (debugger-mode))
(debugger-setup-buffer debugger-args)
(when noninteractive
;; If the backtrace is long, save the beginning
@@ -280,15 +301,14 @@ first will be printed into the backtrace buffer."
(setq debugger-previous-window nil))
;; Restore previous state of debugger-buffer in case we were
;; in a recursive invocation of the debugger, otherwise just
- ;; erase the buffer and put it into fundamental mode.
+ ;; erase the buffer.
(when (buffer-live-p debugger-buffer)
(with-current-buffer debugger-buffer
- (let ((inhibit-read-only t))
- (erase-buffer)
- (if (null debugger-previous-state)
- (fundamental-mode)
- (insert (nth 1 debugger-previous-state))
- (funcall (nth 0 debugger-previous-state))))))
+ (if debugger-previous-state
+ (debugger--restore-buffer-state debugger-previous-state)
+ (setq backtrace-insert-header-function nil)
+ (setq backtrace-frames nil)
+ (backtrace-print))))
(with-timeout-unsuspend debugger-with-timeout-suspend)
(set-match-data debugger-outer-match-data)))
(setq debug-on-next-call debugger-step-after-exit)
@@ -301,112 +321,80 @@ first will be printed into the backtrace buffer."
(message "Error in debug printer: %S" err)
(prin1 obj stream))))
-(defun debugger-insert-backtrace (frames do-xrefs)
- "Format and insert the backtrace FRAMES at point.
-Make functions into cross-reference buttons if DO-XREFS is non-nil."
- (let ((standard-output (current-buffer))
- (eval-buffers eval-buffer-list))
- (require 'help-mode) ; Define `help-function-def' button type.
- (pcase-dolist (`(,evald ,fun ,args ,flags) frames)
- (insert (if (plist-get flags :debug-on-exit)
- "* " " "))
- (let ((fun-file (and do-xrefs (symbol-file fun 'defun)))
- (fun-pt (point)))
- (cond
- ((and evald (not debugger-stack-frame-as-list))
- (debugger--print fun)
- (if args (debugger--print args) (princ "()")))
- (t
- (debugger--print (cons fun args))
- (cl-incf fun-pt)))
- (when fun-file
- (make-text-button fun-pt (+ fun-pt (length (symbol-name fun)))
- :type 'help-function-def
- 'help-args (list fun fun-file))))
- ;; After any frame that uses eval-buffer, insert a line that
- ;; states the buffer position it's reading at.
- (when (and eval-buffers (memq fun '(eval-buffer eval-region)))
- (insert (format " ; Reading at buffer position %d"
- ;; This will get the wrong result if there are
- ;; two nested eval-region calls for the same
- ;; buffer. That's not a very useful case.
- (with-current-buffer (pop eval-buffers)
- (point)))))
- (insert "\n"))))
-
(defun debugger-setup-buffer (args)
"Initialize the `*Backtrace*' buffer for entry to the debugger.
-That buffer should be current already."
- (setq buffer-read-only nil)
- (erase-buffer)
- (set-buffer-multibyte t) ;Why was it nil ? -stef
- (setq buffer-undo-list t)
+That buffer should be current already and in debugger-mode."
+ (setq backtrace-frames (nthcdr
+ ;; Remove debug--implement-debug-on-entry and the
+ ;; advice's `apply' frame.
+ (if (eq (car args) 'debug) 3 1)
+ (backtrace-get-frames 'debug)))
+ (when (eq (car-safe args) 'exit)
+ (setq debugger-value (nth 1 args))
+ (setf (cl-getf (backtrace-frame-flags (car backtrace-frames))
+ :debug-on-exit)
+ nil))
+
+ (setq backtrace-view (plist-put backtrace-view :show-flags t)
+ backtrace-insert-header-function (lambda ()
+ (debugger--insert-header args))
+ backtrace-print-function debugger-print-function)
+ (backtrace-print)
+ ;; Place point on "stack frame 0" (bug#15101).
+ (goto-char (point-min))
+ (search-forward ":" (line-end-position) t)
+ (when (and (< (point) (line-end-position))
+ (= (char-after) ?\s))
+ (forward-char)))
+
+(defun debugger--insert-header (args)
+ "Insert the header for the debugger's Backtrace buffer.
+Include the reason for debugger entry from ARGS."
(insert "Debugger entered")
- (let ((frames (nthcdr
- ;; Remove debug--implement-debug-on-entry and the
- ;; advice's `apply' frame.
- (if (eq (car args) 'debug) 3 1)
- (backtrace-frames 'debug)))
- (print-escape-newlines t)
- (print-escape-control-characters t)
- ;; If you increase print-level, add more depth in call_debugger.
- (print-level 8)
- (print-length 50)
- (pos (point)))
- (pcase (car args)
- ;; lambda is for debug-on-call when a function call is next.
- ;; debug is for debug-on-entry function called.
- ((or `lambda `debug)
- (insert "--entering a function:\n")
- (setq pos (1- (point))))
- ;; Exiting a function.
- (`exit
- (insert "--returning value: ")
- (setq pos (point))
- (setq debugger-value (nth 1 args))
- (debugger--print debugger-value (current-buffer))
- (setf (cl-getf (nth 3 (car frames)) :debug-on-exit) nil)
- (insert ?\n))
- ;; Watchpoint triggered.
- ((and `watchpoint (let `(,symbol ,newval . ,details) (cdr args)))
- (insert
- "--"
- (pcase details
- (`(makunbound nil) (format "making %s void" symbol))
- (`(makunbound ,buffer) (format "killing local value of %s in buffer
%s"
- symbol buffer))
- (`(defvaralias ,_) (format "aliasing %s to %s" symbol newval))
- (`(let ,_) (format "let-binding %s to %S" symbol newval))
- (`(unlet ,_) (format "ending let-binding of %s" symbol))
- (`(set nil) (format "setting %s to %S" symbol newval))
- (`(set ,buffer) (format "setting %s in buffer %s to %S"
- symbol buffer newval))
- (_ (error "unrecognized watchpoint triggered %S" (cdr args))))
- ": ")
- (setq pos (point))
- (insert ?\n))
- ;; Debugger entered for an error.
- (`error
- (insert "--Lisp error: ")
- (setq pos (point))
- (debugger--print (nth 1 args) (current-buffer))
- (insert ?\n))
- ;; debug-on-call, when the next thing is an eval.
- (`t
- (insert "--beginning evaluation of function call form:\n")
- (setq pos (1- (point))))
- ;; User calls debug directly.
- (_
- (insert ": ")
- (setq pos (point))
- (debugger--print
- (if (eq (car args) 'nil)
- (cdr args) args)
- (current-buffer))
- (insert ?\n)))
- (debugger-insert-backtrace frames t)
- ;; Place point on "stack frame 0" (bug#15101).
- (goto-char pos)))
+ (pcase (car args)
+ ;; lambda is for debug-on-call when a function call is next.
+ ;; debug is for debug-on-entry function called.
+ ((or `lambda `debug)
+ (insert "--entering a function:\n"))
+ ;; Exiting a function.
+ (`exit
+ (insert "--returning value: ")
+ (insert (backtrace-print-to-string debugger-value))
+ (insert ?\n))
+ ;; Watchpoint triggered.
+ ((and `watchpoint (let `(,symbol ,newval . ,details) (cdr args)))
+ (insert
+ "--"
+ (pcase details
+ (`(makunbound nil) (format "making %s void" symbol))
+ (`(makunbound ,buffer) (format "killing local value of %s in buffer %s"
+ symbol buffer))
+ (`(defvaralias ,_) (format "aliasing %s to %s" symbol newval))
+ (`(let ,_) (format "let-binding %s to %s" symbol
+ (backtrace-print-to-string newval)))
+ (`(unlet ,_) (format "ending let-binding of %s" symbol))
+ (`(set nil) (format "setting %s to %s" symbol
+ (backtrace-print-to-string newval)))
+ (`(set ,buffer) (format "setting %s in buffer %s to %s"
+ symbol buffer
+ (backtrace-print-to-string newval)))
+ (_ (error "unrecognized watchpoint triggered %S" (cdr args))))
+ ": ")
+ (insert ?\n))
+ ;; Debugger entered for an error.
+ (`error
+ (insert "--Lisp error: ")
+ (insert (backtrace-print-to-string (nth 1 args)))
+ (insert ?\n))
+ ;; debug-on-call, when the next thing is an eval.
+ (`t
+ (insert "--beginning evaluation of function call form:\n"))
+ ;; User calls debug directly.
+ (_
+ (insert ": ")
+ (insert (backtrace-print-to-string (if (eq (car args) 'nil)
+ (cdr args) args)))
+ (insert ?\n))))
(defun debugger-step-through ()
@@ -426,12 +414,12 @@ Enter another debugger on next entry to eval, apply or
funcall."
(unless debugger-may-continue
(error "Cannot continue"))
(message "Continuing.")
- (save-excursion
- ;; Check to see if we've flagged some frame for debug-on-exit, in which
- ;; case we'll probably come back to the debugger soon.
- (goto-char (point-min))
- (if (re-search-forward "^\\* " nil t)
- (setq debugger-will-be-back t)))
+
+ ;; Check to see if we've flagged some frame for debug-on-exit, in which
+ ;; case we'll probably come back to the debugger soon.
+ (dolist (frame backtrace-frames)
+ (when (plist-get (backtrace-frame-flags frame) :debug-on-exit)
+ (setq debugger-will-be-back t)))
(exit-recursive-edit))
(defun debugger-return-value (val)
@@ -446,12 +434,11 @@ will be used, such as in a debug on exit from a frame."
(setq debugger-value val)
(princ "Returning " t)
(debugger--print debugger-value)
- (save-excursion
;; Check to see if we've flagged some frame for debug-on-exit, in which
;; case we'll probably come back to the debugger soon.
- (goto-char (point-min))
- (if (re-search-forward "^\\* " nil t)
- (setq debugger-will-be-back t)))
+ (dolist (frame backtrace-frames)
+ (when (plist-get (backtrace-frame-flags frame) :debug-on-exit)
+ (setq debugger-will-be-back t)))
(exit-recursive-edit))
(defun debugger-jump ()
@@ -473,63 +460,40 @@ removes itself from that hook."
(defun debugger-frame-number (&optional skip-base)
"Return number of frames in backtrace before the one point points at."
- (save-excursion
- (beginning-of-line)
- (if (looking-at " *;;;\\|[a-z]")
- (error "This line is not a function call"))
- (let ((opoint (point))
- (count 0))
- (unless skip-base
+ (let ((index (backtrace-get-index))
+ (count 0))
+ (unless index
+ (error "This line is not a function call"))
+ (unless skip-base
(while (not (eq (cadr (backtrace-frame count)) 'debug))
(setq count (1+ count)))
;; Skip debug--implement-debug-on-entry frame.
(when (eq 'debug--implement-debug-on-entry
(cadr (backtrace-frame (1+ count))))
(setq count (+ 2 count))))
- (goto-char (point-min))
- (when (looking-at "Debugger entered--\\(Lisp error\\|returning
value\\):")
- (goto-char (match-end 0))
- (forward-sexp 1))
- (forward-line 1)
- (while (progn
- (forward-char 2)
- (cond ((debugger--locals-visible-p)
- (goto-char (next-single-char-property-change
- (point) 'locals-visible)))
- ((= (following-char) ?\()
- (forward-sexp 1))
- (t
- (forward-sexp 2)))
- (forward-line 1)
- (<= (point) opoint))
- (if (looking-at " *;;;")
- (forward-line 1))
- (setq count (1+ count)))
- count)))
+ (+ count index)))
(defun debugger-frame ()
"Request entry to debugger when this frame exits.
Applies to the frame whose line point is on in the backtrace."
(interactive)
(backtrace-debug (debugger-frame-number) t)
- (beginning-of-line)
- (if (= (following-char) ? )
- (let ((inhibit-read-only t))
- (delete-char 1)
- (insert ?*)))
- (beginning-of-line))
+ (setf
+ (cl-getf (backtrace-frame-flags (nth (backtrace-get-index)
backtrace-frames))
+ :debug-on-exit)
+ t)
+ (backtrace-update-flags))
(defun debugger-frame-clear ()
"Do not enter debugger when this frame exits.
Applies to the frame whose line point is on in the backtrace."
(interactive)
(backtrace-debug (debugger-frame-number) nil)
- (beginning-of-line)
- (if (= (following-char) ?*)
- (let ((inhibit-read-only t))
- (delete-char 1)
- (insert ? )))
- (beginning-of-line))
+ (setf
+ (cl-getf (backtrace-frame-flags (nth (backtrace-get-index)
backtrace-frames))
+ :debug-on-exit)
+ nil)
+ (backtrace-update-flags))
(defmacro debugger-env-macro (&rest body)
"Run BODY in original environment."
@@ -564,69 +528,10 @@ The environment used is the one when entering the
activation frame at point."
(let ((str (eval-expression-print-format val)))
(if str (princ str t))))))))
-(defun debugger--locals-visible-p ()
- "Are the local variables of the current stack frame visible?"
- (save-excursion
- (move-to-column 2)
- (get-text-property (point) 'locals-visible)))
-
-(defun debugger--insert-locals (locals)
- "Insert the local variables LOCALS at point."
- (cond ((null locals)
- (insert "\n [no locals]"))
- (t
- (let ((print-escape-newlines t))
- (dolist (s+v locals)
- (let ((symbol (car s+v))
- (value (cdr s+v)))
- (insert "\n ")
- (prin1 symbol (current-buffer))
- (insert " = ")
- (debugger--print value (current-buffer))))))))
-
-(defun debugger--show-locals ()
- "For the frame at point, insert locals and add text properties."
- (let* ((nframe (1+ (debugger-frame-number 'skip-base)))
- (base (debugger--backtrace-base))
- (locals (backtrace--locals nframe base))
- (inhibit-read-only t))
- (save-excursion
- (let ((start (progn
- (move-to-column 2)
- (point))))
- (end-of-line)
- (debugger--insert-locals locals)
- (add-text-properties start (point) '(locals-visible t))))))
-
-(defun debugger--hide-locals ()
- "Delete local variables and remove the text property."
- (let* ((col (current-column))
- (end (progn
- (move-to-column 2)
- (next-single-char-property-change (point) 'locals-visible)))
- (start (previous-single-char-property-change end 'locals-visible))
- (inhibit-read-only t))
- (remove-text-properties start end '(locals-visible))
- (goto-char start)
- (end-of-line)
- (delete-region (point) end)
- (move-to-column col)))
-
-(defun debugger-toggle-locals ()
- "Show or hide local variables of the current stack frame."
- (interactive)
- (cond ((debugger--locals-visible-p)
- (debugger--hide-locals))
- (t
- (debugger--show-locals))))
-
(defvar debugger-mode-map
- (let ((map (make-keymap))
- (menu-map (make-sparse-keymap)))
- (set-keymap-parent map button-buffer-map)
- (suppress-keymap map)
- (define-key map "-" 'negative-argument)
+ (let ((map (make-keymap)))
+ (set-keymap-parent map backtrace-mode-map)
(define-key map "b" 'debugger-frame)
(define-key map "c" 'debugger-continue)
(define-key map "j" 'debugger-jump)
@@ -634,63 +539,47 @@ The environment used is the one when entering the
activation frame at point."
(define-key map "u" 'debugger-frame-clear)
(define-key map "d" 'debugger-step-through)
(define-key map "l" 'debugger-list-functions)
- (define-key map "h" 'describe-mode)
- (define-key map "q" 'top-level)
+ (define-key map "q" 'debugger-quit)
(define-key map "e" 'debugger-eval-expression)
- (define-key map "v" 'debugger-toggle-locals) ; "v" is for "variables".
- (define-key map " " 'next-line)
(define-key map "R" 'debugger-record-expression)
- (define-key map "\C-m" 'debug-help-follow)
(define-key map [mouse-2] 'push-button)
- (define-key map [menu-bar debugger] (cons "Debugger" menu-map))
- (define-key menu-map [deb-top]
- '(menu-item "Quit" top-level
- :help "Quit debugging and return to top level"))
- (define-key menu-map [deb-s0] '("--"))
- (define-key menu-map [deb-descr]
- '(menu-item "Describe Debugger Mode" describe-mode
- :help "Display documentation for debugger-mode"))
- (define-key menu-map [deb-hfol]
- '(menu-item "Help Follow" debug-help-follow
- :help "Follow cross-reference"))
- (define-key menu-map [deb-nxt]
- '(menu-item "Next Line" next-line
- :help "Move cursor down"))
- (define-key menu-map [deb-s1] '("--"))
- (define-key menu-map [deb-lfunc]
- '(menu-item "List debug on entry functions" debugger-list-functions
- :help "Display a list of all the functions now set to debug
on entry"))
- (define-key menu-map [deb-fclear]
- '(menu-item "Cancel debug frame" debugger-frame-clear
- :help "Do not enter debugger when this frame exits"))
- (define-key menu-map [deb-frame]
- '(menu-item "Debug frame" debugger-frame
- :help "Request entry to debugger when this frame exits"))
- (define-key menu-map [deb-s2] '("--"))
- (define-key menu-map [deb-ret]
- '(menu-item "Return value..." debugger-return-value
- :help "Continue, specifying value to return."))
- (define-key menu-map [deb-rec]
- '(menu-item "Display and Record Expression" debugger-record-expression
- :help "Display a variable's value and record it in
`*Backtrace-record*' buffer"))
- (define-key menu-map [deb-eval]
- '(menu-item "Eval Expression..." debugger-eval-expression
- :help "Eval an expression, in an environment like that
outside the debugger"))
- (define-key menu-map [deb-jump]
- '(menu-item "Jump" debugger-jump
- :help "Continue to exit from this frame, with all
debug-on-entry suspended"))
- (define-key menu-map [deb-cont]
- '(menu-item "Continue" debugger-continue
- :help "Continue, evaluating this expression without stopping"))
- (define-key menu-map [deb-step]
- '(menu-item "Step through" debugger-step-through
- :help "Proceed, stepping through subexpressions of this expression"))
+ (easy-menu-define nil map ""
+ '("Debugger"
+ ["Step through" debugger-step-through
+ :help "Proceed, stepping through subexpressions of this expression"]
+ ["Continue" debugger-continue
+ :help "Continue, evaluating this expression without stopping"]
+ ["Jump" debugger-jump
+ :help "Continue to exit from this frame, with all debug-on-entry
suspended"]
+ ["Eval Expression..." debugger-eval-expression
+ :help "Eval an expression, in an environment like that outside the
debugger"]
+ ["Display and Record Expression" debugger-record-expression
+ :help "Display a variable's value and record it in
`*Backtrace-record*' buffer"]
+ ["Return value..." debugger-return-value
+ :help "Continue, specifying value to return."]
+ "--"
+ ["Debug frame" debugger-frame
+ :help "Request entry to debugger when this frame exits"]
+ ["Cancel debug frame" debugger-frame-clear
+ :help "Do not enter debugger when this frame exits"]
+ ["List debug on entry functions" debugger-list-functions
+ :help "Display a list of all the functions now set to debug on entry"]
+ "--"
+ ["Next Line" next-line
+ :help "Move cursor down"]
+ ["Help for Symbol" backtrace-help-follow-symbol
+ :help "Show help for symbol at point"]
+ ["Describe Debugger Mode" describe-mode
+ :help "Display documentation for debugger-mode"]
+ "--"
+ ["Quit" debugger-quit
+ :help "Quit debugging and return to top level"]))
map))
(put 'debugger-mode 'mode-class 'special)
-(define-derived-mode debugger-mode fundamental-mode "Debugger"
- "Mode for backtrace buffers, selected in debugger.
+(define-derived-mode debugger-mode backtrace-mode "Debugger"
+ "Mode for debugging Emacs Lisp using a backtrace.
\\<debugger-mode-map>
A line starts with `*' if exiting that frame will call the debugger.
Type \\[debugger-frame] or \\[debugger-frame-clear] to set or remove the `*'.
@@ -704,8 +593,6 @@ which functions will enter the debugger when called.
Complete list of commands:
\\{debugger-mode-map}"
- (setq truncate-lines t)
- (set-syntax-table emacs-lisp-mode-syntax-table)
(add-hook 'kill-buffer-hook
(lambda () (if (> (recursion-depth) 0) (top-level)))
nil t)
@@ -732,27 +619,6 @@ Complete list of commands:
(buffer-substring (line-beginning-position 0)
(line-end-position 0)))))
-(defun debug-help-follow (&optional pos)
- "Follow cross-reference at POS, defaulting to point.
-
-For the cross-reference format, see `help-make-xrefs'."
- (interactive "d")
- ;; Ideally we'd just do (call-interactively 'help-follow) except that this
- ;; assumes we're already in a *Help* buffer and reuses it, so it ends up
- ;; incorrectly "reusing" the *Backtrace* buffer to show the help info.
- (unless pos
- (setq pos (point)))
- (unless (push-button pos)
- ;; check if the symbol under point is a function or variable
- (let ((sym
- (intern
- (save-excursion
- (goto-char pos) (skip-syntax-backward "w_")
- (buffer-substring (point)
- (progn (skip-syntax-forward "w_")
- (point)))))))
- (when (or (boundp sym) (fboundp sym) (facep sym))
- (describe-symbol sym)))))
;; When you change this, you may also need to change the number of
;; frames that the debugger skips.
@@ -853,6 +719,13 @@ To specify a nil argument interactively, exit with an
empty minibuffer."
;;(princ "be set to debug on entry, even if it is in the list.")
)))))
+(defun debugger-quit ()
+ "Quit debugging and return to the top level."
+ (interactive)
+ (if (= (recursion-depth) 0)
+ (quit-window)
+ (top-level)))
+
(defun debug--implement-debug-watch (symbol newval op where)
"Conditionally call the debugger.
This function is called when SYMBOL's value is modified."
diff --git a/lisp/emacs-lisp/easy-mmode.el b/lisp/emacs-lisp/easy-mmode.el
index b83b53a..4d8a502 100644
--- a/lisp/emacs-lisp/easy-mmode.el
+++ b/lisp/emacs-lisp/easy-mmode.el
@@ -474,22 +474,26 @@ See `%s' for more information on %s."
;; The function that calls TURN-ON in each buffer.
(defun ,MODE-enable-in-buffers ()
- (dolist (buf ,MODE-buffers)
- (when (buffer-live-p buf)
- (with-current-buffer buf
- (unless ,MODE-set-explicitly
- (unless (eq ,MODE-major-mode major-mode)
- (if ,mode
- (progn
- (,mode -1)
- (funcall #',turn-on))
- (funcall #',turn-on))))
- (setq ,MODE-major-mode major-mode)))))
+ (let ((buffers ,MODE-buffers))
+ ;; Clear MODE-buffers to avoid scanning the same list of
+ ;; buffers in recursive calls to MODE-enable-in-buffers.
+ ;; Otherwise it could lead to infinite recursion.
+ (setq ,MODE-buffers nil)
+ (dolist (buf buffers)
+ (when (buffer-live-p buf)
+ (with-current-buffer buf
+ (unless ,MODE-set-explicitly
+ (unless (eq ,MODE-major-mode major-mode)
+ (if ,mode
+ (progn
+ (,mode -1)
+ (funcall #',turn-on))
+ (funcall #',turn-on))))
+ (setq ,MODE-major-mode major-mode))))))
(put ',MODE-enable-in-buffers 'definition-name ',global-mode)
(defun ,MODE-check-buffers ()
(,MODE-enable-in-buffers)
- (setq ,MODE-buffers nil)
(remove-hook 'post-command-hook ',MODE-check-buffers))
(put ',MODE-check-buffers 'definition-name ',global-mode)
diff --git a/lisp/emacs-lisp/edebug.el b/lisp/emacs-lisp/edebug.el
index e759c5b..fa418c6 100644
--- a/lisp/emacs-lisp/edebug.el
+++ b/lisp/emacs-lisp/edebug.el
@@ -52,6 +52,7 @@
;;; Code:
+(require 'backtrace)
(require 'macroexp)
(require 'cl-lib)
(eval-when-compile (require 'pcase))
@@ -206,8 +207,7 @@ Use this with caution since it is not debugged."
"Non-nil if Edebug should unwrap results of expressions.
That is, Edebug will try to remove its own instrumentation from the result.
This is useful when debugging macros where the results of expressions
-are instrumented expressions. But don't do this when results might be
-circular or an infinite loop will result."
+are instrumented expressions."
:type 'boolean
:group 'edebug)
@@ -1198,6 +1198,8 @@ purpose by adding an entry to this alist, and setting
(defvar edebug-inside-func) ;; whether code is inside function context.
;; Currently def-form sets this to nil; def-body sets it to t.
+(defvar edebug--cl-macrolet-defs) ;; Fully defined below.
+
(defun edebug-interactive-p-name ()
;; Return a unique symbol for the variable used to store the
;; status of interactive-p for this function.
@@ -1263,25 +1265,59 @@ purpose by adding an entry to this alist, and setting
(defun edebug-unwrap (sexp)
"Return the unwrapped SEXP or return it as is if it is not wrapped.
The SEXP might be the result of wrapping a body, which is a list of
-expressions; a `progn' form will be returned enclosing these forms."
- (if (consp sexp)
- (cond
- ((eq 'edebug-after (car sexp))
- (nth 3 sexp))
- ((eq 'edebug-enter (car sexp))
- (macroexp-progn (nthcdr 2 (nth 1 (nth 3 sexp)))))
- (t sexp);; otherwise it is not wrapped, so just return it.
- )
- sexp))
+expressions; a `progn' form will be returned enclosing these forms.
+Does not unwrap inside vectors, records, structures, or hash tables."
+ (pcase sexp
+ (`(edebug-after ,_before-form ,_after-index ,form)
+ form)
+ (`(lambda ,args (edebug-enter ',_sym ,_arglist
+ (function (lambda nil . ,body))))
+ `(lambda ,args ,@body))
+ (`(closure ,env ,args (edebug-enter ',_sym ,_arglist
+ (function (lambda nil . ,body))))
+ `(closure ,env ,args ,@body))
+ (`(edebug-enter ',_sym ,_args (function (lambda nil . ,body)))
+ (macroexp-progn body))
+ (_ sexp)))
(defun edebug-unwrap* (sexp)
"Return the SEXP recursively unwrapped."
+ (let ((ht (make-hash-table :test 'eq)))
+ (edebug--unwrap1 sexp ht)))
+
+(defun edebug--unwrap1 (sexp hash-table)
+ "Unwrap SEXP using HASH-TABLE of things already unwrapped.
+HASH-TABLE contains the results of unwrapping cons cells within
+SEXP, which are reused to avoid infinite loops when SEXP is or
+contains a circular object."
(let ((new-sexp (edebug-unwrap sexp)))
(while (not (eq sexp new-sexp))
(setq sexp new-sexp
new-sexp (edebug-unwrap sexp)))
(if (consp new-sexp)
- (mapcar #'edebug-unwrap* new-sexp)
+ (let ((result (gethash new-sexp hash-table nil)))
+ (unless result
+ (let ((remainder new-sexp)
+ current)
+ (setq result (cons nil nil)
+ current result)
+ (while
+ (progn
+ (puthash remainder current hash-table)
+ (setf (car current)
+ (edebug--unwrap1 (car remainder) hash-table))
+ (setq remainder (cdr remainder))
+ (cond
+ ((atom remainder)
+ (setf (cdr current)
+ (edebug--unwrap1 remainder hash-table))
+ nil)
+ ((gethash remainder hash-table nil)
+ (setf (cdr current) (gethash remainder hash-table nil))
+ nil)
+ (t (setq current
+ (setf (cdr current) (cons nil nil)))))))))
+ result)
new-sexp)))
@@ -1463,6 +1499,11 @@ expressions; a `progn' form will be returned enclosing
these forms."
;; Helper for edebug-list-form
(let ((spec (get-edebug-spec head)))
(cond
+ ;; Treat cl-macrolet bindings like macros with no spec.
+ ((member head edebug--cl-macrolet-defs)
+ (if edebug-eval-macro-args
+ (edebug-forms cursor)
+ (edebug-sexps cursor)))
(spec
(cond
((consp spec)
@@ -1651,6 +1692,9 @@ expressions; a `progn' form will be returned enclosing
these forms."
;; (function . edebug-match-function)
(lambda-expr . edebug-match-lambda-expr)
(cl-generic-method-args . edebug-match-cl-generic-method-args)
+ (cl-macrolet-expr . edebug-match-cl-macrolet-expr)
+ (cl-macrolet-name . edebug-match-cl-macrolet-name)
+ (cl-macrolet-body . edebug-match-cl-macrolet-body)
(¬ . edebug-match-¬)
(&key . edebug-match-&key)
(place . edebug-match-place)
@@ -1954,6 +1998,43 @@ expressions; a `progn' form will be returned enclosing
these forms."
(edebug-move-cursor cursor)
(list args)))
+(defvar edebug--cl-macrolet-defs nil
+ "List of symbols found within the bindings of enclosing `cl-macrolet'
forms.")
+(defvar edebug--current-cl-macrolet-defs nil
+ "List of symbols found within the bindings of the current `cl-macrolet'
form.")
+
+(defun edebug-match-cl-macrolet-expr (cursor)
+ "Match a `cl-macrolet' form at CURSOR."
+ (let (edebug--current-cl-macrolet-defs)
+ (edebug-match cursor
+ '((&rest (&define cl-macrolet-name cl-macro-list
+ cl-declarations-or-string
+ def-body))
+ cl-declarations cl-macrolet-body))))
+
+(defun edebug-match-cl-macrolet-name (cursor)
+ "Match the name in a `cl-macrolet' binding at CURSOR.
+Collect the names in `edebug--cl-macrolet-defs' where they
+will be checked by `edebug-list-form-args' and treated as
+macros without a spec."
+ (let ((name (edebug-top-element-required cursor "Expected name")))
+ (when (not (symbolp name))
+ (edebug-no-match cursor "Bad name:" name))
+ ;; Change edebug-def-name to avoid conflicts with
+ ;; names at global scope.
+ (setq edebug-def-name (gensym "edebug-anon"))
+ (edebug-move-cursor cursor)
+ (push name edebug--current-cl-macrolet-defs)
+ (list name)))
+
+(defun edebug-match-cl-macrolet-body (cursor)
+ "Match the body of a `cl-macrolet' expression at CURSOR.
+Put the definitions collected in `edebug--current-cl-macrolet-defs'
+into `edebug--cl-macrolet-defs' which is checked in `edebug-list-form-args'."
+ (let ((edebug--cl-macrolet-defs (nconc edebug--current-cl-macrolet-defs
+ edebug--cl-macrolet-defs)))
+ (edebug-match-body cursor)))
+
(defun edebug-match-arg (cursor)
;; set the def-args bound in edebug-defining-form
(let ((edebug-arg (edebug-top-element-required cursor "Expected arg")))
@@ -3611,7 +3692,7 @@ be installed in `emacs-lisp-mode-map'.")
;; misc
(define-key map "?" 'edebug-help)
- (define-key map "d" 'edebug-backtrace)
+ (define-key map "d" 'edebug-pop-to-backtrace)
(define-key map "-" 'negative-argument)
@@ -3869,8 +3950,10 @@ Global commands prefixed by `global-edebug-prefix':
;; (setq debugger 'debug) ; use the standard debugger
;; Note that debug and its utilities must be byte-compiled to work,
-;; since they depend on the backtrace looking a certain way. But
-;; edebug is not dependent on this, yet.
+;; since they depend on the backtrace looking a certain way. Edebug
+;; will work if not byte-compiled, but it will not be able correctly
+;; remove its instrumentation from backtraces unless it is
+;; byte-compiled.
(defun edebug (&optional arg-mode &rest args)
"Replacement for `debug'.
@@ -3900,49 +3983,136 @@ Otherwise call `debug' normally."
(apply #'debug arg-mode args)
))
-
-(defun edebug-backtrace ()
- "Display a non-working backtrace. Better than nothing..."
+;;; Backtrace buffer
+
+(defvar-local edebug-backtrace-frames nil
+ "Stack frames of the current Edebug Backtrace buffer without instrumentation.
+This should be a list of `edebug---frame' objects.")
+(defvar-local edebug-instrumented-backtrace-frames nil
+ "Stack frames of the current Edebug Backtrace buffer with instrumentation.
+This should be a list of `edebug---frame' objects.")
+
+;; Data structure for backtrace frames with information
+;; from Edebug instrumentation found in the backtrace.
+(cl-defstruct
+ (edebug--frame
+ (:constructor edebug--make-frame)
+ (:include backtrace-frame))
+ def-name before-index after-index)
+
+(defun edebug-pop-to-backtrace ()
+ "Display the current backtrace in a `backtrace-mode' window."
(interactive)
(if (or (not edebug-backtrace-buffer)
(null (buffer-name edebug-backtrace-buffer)))
(setq edebug-backtrace-buffer
- (generate-new-buffer "*Backtrace*"))
+ (generate-new-buffer "*Edebug Backtrace*"))
;; Else, could just display edebug-backtrace-buffer.
)
- (with-output-to-temp-buffer (buffer-name edebug-backtrace-buffer)
- (setq edebug-backtrace-buffer standard-output)
- (let ((print-escape-newlines t)
- (print-length 50) ; FIXME cf edebug-safe-prin1-to-string
- last-ok-point)
- (backtrace)
-
- ;; Clean up the backtrace.
- ;; Not quite right for current edebug scheme.
- (set-buffer edebug-backtrace-buffer)
- (setq truncate-lines t)
- (goto-char (point-min))
- (setq last-ok-point (point))
- (if t (progn
-
- ;; Delete interspersed edebug internals.
- (while (re-search-forward "^ (?edebug" nil t)
- (beginning-of-line)
- (cond
- ((looking-at "^ (edebug-after")
- ;; Previous lines may contain code, so just delete this line.
- (setq last-ok-point (point))
- (forward-line 1)
- (delete-region last-ok-point (point)))
-
- ((looking-at (if debugger-stack-frame-as-list
- "^ (edebug"
- "^ edebug"))
- (forward-line 1)
- (delete-region last-ok-point (point))
- )))
- )))))
+ (pop-to-buffer edebug-backtrace-buffer)
+ (unless (derived-mode-p 'backtrace-mode)
+ (backtrace-mode)
+ (add-hook 'backtrace-goto-source-functions 'edebug--backtrace-goto-source))
+ (setq edebug-instrumented-backtrace-frames
+ (backtrace-get-frames 'edebug-debugger
+ :constructor #'edebug--make-frame)
+ edebug-backtrace-frames (edebug--strip-instrumentation
+ edebug-instrumented-backtrace-frames)
+ backtrace-frames edebug-backtrace-frames)
+ (backtrace-print)
+ (goto-char (point-min)))
+
+(defun edebug--strip-instrumentation (frames)
+ "Return a new list of backtrace frames with instrumentation removed.
+Remove frames for Edebug's functions and the lambdas in
+`edebug-enter' wrappers. Fill in the def-name, before-index
+and after-index fields in both FRAMES and the returned list
+of deinstrumented frames, for those frames where the source
+code location is known."
+ (let (skip-next-lambda def-name before-index after-index results
+ (index (length frames)))
+ (dolist (frame (reverse frames))
+ (let ((new-frame (copy-edebug--frame frame))
+ (fun (edebug--frame-fun frame))
+ (args (edebug--frame-args frame)))
+ (cl-decf index)
+ (pcase fun
+ ('edebug-enter
+ (setq skip-next-lambda t
+ def-name (nth 0 args)))
+ ('edebug-after
+ (setq before-index (if (consp (nth 0 args))
+ (nth 1 (nth 0 args))
+ (nth 0 args))
+ after-index (nth 1 args)))
+ ((pred edebug--symbol-not-prefixed-p)
+ (edebug--unwrap-frame new-frame)
+ (edebug--add-source-info new-frame def-name before-index
after-index)
+ (edebug--add-source-info frame def-name before-index after-index)
+ (push new-frame results)
+ (setq before-index nil
+ after-index nil))
+ (`(,(or 'lambda 'closure) . ,_)
+ (unless skip-next-lambda
+ (edebug--unwrap-frame new-frame)
+ (edebug--add-source-info frame def-name before-index after-index)
+ (edebug--add-source-info new-frame def-name before-index
after-index)
+ (push new-frame results))
+ (setq before-index nil
+ after-index nil
+ skip-next-lambda nil)))))
+ results))
+
+(defun edebug--symbol-not-prefixed-p (sym)
+ "Return non-nil if SYM is a symbol not prefixed by \"edebug-\"."
+ (and (symbolp sym)
+ (not (string-prefix-p "edebug-" (symbol-name sym)))))
+
+(defun edebug--unwrap-frame (frame)
+ "Remove Edebug's instrumentation from FRAME.
+Strip it from the function and any unevaluated arguments."
+ (setf (edebug--frame-fun frame) (edebug-unwrap* (edebug--frame-fun frame)))
+ (unless (edebug--frame-evald frame)
+ (let (results)
+ (dolist (arg (edebug--frame-args frame))
+ (push (edebug-unwrap* arg) results))
+ (setf (edebug--frame-args frame) (nreverse results)))))
+
+(defun edebug--add-source-info (frame def-name before-index after-index)
+ "Update FRAME with the additional info needed by an edebug--frame.
+Save DEF-NAME, BEFORE-INDEX and AFTER-INDEX in FRAME."
+ (when (and before-index def-name)
+ (setf (edebug--frame-flags frame)
+ (plist-put (copy-sequence (edebug--frame-flags frame))
+ :source-available t)))
+ (setf (edebug--frame-def-name frame) (and before-index def-name))
+ (setf (edebug--frame-before-index frame) before-index)
+ (setf (edebug--frame-after-index frame) after-index))
+
+(defun edebug--backtrace-goto-source ()
+ (let* ((index (backtrace-get-index))
+ (frame (nth index backtrace-frames)))
+ (when (edebug--frame-def-name frame)
+ (let* ((data (get (edebug--frame-def-name frame) 'edebug))
+ (marker (nth 0 data))
+ (offsets (nth 2 data)))
+ (pop-to-buffer (marker-buffer marker))
+ (goto-char (+ (marker-position marker)
+ (aref offsets (edebug--frame-before-index frame))))))))
+
+(defun edebug-backtrace-show-instrumentation ()
+ "Show Edebug's instrumentation in an Edebug Backtrace buffer."
+ (interactive)
+ (unless (eq backtrace-frames edebug-instrumented-backtrace-frames)
+ (setq backtrace-frames edebug-instrumented-backtrace-frames)
+ (revert-buffer)))
+(defun edebug-backtrace-hide-instrumentation ()
+ "Hide Edebug's instrumentation in an Edebug Backtrace buffer."
+ (interactive)
+ (unless (eq backtrace-frames edebug-backtrace-frames)
+ (setq backtrace-frames edebug-backtrace-frames)
+ (revert-buffer)))
;;; Trace display
@@ -4116,7 +4286,7 @@ It is removed when you hit any char."
["Bounce to Current Point" edebug-bounce-point t]
["View Outside Windows" edebug-view-outside t]
["Previous Result" edebug-previous-result t]
- ["Show Backtrace" edebug-backtrace t]
+ ["Show Backtrace" edebug-pop-to-backtrace t]
["Display Freq Count" edebug-display-freq-count t])
("Eval"
diff --git a/lisp/emacs-lisp/ert.el b/lisp/emacs-lisp/ert.el
index cad2104..eb9695d 100644
--- a/lisp/emacs-lisp/ert.el
+++ b/lisp/emacs-lisp/ert.el
@@ -60,6 +60,7 @@
(require 'cl-lib)
(require 'button)
(require 'debug)
+(require 'backtrace)
(require 'easymenu)
(require 'ewoc)
(require 'find-func)
@@ -677,13 +678,6 @@ and is displayed in front of the value of MESSAGE-FORM."
(cl-defstruct (ert-test-aborted-with-non-local-exit
(:include ert-test-result)))
-(defun ert--print-backtrace (backtrace do-xrefs)
- "Format the backtrace BACKTRACE to the current buffer."
- (let ((print-escape-newlines t)
- (print-level 8)
- (print-length 50))
- (debugger-insert-backtrace backtrace do-xrefs)))
-
;; A container for the state of the execution of a single test and
;; environment data needed during its execution.
(cl-defstruct ert--test-execution-info
@@ -732,7 +726,7 @@ run. ARGS are the arguments to `debugger'."
;; use.
;;
;; Grab the frames above the debugger.
- (backtrace (cdr (backtrace-frames debugger)))
+ (backtrace (cdr (backtrace-get-frames debugger)))
(infos (reverse ert--infos)))
(setf (ert--test-execution-info-result info)
(cl-ecase type
@@ -1406,9 +1400,8 @@ Returns the stats object."
(ert-test-result-with-condition
(message "Test %S backtrace:" (ert-test-name test))
(with-temp-buffer
- (ert--print-backtrace
- (ert-test-result-with-condition-backtrace result)
- nil)
+ (insert (backtrace-to-string
+ (ert-test-result-with-condition-backtrace result)))
(if (not ert-batch-backtrace-right-margin)
(message "%s"
(buffer-substring-no-properties (point-min)
@@ -2450,20 +2443,20 @@ To be used in the ERT results buffer."
(cl-etypecase result
(ert-test-passed (error "Test passed, no backtrace available"))
(ert-test-result-with-condition
- (let ((backtrace (ert-test-result-with-condition-backtrace result))
- (buffer (get-buffer-create "*ERT Backtrace*")))
+ (let ((buffer (get-buffer-create "*ERT Backtrace*")))
(pop-to-buffer buffer)
- (let ((inhibit-read-only t))
- (buffer-disable-undo)
- (erase-buffer)
- (ert-simple-view-mode)
- (set-buffer-multibyte t) ; mimic debugger-setup-buffer
- (setq truncate-lines t)
- (ert--print-backtrace backtrace t)
- (goto-char (point-min))
- (insert (substitute-command-keys "Backtrace for test `"))
- (ert-insert-test-name-button (ert-test-name test))
- (insert (substitute-command-keys "':\n"))))))))
+ (unless (derived-mode-p 'backtrace-mode)
+ (backtrace-mode))
+ (setq backtrace-insert-header-function
+ (lambda () (ert--insert-backtrace-header (ert-test-name test)))
+ backtrace-frames (ert-test-result-with-condition-backtrace
result))
+ (backtrace-print)
+ (goto-char (point-min)))))))
+
+(defun ert--insert-backtrace-header (name)
+ (insert (substitute-command-keys "Backtrace for test `"))
+ (ert-insert-test-name-button name)
+ (insert (substitute-command-keys "':\n")))
(defun ert-results-pop-to-messages-for-test-at-point ()
"Display the part of the *Messages* buffer generated during the test at
point.
diff --git a/lisp/emacs-lisp/lisp-mode.el b/lisp/emacs-lisp/lisp-mode.el
index 4e5b1a7..afb7cbd 100644
--- a/lisp/emacs-lisp/lisp-mode.el
+++ b/lisp/emacs-lisp/lisp-mode.el
@@ -517,6 +517,16 @@ This will generate compile-time constants from BINDINGS."
(defvar lisp-cl-font-lock-keywords lisp-cl-font-lock-keywords-1
"Default expressions to highlight in Lisp modes.")
+;; Support backtrace mode.
+(defconst lisp-el-font-lock-keywords-for-backtraces lisp-el-font-lock-keywords
+ "Default highlighting from Emacs Lisp mod used in Backtrace mode.")
+(defconst lisp-el-font-lock-keywords-for-backtraces-1
lisp-el-font-lock-keywords-1
+ "Subdued highlighting from Emacs Lisp mode used in Backtrace mode.")
+(defconst lisp-el-font-lock-keywords-for-backtraces-2
+ (remove (assoc 'lisp--match-hidden-arg lisp-el-font-lock-keywords-2)
+ lisp-el-font-lock-keywords-2)
+ "Gaudy highlighting from Emacs Lisp mode used in Backtrace mode.")
+
(defun lisp-string-in-doc-position-p (listbeg startpos)
"Return true if a doc string may occur at STARTPOS inside a list.
LISTBEG is the position of the start of the innermost list
@@ -1196,7 +1206,21 @@ ENDPOS is encountered."
(if endpos endpos
;; Get error now if we don't have a complete sexp
;; after point.
- (save-excursion (forward-sexp 1) (point)))))
+ (save-excursion
+ (let ((eol (line-end-position)))
+ (forward-sexp 1)
+ ;; We actually look for a sexp which ends
+ ;; after the current line so that we properly
+ ;; indent things like #s(...). This might not
+ ;; be needed if Bug#15998 is fixed.
+ (condition-case ()
+ (while (and (< (point) eol) (not (eobp)))
+ (forward-sexp 1))
+ ;; But don't signal an error for incomplete
+ ;; sexps following the first complete sexp
+ ;; after point.
+ (scan-error nil)))
+ (point)))))
(save-excursion
(while (let ((indent (lisp-indent-calc-next parse-state))
(ppss (lisp-indent-state-ppss parse-state)))
diff --git a/lisp/emacs-lisp/map-ynp.el b/lisp/emacs-lisp/map-ynp.el
index 61c04ff..a61c0ad 100644
--- a/lisp/emacs-lisp/map-ynp.el
+++ b/lisp/emacs-lisp/map-ynp.el
@@ -257,10 +257,15 @@ C-g to quit (cancel the whole command);
;; either long or short answers.
;; For backward compatibility check if short y/n answers are preferred.
-(defcustom read-answer-short (eq (symbol-function 'yes-or-no-p) 'y-or-n-p)
- "If non-nil, accept short answers to the question."
- :type 'boolean
- :version "27.1"
+(defcustom read-answer-short 'auto
+ "If non-nil, `read-answer' accepts single-character answers.
+If t, accept short (single key-press) answers to the question.
+If nil, require long answers. If `auto', accept short answers if
+the function cell of `yes-or-no-p' is set to `y-or-on-p'."
+ :type '(choice (const :tag "Accept short answers" t)
+ (const :tag "Require long answer" nil)
+ (const :tag "Guess preference" auto))
+ :version "26.2"
:group 'minibuffer)
(defconst read-answer-map--memoize (make-hash-table :weakness 'key :test
'equal))
@@ -290,8 +295,9 @@ When `read-answer-short' is non-nil, accept short answers.
Return a long answer even in case of accepting short ones.
When `use-dialog-box' is t, pop up a dialog window to get user input."
- (custom-reevaluate-setting 'read-answer-short)
- (let* ((short read-answer-short)
+ (let* ((short (if (eq read-answer-short 'auto)
+ (eq (symbol-function 'yes-or-no-p) 'y-or-n-p)
+ read-answer-short))
(answers-with-help
(if (assoc "help" answers)
answers
diff --git a/lisp/emacs-lisp/rx.el b/lisp/emacs-lisp/rx.el
index 85e74f2..bb75901 100644
--- a/lisp/emacs-lisp/rx.el
+++ b/lisp/emacs-lisp/rx.el
@@ -1183,24 +1183,28 @@ enclosed in `(and ...)'.
(pcase-defmacro rx (&rest regexps)
- "Build a `pcase' pattern matching `rx' regexps.
-The REGEXPS are interpreted as by `rx'. The pattern matches if
-the regular expression so constructed matches EXPVAL, as if
-by `string-match'.
+ "Build a `pcase' pattern matching `rx' REGEXPS in sexp form.
+The REGEXPS are interpreted as in `rx'. The pattern matches any
+string that is a match for the regular expression so constructed,
+as if by `string-match'.
In addition to the usual `rx' constructs, REGEXPS can contain the
following constructs:
- (let VAR FORM...) creates a new explicitly numbered submatch
- that matches FORM and binds the match to
- VAR.
- (backref VAR) creates a backreference to the submatch
- introduced by a previous (let VAR ...)
- construct.
-
-The VARs are associated with explicitly numbered submatches
-starting from 1. Multiple occurrences of the same VAR refer to
-the same submatch.
+ (let REF SEXP...) creates a new explicitly named reference to
+ a submatch that matches regular expressions
+ SEXP, and binds the match to REF.
+ (backref REF) creates a backreference to the submatch
+ introduced by a previous (let REF ...)
+ construct. REF can be the same symbol
+ in the first argument of the corresponding
+ (let REF ...) construct, or it can be a
+ submatch number. It matches the referenced
+ submatch.
+
+The REFs are associated with explicitly named submatches starting
+from 1. Multiple occurrences of the same REF refer to the same
+submatch.
If a case matches, the match data is modified as usual so you can
use it in the case body, but you still have to pass the correct
diff --git a/lisp/emacs-lisp/subr-x.el b/lisp/emacs-lisp/subr-x.el
index e03a81c..20eb0d5 100644
--- a/lisp/emacs-lisp/subr-x.el
+++ b/lisp/emacs-lisp/subr-x.el
@@ -211,7 +211,7 @@ The variable list SPEC is the same as in `if-let'."
(defsubst string-join (strings &optional separator)
"Join all STRINGS using SEPARATOR."
- (mapconcat 'identity strings separator))
+ (mapconcat #'identity strings separator))
(define-obsolete-function-alias 'string-reverse 'reverse "25.1")
@@ -219,17 +219,17 @@ The variable list SPEC is the same as in `if-let'."
"Trim STRING of leading string matching REGEXP.
REGEXP defaults to \"[ \\t\\n\\r]+\"."
- (if (string-match (concat "\\`\\(?:" (or regexp "[ \t\n\r]+")"\\)") string)
- (replace-match "" t t string)
+ (if (string-match (concat "\\`\\(?:" (or regexp "[ \t\n\r]+") "\\)") string)
+ (substring string (match-end 0))
string))
(defsubst string-trim-right (string &optional regexp)
"Trim STRING of trailing string matching REGEXP.
REGEXP defaults to \"[ \\t\\n\\r]+\"."
- (if (string-match (concat "\\(?:" (or regexp "[ \t\n\r]+") "\\)\\'") string)
- (replace-match "" t t string)
- string))
+ (let ((i (string-match-p (concat "\\(?:" (or regexp "[ \t\n\r]+") "\\)\\'")
+ string)))
+ (if i (substring string 0 i) string)))
(defsubst string-trim (string &optional trim-left trim-right)
"Trim STRING of leading and trailing strings matching TRIM-LEFT and
TRIM-RIGHT.
diff --git a/lisp/env.el b/lisp/env.el
index e47eb57..7007ba3 100644
--- a/lisp/env.el
+++ b/lisp/env.el
@@ -113,11 +113,11 @@ Changes ENV by side-effect, and returns its new value."
(not keep-empty)
env
(stringp (car env))
- (string-match pattern (car env)))
+ (string-match-p pattern (car env)))
(cdr env)
;; Try to find existing entry for VARIABLE in ENV.
(while (and scan (stringp (car scan)))
- (when (string-match pattern (car scan))
+ (when (string-match-p pattern (car scan))
(if value
(setcar scan (concat variable "=" value))
(if keep-empty
@@ -184,7 +184,7 @@ a side-effect."
(setq variable (encode-coding-string variable locale-coding-system)))
(if (and value (multibyte-string-p value))
(setq value (encode-coding-string value locale-coding-system)))
- (if (string-match "=" variable)
+ (if (string-match-p "=" variable)
(error "Environment variable name `%s' contains `='" variable))
(if (string-equal "TZ" variable)
(set-time-zone-rule value))
diff --git a/lisp/epg-config.el b/lisp/epg-config.el
index 98f458d..fb866df 100644
--- a/lisp/epg-config.el
+++ b/lisp/epg-config.el
@@ -98,11 +98,14 @@ Note that the buffer name starts with a space."
:type 'boolean)
(defconst epg-gpg-minimum-version "1.4.3")
+(defconst epg-gpg2-minimum-version "2.1.6")
(defconst epg-config--program-alist
`((OpenPGP
epg-gpg-program
- ("gpg2" . "2.1.6") ("gpg" . ,epg-gpg-minimum-version))
+ ("gpg2" . ,epg-gpg2-minimum-version)
+ ("gpg" . ((,epg-gpg-minimum-version . "2.0")
+ ,epg-gpg2-minimum-version)))
(CMS
epg-gpgsm-program
("gpgsm" . "2.0.4")))
@@ -228,14 +231,26 @@ version requirement is met."
(epg-config--make-gpg-configuration epg-gpg-program))
;;;###autoload
-(defun epg-check-configuration (config &optional minimum-version)
- "Verify that a sufficient version of GnuPG is installed."
+(defun epg-check-configuration (config &optional req-versions)
+ "Verify that a sufficient version of GnuPG is installed.
+CONFIG should be a `epg-configuration' object (a plist).
+REQ-VERSIONS should be a list with elements of the form (MIN
+. MAX) where MIN and MAX are version strings indicating a
+semi-open range of acceptable versions. REQ-VERSIONS may also be
+a single minimum version string."
(let ((version (alist-get 'version config)))
(unless (stringp version)
(error "Undetermined version: %S" version))
- (unless (version<= (or minimum-version
- epg-gpg-minimum-version)
- version)
+ (catch 'version-ok
+ (pcase-dolist ((or `(,min . ,max)
+ (and min (let max nil)))
+ (if (listp req-versions) req-versions
+ (list req-versions)))
+ (when (and (version<= (or min epg-gpg-minimum-version)
+ version)
+ (or (null max)
+ (version< version max)))
+ (throw 'version-ok t)))
(error "Unsupported version: %s" version))))
;;;###autoload
diff --git a/lisp/eshell/em-dirs.el b/lisp/eshell/em-dirs.el
index ec380e6..5180a07 100644
--- a/lisp/eshell/em-dirs.el
+++ b/lisp/eshell/em-dirs.el
@@ -407,6 +407,7 @@ in the minibuffer:
nil))))
(put 'eshell/cd 'eshell-no-numeric-conversions t)
+(put 'eshell/cd 'eshell-filename-arguments t)
(defun eshell-add-to-dir-ring (path)
"Add PATH to the last-dir-ring, if applicable."
@@ -470,6 +471,7 @@ in the minibuffer:
nil)
(put 'eshell/pushd 'eshell-no-numeric-conversions t)
+(put 'eshell/pushd 'eshell-filename-arguments t)
;;; popd [+n]
(defun eshell/popd (&rest args)
@@ -500,6 +502,7 @@ in the minibuffer:
nil)
(put 'eshell/popd 'eshell-no-numeric-conversions t)
+(put 'eshell/pop 'eshell-filename-arguments t)
(defun eshell/dirs (&optional if-verbose)
"Implementation of dirs in Lisp."
diff --git a/lisp/eshell/em-ls.el b/lisp/eshell/em-ls.el
index 900b289..2b568a9 100644
--- a/lisp/eshell/em-ls.el
+++ b/lisp/eshell/em-ls.el
@@ -334,6 +334,7 @@ instead."
(apply 'eshell-do-ls args)))
(put 'eshell/ls 'eshell-no-numeric-conversions t)
+(put 'eshell/ls 'eshell-filename-arguments t)
(declare-function eshell-glob-regexp "em-glob" (pattern))
diff --git a/lisp/eshell/em-unix.el b/lisp/eshell/em-unix.el
index a18fb85..c912c15 100644
--- a/lisp/eshell/em-unix.el
+++ b/lisp/eshell/em-unix.el
@@ -307,6 +307,7 @@ Remove (unlink) the FILE(s).")
nil))
(put 'eshell/rm 'eshell-no-numeric-conversions t)
+(put 'eshell/rm 'eshell-filename-arguments t)
(defun eshell/mkdir (&rest args)
"Implementation of mkdir in Lisp."
@@ -324,6 +325,7 @@ Create the DIRECTORY(ies), if they do not already exist.")
nil))
(put 'eshell/mkdir 'eshell-no-numeric-conversions t)
+(put 'eshell/mkdir 'eshell-filename-arguments t)
(defun eshell/rmdir (&rest args)
"Implementation of rmdir in Lisp."
@@ -340,6 +342,7 @@ Remove the DIRECTORY(ies), if they are empty.")
nil))
(put 'eshell/rmdir 'eshell-no-numeric-conversions t)
+(put 'eshell/rmdir 'eshell-filename-arguments t)
(defvar no-dereference)
@@ -524,6 +527,7 @@ Rename SOURCE to DEST, or move SOURCE(s) to DIRECTORY.
eshell-mv-overwrite-files))))
(put 'eshell/mv 'eshell-no-numeric-conversions t)
+(put 'eshell/mv 'eshell-filename-arguments t)
(defun eshell/cp (&rest args)
"Implementation of cp in Lisp."
@@ -561,6 +565,7 @@ Copy SOURCE to DEST, or multiple SOURCE(s) to DIRECTORY.")
eshell-cp-overwrite-files preserve)))
(put 'eshell/cp 'eshell-no-numeric-conversions t)
+(put 'eshell/cp 'eshell-filename-arguments t)
(defun eshell/ln (&rest args)
"Implementation of ln in Lisp."
@@ -593,6 +598,7 @@ with `--symbolic'. When creating hard links, each TARGET
must exist.")
eshell-ln-overwrite-files))))
(put 'eshell/ln 'eshell-no-numeric-conversions t)
+(put 'eshell/ln 'eshell-filename-arguments t)
(defun eshell/cat (&rest args)
"Implementation of cat in Lisp.
@@ -645,6 +651,7 @@ Concatenate FILE(s), or standard input, to standard
output.")
(setq eshell-ensure-newline-p nil))))
(put 'eshell/cat 'eshell-no-numeric-conversions t)
+(put 'eshell/cat 'eshell-filename-arguments t)
;; special front-end functions for compilation-mode buffers
@@ -927,6 +934,8 @@ Summarize disk usage of each FILE, recursively for
directories.")
(eshell-print (concat (eshell-du-size-string size)
"total\n"))))))))
+(put 'eshell/du 'eshell-filename-arguments t)
+
(defvar eshell-time-start nil)
(defun eshell-show-elapsed-time ()
@@ -1029,6 +1038,7 @@ Show wall-clock time elapsed during execution of
COMMAND.")
nil)
(put 'eshell/diff 'eshell-no-numeric-conversions t)
+(put 'eshell/diff 'eshell-filename-arguments t)
(defvar locate-history-list)
diff --git a/lisp/eshell/esh-cmd.el b/lisp/eshell/esh-cmd.el
index 61c0ebc..92cac61 100644
--- a/lisp/eshell/esh-cmd.el
+++ b/lisp/eshell/esh-cmd.el
@@ -1304,27 +1304,36 @@ messages, and errors."
"Insert Lisp OBJECT, using ARGS if a function."
(catch 'eshell-external ; deferred to an external command
(let* ((eshell-ensure-newline-p (eshell-interactive-output-p))
- (result
- (if (functionp object)
- (progn
- (setq eshell-last-arguments args
- eshell-last-command-name
- (concat "#<function " (symbol-name object) ">"))
- ;; if any of the arguments are flagged as numbers
- ;; waiting for conversion, convert them now
- (unless (get object 'eshell-no-numeric-conversions)
- (while args
- (let ((arg (car args)))
- (if (and (stringp arg)
- (> (length arg) 0)
- (not (text-property-not-all
- 0 (length arg) 'number t arg)))
- (setcar args (string-to-number arg))))
- (setq args (cdr args))))
- (eshell-apply object eshell-last-arguments))
- (setq eshell-last-arguments args
- eshell-last-command-name "#<Lisp object>")
- (eshell-eval object))))
+ (result
+ (if (functionp object)
+ (progn
+ (setq eshell-last-arguments args
+ eshell-last-command-name
+ (concat "#<function " (symbol-name object) ">"))
+ (let ((numeric (not (get object
+ 'eshell-no-numeric-conversions)))
+ (fname-args (get object 'eshell-filename-arguments)))
+ (when (or numeric fname-args)
+ (while args
+ (let ((arg (car args)))
+ (cond ((and numeric (stringp arg) (> (length arg) 0)
+ (text-property-any 0 (length arg)
+ 'number t arg))
+ ;; If any of the arguments are
+ ;; flagged as numbers waiting for
+ ;; conversion, convert them now.
+ (setcar args (string-to-number arg)))
+ ((and fname-args (stringp arg)
+ (string-equal arg "~"))
+ ;; If any of the arguments match "~",
+ ;; prepend "./" to treat it as a
+ ;; regular file name.
+ (setcar args (concat "./" arg)))))
+ (setq args (cdr args)))))
+ (eshell-apply object eshell-last-arguments))
+ (setq eshell-last-arguments args
+ eshell-last-command-name "#<Lisp object>")
+ (eshell-eval object))))
(if (and eshell-ensure-newline-p
(save-excursion
(goto-char eshell-last-output-end)
diff --git a/lisp/eshell/esh-ext.el b/lisp/eshell/esh-ext.el
index ba5182d..244cc7f 100644
--- a/lisp/eshell/esh-ext.el
+++ b/lisp/eshell/esh-ext.el
@@ -259,6 +259,7 @@ Adds the given PATH to $PATH.")
(eshell-printn dir)))))
(put 'eshell/addpath 'eshell-no-numeric-conversions t)
+(put 'eshell/addpath 'eshell-filename-arguments t)
(defun eshell-script-interpreter (file)
"Extract the script to run from FILE, if it has #!<interp> in it.
diff --git a/lisp/files.el b/lisp/files.el
index eabb3c0..940bacd 100644
--- a/lisp/files.el
+++ b/lisp/files.el
@@ -1830,7 +1830,7 @@ killed."
;; Don't use `find-file' because it may end up using another window
;; in some corner cases, e.g. when the selected window is
;; softly-dedicated.
- (let ((newbuf (find-file-noselect filename wildcards)))
+ (let ((newbuf (find-file-noselect filename nil nil wildcards)))
(switch-to-buffer newbuf)))
(when (eq obuf (current-buffer))
;; This executes if find-file gets an error
@@ -1954,7 +1954,7 @@ started Emacs, set `abbreviated-home-dir' to nil so it
will be recalculated)."
(save-match-data
(string-match "^[a-zA-`]:/$" filename))))
(equal (get 'abbreviated-home-dir 'home)
- (expand-file-name "~")))
+ (save-match-data (expand-file-name "~"))))
(setq filename
(concat "~"
(match-string 1 filename)
@@ -5091,6 +5091,9 @@ Before and after saving the buffer, this function runs
(make-directory dir t)
(error "Canceled")))
(setq setmodes (basic-save-buffer-1)))))
+ ;; We are hunting a nasty error, which happens on hydra.
+ ;; Adding traces might help.
+ (if (getenv "BUG_32226") (message "BUG_32226"))
;; Now we have saved the current buffer. Let's make sure
;; that buffer-file-coding-system is fixed to what
;; actually used for saving by binding it locally.
@@ -5519,6 +5522,21 @@ raised."
(dolist (dir create-list)
(files--ensure-directory dir)))))))
+(defun make-empty-file (filename &optional parents)
+ "Create an empty file FILENAME.
+Optional arg PARENTS, if non-nil then creates parent dirs as needed.
+
+If called interactively, then PARENTS is non-nil."
+ (interactive
+ (let ((filename (read-file-name "Create empty file: ")))
+ (list filename t)))
+ (when (and (file-exists-p filename) (null parents))
+ (signal 'file-already-exists `("File exists" ,filename)))
+ (let ((paren-dir (file-name-directory filename)))
+ (when (and paren-dir (not (file-exists-p paren-dir)))
+ (make-directory paren-dir parents)))
+ (write-region "" nil filename nil 0))
+
(defconst directory-files-no-dot-files-regexp
"^\\([^.]\\|\\.\\([^.]\\|\\..\\)\\).*"
"Regexp matching any file name except \".\" and \"..\".")
diff --git a/lisp/format.el b/lisp/format.el
index 5bf1be3..49d3c71 100644
--- a/lisp/format.el
+++ b/lisp/format.el
@@ -539,6 +539,8 @@ Compare using `equal'."
(setq tail next)))
(cons acopy bcopy)))
+(define-obsolete-function-alias 'format-proper-list-p 'proper-list-p "27.1")
+
(defun format-reorder (items order)
"Arrange ITEMS to follow partial ORDER.
Elements of ITEMS equal to elements of ORDER will be rearranged
diff --git a/lisp/gnus/gnus-art.el b/lisp/gnus/gnus-art.el
index 055f02f..1b0dde9 100644
--- a/lisp/gnus/gnus-art.el
+++ b/lisp/gnus/gnus-art.el
@@ -1626,6 +1626,12 @@ resources when reading email groups (and therefore stops
tracking), but allows loading external resources when reading
from NNTP newsgroups and the like.
+People controlling these external resources won't be able to tell
+that any one person in particular has read the message (since
+it's in a public venue, many people will end up loading that
+resource), but they'll be able to tell that somebody from your IP
+address has accessed the resource.
+
This can also be a function to be evaluated. If so, it will be
called with the group name as the parameter, and should return a
regexp."
diff --git a/lisp/gnus/gnus-sum.el b/lisp/gnus/gnus-sum.el
index e562b30..ceb9842 100644
--- a/lisp/gnus/gnus-sum.el
+++ b/lisp/gnus/gnus-sum.el
@@ -4310,10 +4310,10 @@ If SELECT-ARTICLES, only select those articles from
GROUP."
If FORCE-NEW is not nil, enter HEADER into the DEPENDENCIES table even
if it was already present.
-If `gnus-summary-ignore-duplicates' is nil then duplicate Message-IDs
-will not be entered in the DEPENDENCIES table. Otherwise duplicate
-Message-IDs will be renamed to a unique Message-ID before being
-entered.
+If `gnus-summary-ignore-duplicates' is non-nil then duplicate
+Message-IDs will not be entered in the DEPENDENCIES table.
+Otherwise duplicate Message-IDs will be renamed to a unique
+Message-ID before being entered.
Returns HEADER if it was entered in the DEPENDENCIES. Returns nil otherwise."
(let* ((id (mail-header-id header))
diff --git a/lisp/ielm.el b/lisp/ielm.el
index b4ad69e..8d1efcd 100644
--- a/lisp/ielm.el
+++ b/lisp/ielm.el
@@ -612,17 +612,19 @@ Customized bindings may be defined in `ielm-map', which
currently contains:
;;; User command
;;;###autoload
-(defun ielm nil
+(defun ielm (&optional buf-name)
"Interactively evaluate Emacs Lisp expressions.
-Switches to the buffer `*ielm*', or creates it if it does not exist.
+Switches to the buffer named BUF-NAME if provided (`*ielm*' by default),
+or creates it if it does not exist.
See `inferior-emacs-lisp-mode' for details."
(interactive)
- (let (old-point)
- (unless (comint-check-proc "*ielm*")
- (with-current-buffer (get-buffer-create "*ielm*")
+ (let (old-point
+ (buf-name (or buf-name "*ielm*")))
+ (unless (comint-check-proc buf-name)
+ (with-current-buffer (get-buffer-create buf-name)
(unless (zerop (buffer-size)) (setq old-point (point)))
(inferior-emacs-lisp-mode)))
- (pop-to-buffer-same-window "*ielm*")
+ (pop-to-buffer-same-window buf-name)
(when old-point (push-mark old-point))))
(provide 'ielm)
diff --git a/lisp/imenu.el b/lisp/imenu.el
index edca51e..7285b10 100644
--- a/lisp/imenu.el
+++ b/lisp/imenu.el
@@ -832,15 +832,14 @@ depending on PATTERNS."
(dolist (item index-alist)
(when (listp item)
(setcdr item (sort (cdr item) 'imenu--sort-by-position))))
- (let ((main-element (assq nil index-alist)))
- (nconc (delq main-element (delq 'dummy index-alist))
- (cdr main-element)))
;; Remove any empty menus. That can happen because of skipping
;; things inside comments or strings.
- (when (consp (car index-alist))
- (setq index-alist (cl-delete-if-not
- (lambda (it) (cdr it))
- index-alist)))))
+ (setq index-alist (cl-delete-if
+ (lambda (it) (and (consp it) (null (cdr it))))
+ index-alist))
+ (let ((main-element (assq nil index-alist)))
+ (nconc (delq main-element (delq 'dummy index-alist))
+ (cdr main-element)))))
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;;
diff --git a/lisp/indent.el b/lisp/indent.el
index 4506321..73a7d0e 100644
--- a/lisp/indent.el
+++ b/lisp/indent.el
@@ -292,7 +292,8 @@ indentation by specifying a large negative ARG."
"Indent current line to COLUMN.
This function removes or adds spaces and tabs at beginning of line
only if necessary. It leaves point at end of indentation."
- (back-to-indentation)
+ (beginning-of-line 1)
+ (skip-chars-forward " \t")
(let ((cur-col (current-column)))
(cond ((< cur-col column)
(if (>= (- column (* (/ cur-col tab-width) tab-width)) tab-width)
@@ -303,8 +304,10 @@ only if necessary. It leaves point at end of indentation."
(delete-region (progn (move-to-column column t) (point))
;; The `move-to-column' call may replace
;; tabs with spaces, so we can't reuse the
- ;; previous `back-to-indentation' point.
- (progn (back-to-indentation) (point)))))))
+ ;; previous start point.
+ (progn (beginning-of-line 1)
+ (skip-chars-forward " \t")
+ (point)))))))
(defun current-left-margin ()
"Return the left margin to use for this line.
diff --git a/lisp/international/fontset.el b/lisp/international/fontset.el
index a023d4f..d4ade3c 100644
--- a/lisp/international/fontset.el
+++ b/lisp/international/fontset.el
@@ -79,7 +79,7 @@
("cns11643.92p7-0" . chinese-cns11643-7)
("big5" . big5)
("viscii" . viscii)
- ("tis620" . tis620-2533)
+ ("tis620" . thai-iso8859-11)
("microsoft-cp1251" . windows-1251)
("koi8-r" . koi8-r)
("jisx0213.2000-1" . japanese-jisx0213-1)
@@ -139,7 +139,7 @@
(cyrillic-iso8859-5 . iso-8859-5)
(greek-iso8859-7 . iso-8859-7)
(arabic-iso8859-6 . iso-8859-6)
- (thai-tis620 . tis620-2533)
+ (thai-tis620 . thai-iso8859-11)
(latin-jisx0201 . jisx0201)
(katakana-jisx0201 . jisx0201)
(chinese-big5-1 . big5)
@@ -1233,11 +1233,12 @@ Done when `mouse-set-font' is called."
(latin-iso8859-15 . latin)
(latin-iso8859-16 . latin)
(latin-jisx0201 . latin)
+ (thai-iso8859-11 . thai)
(thai-tis620 . thai)
(cyrillic-iso8859-5 . cyrillic)
(arabic-iso8859-6 . arabic)
- (greek-iso8859-7 . latin)
- (hebrew-iso8859-8 . latin)
+ (greek-iso8859-7 . greek)
+ (hebrew-iso8859-8 . hebrew)
(katakana-jisx0201 . kana)
(chinese-gb2312 . han)
(chinese-gbk . han)
diff --git a/lisp/international/mule-cmds.el b/lisp/international/mule-cmds.el
index cf6a8c7..2bde83f 100644
--- a/lisp/international/mule-cmds.el
+++ b/lisp/international/mule-cmds.el
@@ -300,8 +300,7 @@ wrong, use this command again to toggle back to the right
mode."
(cmd (key-binding keyseq))
prefix)
;; read-key-sequence ignores quit, so make an explicit check.
- ;; Like many places, this assumes quit == C-g, but it need not be.
- (if (equal last-input-event ?\C-g)
+ (if (equal last-input-event (nth 3 (current-input-mode)))
(keyboard-quit))
(when (memq cmd '(universal-argument digit-argument))
(call-interactively cmd)
@@ -314,16 +313,16 @@ wrong, use this command again to toggle back to the right
mode."
(let ((current-prefix-arg prefix-arg)
;; Have to bind `last-command-event' here so that
;; `digit-argument', for instance, can compute the
- ;; prefix arg.
+ ;; `prefix-arg'.
(last-command-event (aref keyseq 0)))
(call-interactively cmd)))
;; This is the final call to `universal-argument-other-key', which
- ;; set's the final `prefix-arg.
+ ;; sets the final `prefix-arg'.
(let ((current-prefix-arg prefix-arg))
(call-interactively cmd))
- ;; Read the command to execute with the given prefix arg.
+ ;; Read the command to execute with the given `prefix-arg'.
(setq prefix prefix-arg
keyseq (read-key-sequence nil t)
cmd (key-binding keyseq)))
diff --git a/lisp/international/mule-conf.el b/lisp/international/mule-conf.el
index dc09570..a635c67 100644
--- a/lisp/international/mule-conf.el
+++ b/lisp/international/mule-conf.el
@@ -201,6 +201,7 @@
;; plus nbsp
(define-iso-single-byte-charset 'iso-8859-11 'thai-iso8859-11
"ISO/IEC 8859/11" "Latin/Thai" 166 ?T nil "8859-11")
+(define-charset-alias 'tis620-2533 'thai-iso8859-11)
;; 8859-12 doesn't (yet?) exist.
@@ -229,14 +230,6 @@
:code-space [32 127]
:code-offset #x0E00)
-;; Fixme: doc for this, c.f. above
-(define-charset 'tis620-2533
- "TIS620.2533"
- :short-name "TIS620.2533"
- :ascii-compatible-p t
- :code-space [0 255]
- :superset '(ascii eight-bit-control (thai-tis620 . 128)))
-
(define-charset 'jisx0201
"JISX0201"
:short-name "JISX0201"
diff --git a/lisp/international/mule-diag.el b/lisp/international/mule-diag.el
index 87a2e99..c9829e3 100644
--- a/lisp/international/mule-diag.el
+++ b/lisp/international/mule-diag.el
@@ -355,7 +355,8 @@ meanings of these arguments."
(:iso-revision-number "ISO revision number: "
number-to-string)
(:supplementary-p
- "Used only as a parent of some other charset." nil)))
+ "Used only as a parent or a subset of some other charset,
+or provided just for backward compatibility." nil)))
(let ((val (get-charset-property charset (car elt))))
(when val
(if (cadr elt) (insert (cadr elt)))
diff --git a/lisp/international/quail.el b/lisp/international/quail.el
index eece836..ec15cca 100644
--- a/lisp/international/quail.el
+++ b/lisp/international/quail.el
@@ -1394,12 +1394,13 @@ Return the input string."
(generated-events nil) ;FIXME: What is this?
(input-method-function nil)
(modified-p (buffer-modified-p))
- last-command-event last-command this-command)
+ last-command-event last-command this-command inhibit-record)
(setq quail-current-key ""
quail-current-str ""
quail-translating t)
(if key
- (setq unread-command-events (cons key unread-command-events)))
+ (setq unread-command-events (cons key unread-command-events)
+ inhibit-record t))
(while quail-translating
(set-buffer-modified-p modified-p)
(quail-show-guidance)
@@ -1408,8 +1409,13 @@ Return the input string."
(or input-method-previous-message "")
quail-current-str
quail-guidance-str)))
+ ;; We inhibit record_char only for the first key,
+ ;; because it was already recorded before read_char
+ ;; called quail-input-method.
+ (inhibit--record-char inhibit-record)
(keyseq (read-key-sequence prompt nil nil t))
(cmd (lookup-key (quail-translation-keymap) keyseq)))
+ (setq inhibit-record nil)
(if (if key
(and (commandp cmd) (not (eq cmd 'quail-other-command)))
(eq cmd 'quail-self-insert-command))
@@ -1453,14 +1459,15 @@ Return the input string."
(generated-events nil) ;FIXME: What is this?
(input-method-function nil)
(modified-p (buffer-modified-p))
- last-command-event last-command this-command)
+ last-command-event last-command this-command inhibit-record)
(setq quail-current-key ""
quail-current-str ""
quail-translating t
quail-converting t
quail-conversion-str "")
(if key
- (setq unread-command-events (cons key unread-command-events)))
+ (setq unread-command-events (cons key unread-command-events)
+ inhibit-record t))
(while quail-converting
(set-buffer-modified-p modified-p)
(or quail-translating
@@ -1476,8 +1483,13 @@ Return the input string."
quail-conversion-str
quail-current-str
quail-guidance-str)))
+ ;; We inhibit record_char only for the first key,
+ ;; because it was already recorded before read_char
+ ;; called quail-input-method.
+ (inhibit--record-char inhibit-record)
(keyseq (read-key-sequence prompt nil nil t))
(cmd (lookup-key (quail-conversion-keymap) keyseq)))
+ (setq inhibit-record nil)
(if (if key (commandp cmd) (eq cmd 'quail-self-insert-command))
(progn
(setq last-command-event (aref keyseq (1- (length keyseq)))
diff --git a/lisp/jsonrpc.el b/lisp/jsonrpc.el
index b2ccea5..a137616 100644
--- a/lisp/jsonrpc.el
+++ b/lisp/jsonrpc.el
@@ -6,7 +6,7 @@
;; Maintainer: João Távora <address@hidden>
;; Keywords: processes, languages, extensions
;; Package-Requires: ((emacs "25.2"))
-;; Version: 1.0.0
+;; Version: 1.0.2
;; This is an Elpa :core package. Don't use functionality that is not
;; compatible with Emacs 25.2.
@@ -74,7 +74,11 @@
:documentation "A hash table of request ID to continuation lambdas.")
(-events-buffer
:accessor jsonrpc--events-buffer
- :documentation "A buffer pretty-printing the JSON-RPC RPC events")
+ :documentation "A buffer pretty-printing the JSONRPC events")
+ (-events-buffer-scrollback-size
+ :initarg :events-buffer-scrollback-size
+ :accessor jsonrpc--events-buffer-scrollback-size
+ :documentation "If non-nil, maximum size of events buffer.")
(-deferred-actions
:initform (make-hash-table :test #'equal)
:accessor jsonrpc--deferred-actions
@@ -193,9 +197,7 @@ dispatcher in CONNECTION."
(when timer (cancel-timer timer)))
(remhash id (jsonrpc--request-continuations connection))
(if error (funcall (nth 1 continuations) error)
- (funcall (nth 0 continuations) result)))
- (;; An abnormal situation
- id (jsonrpc--warn "No continuation for id %s" id)))
+ (funcall (nth 0 continuations) result))))
(jsonrpc--call-deferred connection))))
@@ -256,17 +258,30 @@ Returns nil."
(apply #'jsonrpc--async-request-1 connection method params args)
nil)
-(cl-defun jsonrpc-request (connection method params &key deferred timeout)
+(cl-defun jsonrpc-request (connection
+ method params &key
+ deferred timeout
+ cancel-on-input
+ cancel-on-input-retval)
"Make a request to CONNECTION, wait for a reply.
Like `jsonrpc-async-request' for CONNECTION, METHOD and PARAMS,
-but synchronous, i.e. this function doesn't exit until anything
-interesting (success, error or timeout) happens. Furthermore, it
-only exits locally (returning the JSONRPC result object) if the
-request is successful, otherwise exit non-locally with an error
-of type `jsonrpc-error'.
+but synchronous.
-DEFERRED is passed to `jsonrpc-async-request', which see."
+Except in the case of a non-nil CANCEL-ON-INPUT (explained
+below), this function doesn't exit until anything interesting
+happens (success reply, error reply, or timeout). Furthermore,
+it only exits locally (returning the JSONRPC result object) if
+the request is successful, otherwise it exits non-locally with an
+error of type `jsonrpc-error'.
+
+DEFERRED is passed to `jsonrpc-async-request', which see.
+
+If CANCEL-ON-INPUT is non-nil and the user inputs something while
+the functino is waiting, then it exits immediately, returning
+CANCEL-ON-INPUT-RETVAL. Any future replies (normal or error) are
+ignored."
(let* ((tag (cl-gensym "jsonrpc-request-catch-tag")) id-and-timer
+ cancelled
(retval
(unwind-protect ; protect against user-quit, for example
(catch tag
@@ -274,19 +289,27 @@ DEFERRED is passed to `jsonrpc-async-request', which see."
id-and-timer
(jsonrpc--async-request-1
connection method params
- :success-fn (lambda (result) (throw tag `(done ,result)))
+ :success-fn (lambda (result)
+ (unless cancelled
+ (throw tag `(done ,result))))
:error-fn
(jsonrpc-lambda
(&key code message data)
- (throw tag `(error (jsonrpc-error-code . ,code)
- (jsonrpc-error-message . ,message)
- (jsonrpc-error-data . ,data))))
+ (unless cancelled
+ (throw tag `(error (jsonrpc-error-code . ,code)
+ (jsonrpc-error-message . ,message)
+ (jsonrpc-error-data . ,data)))))
:timeout-fn
(lambda ()
- (throw tag '(error (jsonrpc-error-message . "Timed out"))))
+ (unless cancelled
+ (throw tag '(error (jsonrpc-error-message . "Timed
out")))))
:deferred deferred
:timeout timeout))
- (while t (accept-process-output nil 30)))
+ (cond (cancel-on-input
+ (while (sit-for 30))
+ (setq cancelled t)
+ `(cancelled ,cancel-on-input-retval))
+ (t (while t (accept-process-output nil 30)))))
(pcase-let* ((`(,id ,timer) id-and-timer))
(remhash id (jsonrpc--request-continuations connection))
(remhash (list deferred (current-buffer))
@@ -641,15 +664,26 @@ originated."
(if type
(format "-%s" subtype)))))
(goto-char (point-max))
- (let ((msg (format "%s%s%s %s:\n%s\n"
- type
- (if id (format " (id:%s)" id) "")
- (if error " ERROR" "")
- (current-time-string)
- (pp-to-string message))))
- (when error
- (setq msg (propertize msg 'face 'error)))
- (insert-before-markers msg))))))
+ (prog1
+ (let ((msg (format "%s%s%s %s:\n%s\n"
+ type
+ (if id (format " (id:%s)" id) "")
+ (if error " ERROR" "")
+ (current-time-string)
+ (pp-to-string message))))
+ (when error
+ (setq msg (propertize msg 'face 'error)))
+ (insert-before-markers msg))
+ ;; Trim the buffer if it's too large
+ (let ((max (jsonrpc--events-buffer-scrollback-size connection)))
+ (when max
+ (save-excursion
+ (goto-char (point-min))
+ (while (> (buffer-size) max)
+ (delete-region (point) (progn (forward-line 1)
+ (forward-sexp 1)
+ (forward-line 2)
+ (point))))))))))))
(provide 'jsonrpc)
;;; jsonrpc.el ends here
diff --git a/lisp/language/thai.el b/lisp/language/thai.el
index a896fe5..c655845 100644
--- a/lisp/language/thai.el
+++ b/lisp/language/thai.el
@@ -36,7 +36,7 @@
"8-bit encoding for ASCII (MSB=0) and Thai TIS620 (MSB=1)."
:coding-type 'charset
:mnemonic ?T
- :charset-list '(tis620-2533))
+ :charset-list '(thai-iso8859-11))
(define-coding-system-alias 'th-tis620 'thai-tis620)
(define-coding-system-alias 'tis620 'thai-tis620)
@@ -47,7 +47,7 @@
(charset thai-tis620)
(coding-system thai-tis620 iso-8859-11 cp874)
(coding-priority thai-tis620)
- (nonascii-translation . tis620-2533)
+ (nonascii-translation . iso-8859-11)
(input-method . "thai-kesmanee")
(unibyte-display . thai-tis620)
(features thai-util)
diff --git a/lisp/ldefs-boot.el b/lisp/ldefs-boot.el
index 5f26eba..3bd775f 100644
--- a/lisp/ldefs-boot.el
+++ b/lisp/ldefs-boot.el
@@ -176,12 +176,18 @@ Optional arg BUFFER-FILE overrides `buffer-file-name'.
\(fn &optional FILE-NAME BUFFER-FILE)" nil nil)
(autoload 'add-change-log-entry "add-log" "\
-Find change log file, and add an entry for today and an item for this file.
-Optional arg WHOAMI (interactive prefix) non-nil means prompt for user
-name and email (stored in `add-log-full-name' and `add-log-mailing-address').
-
-Second arg FILE-NAME is file name of the change log.
-If nil, use the value of `change-log-default-name'.
+Find ChangeLog buffer, add an entry for today and an item for this file.
+Optional arg WHOAMI (interactive prefix) non-nil means prompt for
+user name and email (stored in `add-log-full-name'
+and `add-log-mailing-address').
+
+Second arg CHANGELOG-FILE-NAME is the file name of the change log.
+If nil, use the value of `change-log-default-name'. If the file
+thus named exists, it is used for the new entry. If it doesn't
+exist, it is created, unless `add-log-dont-create-changelog-file' is t,
+in which case a suitably named buffer that doesn't visit any file
+is used for keeping entries pertaining to CHANGELOG-FILE-NAME's
+directory.
Third arg OTHER-WINDOW non-nil means visit in other window.
@@ -204,7 +210,7 @@ notices.
Today's date is calculated according to `add-log-time-zone-rule' if
non-nil, otherwise in local time.
-\(fn &optional WHOAMI FILE-NAME OTHER-WINDOW NEW-ENTRY
PUT-NEW-ENTRY-ON-NEW-LINE)" t nil)
+\(fn &optional WHOAMI CHANGELOG-FILE-NAME OTHER-WINDOW NEW-ENTRY
PUT-NEW-ENTRY-ON-NEW-LINE)" t nil)
(autoload 'add-change-log-entry-other-window "add-log" "\
Find change log file in other window and add entry and item.
@@ -577,9 +583,11 @@ Return t if `allout-mode' is active in current buffer.
(autoload 'allout-mode "allout" "\
Toggle Allout outline mode.
-With a prefix argument ARG, enable Allout outline mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Allout mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\\<allout-mode-map-value>
Allout outline mode is a minor mode that provides extensive
@@ -890,9 +898,11 @@ See `allout-widgets-mode' for allout widgets mode
features.")
(autoload 'allout-widgets-mode "allout-widgets" "\
Toggle Allout Widgets mode.
-With a prefix argument ARG, enable Allout Widgets mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Allout-Widgets mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Allout Widgets mode is an extension of Allout mode that provides
graphical decoration of outline structure. It is meant to
@@ -1300,7 +1310,12 @@ Entering array mode calls the function `array-mode-hook'.
(autoload 'artist-mode "artist" "\
Toggle Artist mode.
-With argument ARG, turn Artist mode on if ARG is positive.
+
+If called interactively, enable Artist mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
Artist lets you draw lines, squares, rectangles and poly-lines,
ellipses and circles with your mouse and/or keyboard.
@@ -1571,9 +1586,6 @@ for a description of this minor mode.")
(autoload 'autoarg-mode "autoarg" "\
Toggle Autoarg mode, a global minor mode.
-With a prefix argument ARG, enable Autoarg mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
\\<autoarg-mode-map>
In Autoarg mode, digits are bound to `digit-argument', i.e. they
@@ -1607,9 +1619,11 @@ or call the function `autoarg-kp-mode'.")
(autoload 'autoarg-kp-mode "autoarg" "\
Toggle Autoarg-KP mode, a global minor mode.
-With a prefix argument ARG, enable Autoarg-KP mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Autoarg-Kp mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\\<autoarg-kp-mode-map>
This is similar to `autoarg-mode' but rebinds the keypad keys
@@ -1663,9 +1677,11 @@ or call the function `auto-insert-mode'.")
(autoload 'auto-insert-mode "autoinsert" "\
Toggle Auto-insert mode, a global minor mode.
-With a prefix argument ARG, enable Auto-insert mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Auto-Insert mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Auto-insert mode is enabled, when new files are created you can
insert a template for the file depending on the mode of the buffer.
@@ -1735,9 +1751,11 @@ should be non-nil).
(autoload 'auto-revert-mode "autorevert" "\
Toggle reverting buffer when the file changes (Auto-Revert Mode).
-With a prefix argument ARG, enable Auto-Revert Mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Auto-Revert mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Auto-Revert Mode is a minor mode that affects only the current
buffer. When enabled, it reverts the buffer when the file on
@@ -1762,9 +1780,11 @@ This function is designed to be added to hooks, for
example:
(autoload 'auto-revert-tail-mode "autorevert" "\
Toggle reverting tail of buffer when the file grows.
-With a prefix argument ARG, enable Auto-Revert Tail Mode if ARG
-is positive, and disable it otherwise. If called from Lisp,
-enable the mode if ARG is omitted or nil.
+
+If called interactively, enable Auto-Revert-Tail mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Auto-Revert Tail Mode is enabled, the tail of the file is
constantly followed, as with the shell command `tail -f'. This
@@ -1803,9 +1823,11 @@ or call the function `global-auto-revert-mode'.")
(autoload 'global-auto-revert-mode "autorevert" "\
Toggle Global Auto-Revert Mode.
-With a prefix argument ARG, enable Global Auto-Revert Mode if ARG
-is positive, and disable it otherwise. If called from Lisp,
-enable the mode if ARG is omitted or nil.
+
+If called interactively, enable Global Auto-Revert mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Global Auto-Revert Mode is a global minor mode that reverts any
buffer associated with a file when the file changes on disk. Use
@@ -1921,9 +1943,11 @@ or call the function `display-battery-mode'.")
(autoload 'display-battery-mode "battery" "\
Toggle battery status display in mode line (Display Battery mode).
-With a prefix argument ARG, enable Display Battery mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Display-Battery mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
The text displayed in the mode line is controlled by
`battery-mode-line-format' and `battery-status-function'.
@@ -2331,7 +2355,7 @@ BOOKMARK is usually a bookmark name (a string). It can
also be a
bookmark record, but this is usually only done by programmatic callers.
If DISPLAY-FUNC is non-nil, it is a function to invoke to display the
-bookmark. It defaults to `switch-to-buffer'. A typical value for
+bookmark. It defaults to `pop-to-buffer-same-window'. A typical value for
DISPLAY-FUNC would be `switch-to-buffer-other-window'.
\(fn BOOKMARK &optional DISPLAY-FUNC)" t nil)
@@ -2897,15 +2921,22 @@ columns on its right towards the left.
(autoload 'bug-reference-mode "bug-reference" "\
Toggle hyperlinking bug references in the buffer (Bug Reference mode).
-With a prefix argument ARG, enable Bug Reference mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Bug-Reference mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
(autoload 'bug-reference-prog-mode "bug-reference" "\
Like `bug-reference-mode', but only buttonize in comments and strings.
+If called interactively, enable Bug-Reference-Prog mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(if (fboundp 'register-definition-prefixes) (register-definition-prefixes
"bug-reference" '("bug-reference-")))
@@ -4691,9 +4722,11 @@ Prefix argument is the same as for `checkdoc-defun'
(autoload 'checkdoc-minor-mode "checkdoc" "\
Toggle automatic docstring checking (Checkdoc minor mode).
-With a prefix argument ARG, enable Checkdoc minor mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Checkdoc minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
In Checkdoc minor mode, the usual bindings for `eval-defun' which is
bound to \\<checkdoc-minor-mode-map>\\[checkdoc-eval-defun] and
`checkdoc-eval-current-buffer' are overridden to include
@@ -4933,6 +4966,11 @@ This can be needed when using code byte-compiled using
the old
macro-expansion of `cl-defstruct' that used vectors objects instead
of record objects.
+If called interactively, enable Cl-Old-Struct-Compat mode if ARG is positive,
and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(if (fboundp 'register-definition-prefixes) (register-definition-prefixes
"cl-lib" '("cl-")))
@@ -5150,7 +5188,7 @@ REGEXP-GROUP is the regular expression group in REGEXP to
use.
\(fn PROCESS COMMAND REGEXP REGEXP-GROUP)" nil nil)
-(if (fboundp 'register-definition-prefixes) (register-definition-prefixes
"comint" '("comint-" "send-invisible" "shell-strip-ctrl-m")))
+(if (fboundp 'register-definition-prefixes) (register-definition-prefixes
"comint" '("comint-")))
;;;***
@@ -5346,9 +5384,11 @@ Runs `compilation-mode-hook' with `run-mode-hooks'
(which see).
(autoload 'compilation-shell-minor-mode "compile" "\
Toggle Compilation Shell minor mode.
-With a prefix argument ARG, enable Compilation Shell minor mode
-if ARG is positive, and disable it otherwise. If called from
-Lisp, enable the mode if ARG is omitted or nil.
+
+If called interactively, enable Compilation-Shell minor mode if ARG is
positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Compilation Shell minor mode is enabled, all the
error-parsing commands of the Compilation major mode are
@@ -5359,9 +5399,11 @@ See `compilation-mode'.
(autoload 'compilation-minor-mode "compile" "\
Toggle Compilation minor mode.
-With a prefix argument ARG, enable Compilation minor mode if ARG
-is positive, and disable it otherwise. If called from Lisp,
-enable the mode if ARG is omitted or nil.
+
+If called interactively, enable Compilation minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Compilation minor mode is enabled, all the error-parsing
commands of Compilation major mode are available. See
@@ -5394,9 +5436,11 @@ or call the function `dynamic-completion-mode'.")
(autoload 'dynamic-completion-mode "completion" "\
Toggle dynamic word-completion on or off.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Dynamic-Completion mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
@@ -5959,9 +6003,11 @@ or call the function `cua-mode'.")
(autoload 'cua-mode "cua-base" "\
Toggle Common User Access style editing (CUA mode).
-With a prefix argument ARG, enable CUA mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Cua mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
CUA mode is a global minor mode. When enabled, typed text
replaces the active selection, and you can use C-z, C-x, C-c, and
@@ -6006,6 +6052,11 @@ Enable CUA selection mode without the C-z/C-x/C-c/C-v
bindings.
Toggle the region as rectangular.
Activates the region if needed. Only lasts until the region is deactivated.
+If called interactively, enable Cua-Rectangle-Mark mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(if (fboundp 'register-definition-prefixes) (register-definition-prefixes
"cua-rect" '("cua-")))
@@ -6021,6 +6072,11 @@ Activates the region if needed. Only lasts until the
region is deactivated.
(autoload 'cursor-intangible-mode "cursor-sensor" "\
Keep cursor outside of any `cursor-intangible' text property.
+If called interactively, enable Cursor-Intangible mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(autoload 'cursor-sensor-mode "cursor-sensor" "\
@@ -6031,6 +6087,11 @@ where WINDOW is the affected window, OLDPOS is the last
known position of
the cursor and DIR can be `entered' or `left' depending on whether the cursor
is entering the area covered by the text-property property or leaving it.
+If called interactively, enable Cursor-Sensor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(if (fboundp 'register-definition-prefixes) (register-definition-prefixes
"cursor-sensor" '("cursor-sensor-")))
@@ -6421,16 +6482,17 @@ Mode used for cvs status output.
(autoload 'cwarn-mode "cwarn" "\
Minor mode that highlights suspicious C and C++ constructions.
+If called interactively, enable Cwarn mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
Suspicious constructs are highlighted using `font-lock-warning-face'.
Note, in addition to enabling this minor mode, the major mode must
be included in the variable `cwarn-configuration'. By default C and
C++ modes are included.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
-
\(fn &optional ARG)" t nil)
(define-obsolete-function-alias 'turn-on-cwarn-mode 'cwarn-mode "24.1")
@@ -6849,12 +6911,11 @@ or call the function `delete-selection-mode'.")
(autoload 'delete-selection-mode "delsel" "\
Toggle Delete Selection mode.
-Interactively, with a prefix argument, enable
-Delete Selection mode if the prefix argument is positive,
-and disable it otherwise. If called from Lisp, toggle
-the mode if ARG is `toggle', disable the mode if ARG is
-a non-positive integer, and enable the mode otherwise
-\(including if ARG is omitted or nil or a positive integer).
+
+If called interactively, enable Delete-Selection mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Delete Selection mode is enabled, typed text replaces the selection
if the selection is active. Otherwise, typed text is just inserted at
@@ -7006,9 +7067,11 @@ or call the function `desktop-save-mode'.")
(autoload 'desktop-save-mode "desktop" "\
Toggle desktop saving (Desktop Save mode).
-With a prefix argument ARG, enable Desktop Save mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode if ARG
-is omitted or nil.
+
+If called interactively, enable Desktop-Save mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Desktop Save mode is enabled, the state of Emacs is saved from
one session to another. In particular, Emacs will save the desktop when
@@ -7371,9 +7434,11 @@ a diff with \\[diff-reverse-direction].
(autoload 'diff-minor-mode "diff-mode" "\
Toggle Diff minor mode.
-With a prefix argument ARG, enable Diff minor mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Diff minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\\{diff-minor-mode-map}
@@ -7549,9 +7614,11 @@ Keybindings:
(autoload 'dirtrack-mode "dirtrack" "\
Toggle directory tracking in shell buffers (Dirtrack mode).
-With a prefix argument ARG, enable Dirtrack mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Dirtrack mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
This method requires that your shell prompt contain the current
working directory at all times, and that you set the variable
@@ -7723,6 +7790,11 @@ in `.emacs'.
Toggle display of line numbers in the buffer.
This uses `display-line-numbers' internally.
+If called interactively, enable Display-Line-Numbers mode if ARG is positive,
and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
To change the type of line numbers displayed by default,
customize `display-line-numbers-type'. To change the type while
the mode is on, set `display-line-numbers' directly.
@@ -7856,9 +7928,11 @@ to the next best mode.
(autoload 'doc-view-minor-mode "doc-view" "\
Toggle displaying buffer via Doc View (Doc View minor mode).
-With a prefix argument ARG, enable Doc View minor mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Doc-View minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
See the command `doc-view-mode' for more information on this mode.
@@ -7918,9 +7992,11 @@ Switch to *doctor* buffer and start giving psychotherapy.
(autoload 'double-mode "double" "\
Toggle special insertion on double keypresses (Double mode).
-With a prefix argument ARG, enable Double mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Double mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Double mode is enabled, some keys will insert different
strings when pressed twice. See `double-map' for details.
@@ -7975,7 +8051,9 @@ non-positive integer, and enables the mode otherwise
(including
if the argument is omitted or nil or a positive integer).
If DOC is nil, give the mode command a basic doc-string
-documenting what its argument does.
+documenting what its argument does. If the word \"ARG\" does not
+appear in DOC, a paragraph is added to DOC explaining
+usage of the mode argument.
Optional INIT-VALUE is the initial value of the mode's variable.
Optional LIGHTER is displayed in the mode line when the mode is on.
@@ -8785,9 +8863,11 @@ or call the function `global-ede-mode'.")
(autoload 'global-ede-mode "ede" "\
Toggle global EDE (Emacs Development Environment) mode.
-With a prefix argument ARG, enable global EDE mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Global Ede mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
This global minor mode enables `ede-minor-mode' in all buffers in
an EDE controlled project.
@@ -9797,9 +9877,11 @@ or call the function `electric-pair-mode'.")
(autoload 'electric-pair-mode "elec-pair" "\
Toggle automatic parens pairing (Electric Pair mode).
-With a prefix argument ARG, enable Electric Pair mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Electric-Pair mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Electric Pair mode is a global minor mode. When enabled, typing
an open parenthesis automatically inserts the corresponding
@@ -9814,6 +9896,11 @@ To toggle the mode in a single buffer, use
`electric-pair-local-mode'.
(autoload 'electric-pair-local-mode "elec-pair" "\
Toggle `electric-pair-mode' only in this buffer.
+If called interactively, enable Electric-Pair-Local mode if ARG is positive,
and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(if (fboundp 'register-definition-prefixes) (register-definition-prefixes
"elec-pair" '("electric-pair-")))
@@ -10053,9 +10140,7 @@ displayed.
(autoload 'emacs-lock-mode "emacs-lock" "\
Toggle Emacs Lock mode in the current buffer.
If called with a plain prefix argument, ask for the locking mode
-to be used. With any other prefix ARG, turn mode on if ARG is
-positive, off otherwise. If called from Lisp, enable the mode if
-ARG is omitted or nil.
+to be used.
Initially, if the user does not pass an explicit locking mode, it
defaults to `emacs-lock-default-locking-mode' (which see);
@@ -10070,6 +10155,9 @@ When called from Elisp code, ARG can be any locking
mode:
Other values are interpreted as usual.
+See also `emacs-lock-unlockable-modes', which exempts buffers under
+some major modes from being locked under some circumstances.
+
\(fn &optional ARG)" t nil)
(if (fboundp 'register-definition-prefixes) (register-definition-prefixes
"emacs-lock" '("emacs-lock-" "toggle-emacs-lock")))
@@ -10161,9 +10249,10 @@ Minor mode for editing text/enriched files.
These are files with embedded formatting information in the MIME standard
text/enriched format.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+If called interactively, enable Enriched mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Turning the mode on or off runs `enriched-mode-hook'.
@@ -10432,9 +10521,11 @@ Encrypt marked files.
(autoload 'epa-mail-mode "epa-mail" "\
A minor-mode for composing encrypted/clearsigned mails.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable epa-mail mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
@@ -10497,9 +10588,11 @@ or call the function `epa-global-mail-mode'.")
(autoload 'epa-global-mail-mode "epa-mail" "\
Minor mode to hook EasyPG into Mail mode.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Epa-Global-Mail mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
@@ -10545,8 +10638,13 @@ Return a list of internal configuration parameters of
`epg-gpg-program'.
(autoload 'epg-check-configuration "epg-config" "\
Verify that a sufficient version of GnuPG is installed.
+CONFIG should be a `epg-configuration' object (a plist).
+REQ-VERSIONS should be a list with elements of the form (MIN
+. MAX) where MIN and MAX are version strings indicating a
+semi-open range of acceptable versions. REQ-VERSIONS may also be
+a single minimum version string.
-\(fn CONFIG &optional MINIMUM-VERSION)" nil nil)
+\(fn CONFIG &optional REQ-VERSIONS)" nil nil)
(autoload 'epg-expand-group "epg-config" "\
Look at CONFIG and try to expand GROUP.
@@ -12087,10 +12185,14 @@ a top-level keymap, `text-scale-increase' or
(autoload 'buffer-face-mode "face-remap" "\
Minor mode for a buffer-specific default face.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil. When enabled, the face specified by the
-variable `buffer-face-mode-face' is used to display the buffer text.
+
+If called interactively, enable Buffer-Face mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
+When enabled, the face specified by the variable
+`buffer-face-mode-face' is used to display the buffer text.
\(fn &optional ARG)" t nil)
@@ -12972,9 +13074,11 @@ region is invalid.
(autoload 'flymake-mode "flymake" "\
Toggle Flymake mode on or off.
-With a prefix argument ARG, enable Flymake mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil, and toggle it if ARG is `toggle'.
+
+If called interactively, enable Flymake mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Flymake is an Emacs minor mode for on-the-fly syntax checking.
Flymake collects diagnostic information from multiple sources,
@@ -13060,9 +13164,11 @@ Turn on `flyspell-mode' for comments and strings.
(autoload 'flyspell-mode "flyspell" "\
Toggle on-the-fly spell checking (Flyspell mode).
-With a prefix argument ARG, enable Flyspell mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Flyspell mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Flyspell mode is a buffer-local minor mode. When enabled, it
spawns a single Ispell process and checks each word. The default
@@ -13110,6 +13216,9 @@ Turn Flyspell mode off.
(autoload 'flyspell-region "flyspell" "\
Flyspell text between BEG and END.
+Make sure `flyspell-mode' is turned on if you want the highlight
+of a misspelled word removed when you've corrected it.
+
\(fn BEG END)" t nil)
(autoload 'flyspell-buffer "flyspell" "\
@@ -13144,9 +13253,11 @@ Turn off Follow mode. Please see the function
`follow-mode'.
(autoload 'follow-mode "follow" "\
Toggle Follow mode.
-With a prefix argument ARG, enable Follow mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Follow mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Follow mode is a minor mode that combines windows into one tall
virtual window. This is accomplished by two main techniques:
@@ -13267,9 +13378,11 @@ selected if the original window is the first one in
the frame.
(autoload 'footnote-mode "footnote" "\
Toggle Footnote mode.
-With a prefix argument ARG, enable Footnote mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Footnote mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Footnote mode is a buffer-local minor mode. If enabled, it
provides footnote support for `message-mode'. To get started,
@@ -13691,6 +13804,11 @@ being transferred. This list may grow up to a size of
`gdb-debug-log-max' after which the oldest element (at the end of
the list) is deleted every time a new one is added (at the front).
+If called interactively, enable Gdb-Enable-Debug mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(autoload 'gdb "gdb-mi" "\
@@ -13859,10 +13977,14 @@ regular expression that can be used as an element of
(autoload 'glasses-mode "glasses" "\
Minor mode for making identifiers likeThis readable.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil. When this mode is active, it tries to
-add virtual separators (like underscores) at places they belong to.
+
+If called interactively, enable Glasses mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
+When this mode is active, it tries to add virtual
+separators (like underscores) at places they belong to.
\(fn &optional ARG)" t nil)
@@ -14469,6 +14591,11 @@ If FORCE is non-nil, replace the old ones.
(autoload 'gnus-mailing-list-mode "gnus-ml" "\
Minor mode for providing mailing-list commands.
+If called interactively, enable Gnus-Mailing-List mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\\{gnus-mailing-list-mode-map}
\(fn &optional ARG)" t nil)
@@ -14889,7 +15016,14 @@ number with fewer than this number of bits, the
handshake is
rejected. (The smaller the prime number, the less secure the
key exchange is against man-in-the-middle attacks.)
-A value of nil says to use the default GnuTLS value.")
+A value of nil says to use the default GnuTLS value.
+
+The default value of this variable is such that virtually any
+connection can be established, whether this connection can be
+considered cryptographically \"safe\" or not. However, Emacs
+network security is handled at a higher level via
+`open-network-stream' and the Network Security Manager. See Info
+node `(emacs) Network Security'.")
(custom-autoload 'gnutls-min-prime-bits "gnutls" t)
@@ -14951,15 +15085,22 @@ Also fontifies the buffer appropriately (see
`goto-address-fontify-p' and
(autoload 'goto-address-mode "goto-addr" "\
Minor mode to buttonize URLs and e-mail addresses in the current buffer.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Goto-Address mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
(autoload 'goto-address-prog-mode "goto-addr" "\
Like `goto-address-mode', but only for comments and strings.
+If called interactively, enable Goto-Address-Prog mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(if (fboundp 'register-definition-prefixes) (register-definition-prefixes
"goto-addr" '("goto-address-")))
@@ -15017,7 +15158,7 @@ List of hook functions run by `grep-process-setup' (see
`run-hooks').")
(custom-autoload 'grep-setup-hook "grep" t)
-(defconst grep-regexp-alist `((,(concat "^\\(?:"
"\\(?1:[^\0\n]+\\)\\(?3:\0\\)\\(?2:[0-9]+\\):" "\\|"
"\\(?1:[^\n:]+?[^\n/:]\\):[\11 ]*\\(?2:[1-9][0-9]*\\)[\11 ]*:" "\\)") 1 2
(,(lambda nil (when grep-highlight-matches (let* ((beg (match-end 0)) (end
(save-excursion (goto-char beg) (line-end-position))) (mbeg (text-property-any
beg end 'font-lock-face 'grep-match-face))) (when mbeg (- mbeg beg))))) \,
(lambda nil (when grep-highlight-matches (let* ((beg (match-end 0)) (end
(save-excursio [...]
+(defconst grep-regexp-alist `((,(concat "^\\(?:"
"\\(?1:[^\0\n]+\\)\\(?3:\0\\)\\(?2:[0-9]+\\):" "\\|" "\\(?1:"
"\\(?:[a-zA-Z]:\\)?" "[^\n:]+?[^\n/:]\\):[\11 ]*\\(?2:[1-9][0-9]*\\)[\11 ]*:"
"\\)") 1 2 (,(lambda nil (when grep-highlight-matches (let* ((beg (match-end
0)) (end (save-excursion (goto-char beg) (line-end-position))) (mbeg
(text-property-any beg end 'font-lock-face 'grep-match-face))) (when mbeg (-
mbeg beg))))) \, (lambda nil (when grep-highlight-matches (let* ((beg
(match-end [...]
Regexp used to match grep hits.
See `compilation-error-regexp-alist' for format details.")
@@ -15259,9 +15400,11 @@ or call the function `gud-tooltip-mode'.")
(autoload 'gud-tooltip-mode "gud" "\
Toggle the display of GUD tooltips.
-With a prefix argument ARG, enable the feature if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-it if ARG is omitted or nil.
+
+If called interactively, enable Gud-Tooltip mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
@@ -15944,9 +16087,11 @@ This discards the buffer's undo information.
(autoload 'hi-lock-mode "hi-lock" "\
Toggle selective highlighting of patterns (Hi Lock mode).
-With a prefix argument ARG, enable Hi Lock mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Hi-Lock mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Hi Lock mode is automatically enabled when you invoke any of the
highlighting commands listed below, such as \\[highlight-regexp].
@@ -16114,9 +16259,11 @@ be found in variable `hi-lock-interactive-patterns'.
(autoload 'hide-ifdef-mode "hideif" "\
Toggle features to hide/show #ifdef blocks (Hide-Ifdef mode).
-With a prefix argument ARG, enable Hide-Ifdef mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Hide-Ifdef mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Hide-Ifdef mode is a buffer-local minor mode for use with C and
C-like major modes. When enabled, code within #ifdef constructs
@@ -16191,9 +16338,11 @@ whitespace. Case does not matter.")
(autoload 'hs-minor-mode "hideshow" "\
Minor mode to selectively hide/show code and comment blocks.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Hs minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When hideshow minor mode is on, the menu bar is augmented with hideshow
commands and the hideshow commands are enabled.
@@ -16227,9 +16376,11 @@ Unconditionally turn off `hs-minor-mode'.
(autoload 'highlight-changes-mode "hilit-chg" "\
Toggle highlighting changes in this buffer (Highlight Changes mode).
-With a prefix argument ARG, enable Highlight Changes mode if ARG
-is positive, and disable it otherwise. If called from Lisp,
-enable the mode if ARG is omitted or nil.
+
+If called interactively, enable Highlight-Changes mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Highlight Changes is enabled, changes are marked with a text
property. Normally they are displayed in a distinctive face, but
@@ -16250,9 +16401,11 @@ buffer with the contents of a file
(autoload 'highlight-changes-visible-mode "hilit-chg" "\
Toggle visibility of highlighting due to Highlight Changes mode.
-With a prefix argument ARG, enable Highlight Changes Visible mode
-if ARG is positive, and disable it otherwise. If called from
-Lisp, enable the mode if ARG is omitted or nil.
+
+If called interactively, enable Highlight-Changes-Visible mode if ARG is
positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Highlight Changes Visible mode only has an effect when Highlight
Changes mode is on. When enabled, the changed text is displayed
@@ -16395,9 +16548,11 @@ argument VERBOSE non-nil makes the function verbose.
(autoload 'hl-line-mode "hl-line" "\
Toggle highlighting of the current line (Hl-Line mode).
-With a prefix argument ARG, enable Hl-Line mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Hl-Line mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Hl-Line mode is a buffer-local minor mode. If
`hl-line-sticky-flag' is non-nil, Hl-Line mode highlights the
@@ -16425,9 +16580,11 @@ or call the function `global-hl-line-mode'.")
(autoload 'global-hl-line-mode "hl-line" "\
Toggle line highlighting in all buffers (Global Hl-Line mode).
-With a prefix argument ARG, enable Global Hl-Line mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Global Hl-Line mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
If `global-hl-line-sticky-flag' is non-nil, Global Hl-Line mode
highlights the line about the current buffer's point in all live
@@ -16841,9 +16998,11 @@ or call the function `icomplete-mode'.")
(autoload 'icomplete-mode "icomplete" "\
Toggle incremental minibuffer completion (Icomplete mode).
-With a prefix argument ARG, enable Icomplete mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Icomplete mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When this global minor mode is enabled, typing in the minibuffer
continuously displays a list of possible completions that match
@@ -17392,10 +17551,11 @@ DEF, if non-nil, is the default value.
(autoload 'ielm "ielm" "\
Interactively evaluate Emacs Lisp expressions.
-Switches to the buffer `*ielm*', or creates it if it does not exist.
+Switches to the buffer named BUF-NAME if provided (`*ielm*' by default),
+or creates it if it does not exist.
See `inferior-emacs-lisp-mode' for details.
-\(fn)" t nil)
+\(fn &optional BUF-NAME)" t nil)
(if (fboundp 'register-definition-prefixes) (register-definition-prefixes
"ielm" '("ielm-" "inferior-emacs-lisp-mode")))
@@ -17415,9 +17575,12 @@ See `inferior-emacs-lisp-mode' for details.
(autoload 'iimage-mode "iimage" "\
Toggle Iimage mode on or off.
-With a prefix argument ARG, enable Iimage mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil, and toggle it if ARG is `toggle'.
+
+If called interactively, enable Iimage mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\\{iimage-mode-map}
\(fn &optional ARG)" t nil)
@@ -17710,6 +17873,11 @@ Setup easy-to-use keybindings for the commands to be
used in dired mode.
Note that n, p and <down> and <up> will be hijacked and bound to
`image-dired-dired-x-line'.
+If called interactively, enable Image-Dired minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(define-obsolete-function-alias 'image-dired-setup-dired-keybindings
'image-dired-minor-mode "26.1")
@@ -17813,9 +17981,11 @@ or call the function `auto-image-file-mode'.")
(autoload 'auto-image-file-mode "image-file" "\
Toggle visiting of image files as images (Auto Image File mode).
-With a prefix argument ARG, enable Auto Image File mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Auto-Image-File mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
An image file is one whose name has an extension in
`image-file-name-extensions', or matches a regexp in
@@ -17842,9 +18012,11 @@ Key bindings:
(autoload 'image-minor-mode "image-mode" "\
Toggle Image minor mode in this buffer.
-With a prefix argument ARG, enable Image minor mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Image minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Image minor mode provides the key \\<image-mode-map>\\[image-toggle-display],
to switch back to `image-mode' and display an image file as the
@@ -17907,9 +18079,9 @@ string (which specifies the title of a submenu into
which the
matches are put).
REGEXP is a regular expression matching a definition construct
which is to be displayed in the menu. REGEXP may also be a
-function, called without arguments. It is expected to search
-backwards. It must return true and set `match-data' if it finds
-another element.
+function of no arguments. If REGEXP is a function, it is
+expected to search backwards, return non-nil if it finds a
+definition construct, and set `match-data' for that construct.
INDEX is an integer specifying which subexpression of REGEXP
matches the definition's name; this subexpression is displayed as
the menu item.
@@ -18824,9 +18996,11 @@ available on the net.
(autoload 'ispell-minor-mode "ispell" "\
Toggle last-word spell checking (Ispell minor mode).
-With a prefix argument ARG, enable Ispell minor mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable ISpell minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Ispell minor mode is a buffer-local minor mode. When enabled,
typing SPC or RET warns you if the previous word is incorrectly
@@ -19510,9 +19684,11 @@ generations (this defaults to 1).
(autoload 'linum-mode "linum" "\
Toggle display of line numbers in the left margin (Linum mode).
-With a prefix argument ARG, enable Linum mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Linum mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Linum mode is a buffer-local minor mode.
@@ -20085,9 +20261,11 @@ or call the function `mail-abbrevs-mode'.")
(autoload 'mail-abbrevs-mode "mailabbrev" "\
Toggle abbrev expansion of mail aliases (Mail Abbrevs mode).
-With a prefix argument ARG, enable Mail Abbrevs mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Mail-Abbrevs mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Mail Abbrevs mode is a global minor mode. When enabled,
abbrev-like expansion is performed when editing certain mail
@@ -20431,9 +20609,11 @@ Default bookmark handler for Man buffers.
(autoload 'master-mode "master" "\
Toggle Master mode.
-With a prefix argument ARG, enable Master mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Master mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Master mode is enabled, you can scroll the slave buffer
using the following commands:
@@ -20465,9 +20645,11 @@ or call the function
`minibuffer-depth-indicate-mode'.")
(autoload 'minibuffer-depth-indicate-mode "mb-depth" "\
Toggle Minibuffer Depth Indication mode.
-With a prefix argument ARG, enable Minibuffer Depth Indication
-mode if ARG is positive, and disable it otherwise. If called
-from Lisp, enable the mode if ARG is omitted or nil.
+
+If called interactively, enable Minibuffer-Depth-Indicate mode if ARG is
positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Minibuffer Depth Indication mode is a global minor mode. When
enabled, any recursive use of the minibuffer will show the
@@ -21095,6 +21277,11 @@ or call the function `midnight-mode'.")
(autoload 'midnight-mode "midnight" "\
Non-nil means run `midnight-hook' at midnight.
+If called interactively, enable Midnight mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(autoload 'clean-buffer-list "midnight" "\
@@ -21137,9 +21324,11 @@ or call the function
`minibuffer-electric-default-mode'.")
(autoload 'minibuffer-electric-default-mode "minibuf-eldef" "\
Toggle Minibuffer Electric Default mode.
-With a prefix argument ARG, enable Minibuffer Electric Default
-mode if ARG is positive, and disable it otherwise. If called
-from Lisp, enable the mode if ARG is omitted or nil.
+
+If called interactively, enable Minibuffer-Electric-Default mode if ARG is
positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Minibuffer Electric Default mode is a global minor mode. When
enabled, minibuffer prompts that show a default value only show
@@ -21722,9 +21911,11 @@ or call the function `msb-mode'.")
(autoload 'msb-mode "msb" "\
Toggle Msb mode.
-With a prefix argument ARG, enable Msb mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Msb mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
This mode overrides the binding(s) of `mouse-buffer-menu' to provide a
different buffer menu using the function `msb'.
@@ -23390,6 +23581,11 @@ modes. The following keys behave as if Org mode were
active, if
the cursor is on a headline, or on a plain list item (both as
defined by Org mode).
+If called interactively, enable OrgStruct mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(autoload 'turn-on-orgstruct "org" "\
@@ -24302,9 +24498,11 @@ Turning on outline mode calls the value of
`text-mode-hook' and then of
(autoload 'outline-minor-mode "outline" "\
Toggle Outline minor mode.
-With a prefix argument ARG, enable Outline minor mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Outline minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
See the command `outline-mode' for more information on this mode.
@@ -24579,9 +24777,11 @@ or call the function `show-paren-mode'.")
(autoload 'show-paren-mode "paren" "\
Toggle visualization of matching parens (Show Paren mode).
-With a prefix argument ARG, enable Show Paren mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Show-Paren mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Show Paren mode is a global minor mode. When enabled, any
matching parenthesis is highlighted in `show-paren-style' after
@@ -25300,9 +25500,11 @@ or call the function `pixel-scroll-mode'.")
(autoload 'pixel-scroll-mode "pixel-scroll" "\
A minor mode to scroll text pixel-by-pixel.
-With a prefix argument ARG, enable Pixel Scroll mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable Pixel Scroll mode
-if ARG is omitted or nil.
+
+If called interactively, enable Pixel-Scroll mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
@@ -26959,9 +27161,11 @@ or call the function `rcirc-track-minor-mode'.")
(autoload 'rcirc-track-minor-mode "rcirc" "\
Global minor mode for tracking activity in rcirc buffers.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Rcirc-Track minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
@@ -27005,9 +27209,11 @@ or call the function `recentf-mode'.")
(autoload 'recentf-mode "recentf" "\
Toggle \"Open Recent\" menu (Recentf mode).
-With a prefix argument ARG, enable Recentf mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-Recentf mode if ARG is omitted or nil.
+
+If called interactively, enable Recentf mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Recentf mode is enabled, a \"Open Recent\" submenu is
displayed in the \"File\" menu, containing a list of files that
@@ -27157,6 +27363,12 @@ with a prefix argument, prompt for START-AT and FORMAT.
(autoload 'rectangle-mark-mode "rect" "\
Toggle the region as rectangular.
+
+If called interactively, enable Rectangle-Mark mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
Activates the region if needed. Only lasts until the region is deactivated.
\(fn &optional ARG)" t nil)
@@ -27184,9 +27396,11 @@ Activates the region if needed. Only lasts until the
region is deactivated.
(autoload 'refill-mode "refill" "\
Toggle automatic refilling (Refill mode).
-With a prefix argument ARG, enable Refill mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Refill mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Refill mode is a buffer-local minor mode. When enabled, the
current paragraph is refilled as you edit. Self-inserting
@@ -27216,6 +27430,11 @@ Turn on RefTeX mode.
(autoload 'reftex-mode "reftex" "\
Minor mode with distinct support for \\label, \\ref and \\cite in LaTeX.
+If called interactively, enable Reftex mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\\<reftex-mode-map>A Table of Contents of the entire (multifile) document with
browsing
capabilities is available with `\\[reftex-toc]'.
@@ -27560,9 +27779,11 @@ first comment line visible (if point is in a comment).
(autoload 'reveal-mode "reveal" "\
Toggle uncloaking of invisible text near point (Reveal mode).
-With a prefix argument ARG, enable Reveal mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-Reveal mode if ARG is omitted or nil.
+
+If called interactively, enable Reveal mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Reveal mode is a buffer-local minor mode. When enabled, it
reveals invisible text around point.
@@ -27583,9 +27804,10 @@ or call the function `global-reveal-mode'.")
Toggle Reveal mode in all buffers (Global Reveal mode).
Reveal mode renders invisible text around point visible again.
-With a prefix argument ARG, enable Global Reveal mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+If called interactively, enable Global Reveal mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
@@ -28303,9 +28525,11 @@ highlighting.
(autoload 'rst-minor-mode "rst" "\
Toggle ReST minor mode.
-With a prefix argument ARG, enable ReST minor mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Rst minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When ReST minor mode is enabled, the ReST mode keybindings
are installed on top of the major mode bindings. Use this
@@ -28352,9 +28576,11 @@ Use the command `ruler-mode' to change this variable.")
(autoload 'ruler-mode "ruler-mode" "\
Toggle display of ruler in header line (Ruler mode).
-With a prefix argument ARG, enable Ruler mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Ruler mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
@@ -28738,9 +28964,11 @@ or call the function `savehist-mode'.")
(autoload 'savehist-mode "savehist" "\
Toggle saving of minibuffer history (Savehist mode).
-With a prefix argument ARG, enable Savehist mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Savehist mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Savehist mode is enabled, minibuffer history is saved
periodically and when exiting Emacs. When Savehist mode is
@@ -28775,6 +29003,11 @@ Non-nil means automatically save place in each file.
This means when you visit a file, point goes to the last place
where it was when you previously visited the same file.
+If called interactively, enable Save-Place mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(autoload 'save-place-local-mode "saveplace" "\
@@ -28783,8 +29016,10 @@ If this mode is enabled, point is recorded when you
kill the buffer
or exit Emacs. Visiting this file again will go to that position,
even in a later Emacs session.
-If called with a prefix arg, the mode is enabled if and only if
-the argument is positive.
+If called interactively, enable Save-Place-Local mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
To save places automatically in all files, put this in your init
file:
@@ -28875,9 +29110,11 @@ or call the function `scroll-all-mode'.")
(autoload 'scroll-all-mode "scroll-all" "\
Toggle shared scrolling in same-frame windows (Scroll-All mode).
-With a prefix argument ARG, enable Scroll-All mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Scroll-All mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Scroll-All mode is enabled, scrolling commands invoked in
one window apply to all visible windows in the same frame.
@@ -28900,12 +29137,16 @@ one window apply to all visible windows in the same
frame.
(autoload 'scroll-lock-mode "scroll-lock" "\
Buffer-local minor mode for pager-like scrolling.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil. When enabled, keys that normally move
-point by line or paragraph will scroll the buffer by the
-respective amount of lines instead and point will be kept
-vertically fixed relative to window boundaries during scrolling.
+
+If called interactively, enable Scroll-Lock mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
+When enabled, keys that normally move point by line or paragraph
+will scroll the buffer by the respective amount of lines instead
+and point will be kept vertically fixed relative to window
+boundaries during scrolling.
\(fn &optional ARG)" t nil)
@@ -28964,9 +29205,11 @@ or call the function `semantic-mode'.")
(autoload 'semantic-mode "semantic" "\
Toggle parser features (Semantic mode).
-With a prefix argument ARG, enable Semantic mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-Semantic mode if ARG is omitted or nil.
+
+If called interactively, enable Semantic mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
In Semantic mode, Emacs parses the buffers you visit for their
semantic content. This information is used by a variety of
@@ -29925,9 +30168,11 @@ or call the function `server-mode'.")
(autoload 'server-mode "server" "\
Toggle Server mode.
-With a prefix argument ARG, enable Server mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-Server mode if ARG is omitted or nil.
+
+If called interactively, enable Server mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Server mode runs a process that accepts commands from the
`emacsclient' program. See Info node `Emacs server' and
@@ -30550,9 +30795,12 @@ buffer names.
(autoload 'smerge-mode "smerge-mode" "\
Minor mode to simplify editing output from the diff3 program.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Smerge mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\\{smerge-mode-map}
\(fn &optional ARG)" t nil)
@@ -31865,9 +32113,11 @@ or call the function `strokes-mode'.")
(autoload 'strokes-mode "strokes" "\
Toggle Strokes mode, a global minor mode.
-With a prefix argument ARG, enable Strokes mode if ARG is
-positive, and disable it otherwise. If called from Lisp,
-enable the mode if ARG is omitted or nil.
+
+If called interactively, enable Strokes mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\\<strokes-mode-map>
Strokes are pictographic mouse gestures which invoke commands.
@@ -31934,9 +32184,11 @@ Studlify-case the current buffer.
(autoload 'subword-mode "subword" "\
Toggle subword movement and editing (Subword mode).
-With a prefix argument ARG, enable Subword mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Subword mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Subword mode is a buffer-local minor mode. Enabling it changes
the definition of a word so that word-based commands stop inside
@@ -31956,8 +32208,6 @@ called a `subword'. Here are some examples:
This mode changes the definition of a word so that word commands
treat nomenclature boundaries as word boundaries.
-\\{subword-mode-map}
-
\(fn &optional ARG)" t nil)
(defvar global-subword-mode nil "\
@@ -31984,9 +32234,11 @@ See `subword-mode' for more information on Subword
mode.
(autoload 'superword-mode "subword" "\
Toggle superword movement and editing (Superword mode).
-With a prefix argument ARG, enable Superword mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Superword mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Superword mode is a buffer-local minor mode. Enabling it changes
the definition of words such that symbols characters are treated
@@ -32081,9 +32333,11 @@ or call the function `gpm-mouse-mode'.")
(autoload 'gpm-mouse-mode "t-mouse" "\
Toggle mouse support in GNU/Linux consoles (GPM Mouse mode).
-With a prefix argument ARG, enable GPM Mouse mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Gpm-Mouse mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
This allows the use of the mouse when operating on a GNU/Linux console,
in the same way as you can use the mouse under X11.
@@ -32481,6 +32735,11 @@ location is indicated by
`table-word-continuation-char'. This
variable's value can be toggled by \\[table-fixed-width-mode] at
run-time.
+If called interactively, enable Table-Fixed-Width mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
\(fn &optional ARG)" t nil)
(autoload 'table-query-dimension "table" "\
@@ -33667,6 +33926,11 @@ This function is meant to be used as a
`post-self-insert-hook'.
(autoload 'tildify-mode "tildify" "\
Adds electric behavior to space character.
+If called interactively, enable Tildify mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
When space is inserted into a buffer in a position where hard space is required
instead (determined by `tildify-space-pattern' and `tildify-space-predicates'),
that space character is replaced by a hard space specified by
@@ -33712,9 +33976,11 @@ or call the function `display-time-mode'.")
(autoload 'display-time-mode "time" "\
Toggle display of time, load level, and mail flag in mode lines.
-With a prefix argument ARG, enable Display Time mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-it if ARG is omitted or nil.
+
+If called interactively, enable Display-Time mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When Display Time mode is enabled, it updates every minute (you
can control the number of seconds between updates by customizing
@@ -34571,6 +34837,11 @@ or call the function `type-break-mode'.")
Enable or disable typing-break mode.
This is a minor mode, but it is global to all buffers by default.
+If called interactively, enable Type-Break mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
+
When this mode is enabled, the user is encouraged to take typing breaks at
appropriate intervals; either after a specified amount of time or when the
user has exceeded a keystroke threshold. When the time arrives, the user
@@ -34579,9 +34850,6 @@ again in a short period of time. The idea is to give
the user enough time
to find a good breaking point in his or her work, but be sufficiently
annoying to discourage putting typing breaks off indefinitely.
-A negative prefix argument disables this mode.
-No argument or any non-negative argument enables it.
-
The user may enable or disable this mode by setting the variable of the
same name, though setting it in that way doesn't reschedule a break or
reset the keystroke counter.
@@ -35105,9 +35373,11 @@ or call the function `url-handler-mode'.")
(autoload 'url-handler-mode "url-handlers" "\
Toggle using `url' library for URL filenames (URL Handler mode).
-With a prefix argument ARG, enable URL Handler mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Url-Handler mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
@@ -37252,9 +37522,11 @@ own View-like bindings.
(autoload 'view-mode "view" "\
Toggle View mode, a minor mode for viewing text but not editing it.
-With a prefix argument ARG, enable View mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable View mode
-if ARG is omitted or nil.
+
+If called interactively, enable View mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
When View mode is enabled, commands that do not change the buffer
contents are available as usual. Kill commands insert text in
@@ -37628,9 +37900,11 @@ or call the function `which-function-mode'.")
(autoload 'which-function-mode "which-func" "\
Toggle mode line display of current function (Which Function mode).
-With a prefix argument ARG, enable Which Function mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Which-Function mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Which Function mode is a global minor mode. When enabled, the
current function name is continuously displayed in the mode line,
@@ -37648,11 +37922,11 @@ in certain major modes.
(autoload 'whitespace-mode "whitespace" "\
Toggle whitespace visualization (Whitespace mode).
-With a prefix argument ARG, enable Whitespace mode if ARG is
-positive, and disable it otherwise.
-If called from Lisp, also enables the mode if ARG is omitted or nil,
-and toggles it if ARG is `toggle'.
+If called interactively, enable Whitespace mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
See also `whitespace-style', `whitespace-newline' and
`whitespace-display-mappings'.
@@ -37661,11 +37935,11 @@ See also `whitespace-style', `whitespace-newline' and
(autoload 'whitespace-newline-mode "whitespace" "\
Toggle newline visualization (Whitespace Newline mode).
-With a prefix argument ARG, enable Whitespace Newline mode if ARG
-is positive, and disable it otherwise.
-If called from Lisp, also enables the mode if ARG is omitted or nil,
-and toggles it if ARG is `toggle'.
+If called interactively, enable Whitespace-Newline mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Use `whitespace-newline-mode' only for NEWLINE visualization
exclusively. For other visualizations, including NEWLINE
@@ -37688,11 +37962,11 @@ or call the function `global-whitespace-mode'.")
(autoload 'global-whitespace-mode "whitespace" "\
Toggle whitespace visualization globally (Global Whitespace mode).
-With a prefix argument ARG, enable Global Whitespace mode if ARG
-is positive, and disable it otherwise.
-If called from Lisp, also enables the mode if ARG is omitted or nil,
-and toggles it if ARG is `toggle'.
+If called interactively, enable Global Whitespace mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
See also `whitespace-style', `whitespace-newline' and
`whitespace-display-mappings'.
@@ -37711,11 +37985,11 @@ or call the function
`global-whitespace-newline-mode'.")
(autoload 'global-whitespace-newline-mode "whitespace" "\
Toggle global newline visualization (Global Whitespace Newline mode).
-With a prefix argument ARG, enable Global Whitespace Newline mode
-if ARG is positive, and disable it otherwise.
-If called from Lisp, also enables the mode if ARG is omitted or nil,
-and toggles it if ARG is `toggle'.
+If called interactively, enable Global Whitespace-Newline mode if ARG is
positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Use `global-whitespace-newline-mode' only for NEWLINE
visualization exclusively. For other visualizations, including
@@ -38037,9 +38311,11 @@ Show widget browser for WIDGET in other window.
(autoload 'widget-minor-mode "wid-browse" "\
Minor mode for traversing widgets.
-With a prefix argument ARG, enable the mode if ARG is positive,
-and disable it otherwise. If called from Lisp, enable the mode
-if ARG is omitted or nil.
+
+If called interactively, enable Widget minor mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
\(fn &optional ARG)" t nil)
@@ -38161,9 +38437,11 @@ or call the function `winner-mode'.")
(autoload 'winner-mode "winner" "\
Toggle Winner mode on or off.
-With a prefix argument ARG, enable Winner mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil, and toggle it if ARG is `toggle'.
+
+If called interactively, enable Winner mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Winner mode is a global minor mode that records the changes in
the window configuration (i.e. how the frames are partitioned
@@ -38373,6 +38651,12 @@ With prefix argument, prompt for the identifier.
\(fn IDENTIFIER)" t nil)
+(autoload 'xref-find-definitions-at-mouse "xref" "\
+Find the definition of identifier at or around mouse click.
+This command is intended to be bound to a mouse event.
+
+\(fn EVENT)" t nil)
+
(autoload 'xref-find-apropos "xref" "\
Find all meaningful symbols that match PATTERN.
The argument has the same meaning as in `apropos'.
@@ -38425,9 +38709,11 @@ or call the function `xterm-mouse-mode'.")
(autoload 'xterm-mouse-mode "xt-mouse" "\
Toggle XTerm mouse mode.
-With a prefix argument ARG, enable XTerm mouse mode if ARG is
-positive, and disable it otherwise. If called from Lisp, enable
-the mode if ARG is omitted or nil.
+
+If called interactively, enable Xterm-Mouse mode if ARG is positive, and
+disable it if ARG is zero or negative. If called from Lisp,
+also enable the mode if ARG is omitted or nil, and toggle it
+if ARG is `toggle'; disable the mode otherwise.
Turn it on to use Emacs mouse commands, and off to use xterm mouse commands.
This works in terminal emulators compatible with xterm. It only
diff --git a/lisp/mh-e/mh-comp.el b/lisp/mh-e/mh-comp.el
index 2b49fae..5c474b4 100644
--- a/lisp/mh-e/mh-comp.el
+++ b/lisp/mh-e/mh-comp.el
@@ -77,6 +77,14 @@ Default is \"components\".
If not an absolute file name, the file is searched for first in the
user's MH directory, then in the system MH lib directory.")
+(defvar mh-dist-formfile "distcomps"
+ "Name of file to be used as a skeleton for redistributing messages.
+
+Default is \"distcomps\".
+
+If not an absolute file name, the file is searched for first in the
+user's MH directory, then in the system MH lib directory.")
+
(defvar mh-repl-formfile "replcomps"
"Name of file to be used as a skeleton for replying to messages.
@@ -413,7 +421,7 @@ See also `mh-send'."
(interactive (list (mh-get-msg-num t)))
(let* ((from-folder mh-current-folder)
(config (current-window-configuration))
- (components-file (mh-bare-components))
+ (components-file (mh-bare-components mh-comp-formfile))
(draft
(cond ((and mh-draft-folder (equal from-folder mh-draft-folder))
(pop-to-buffer (find-file-noselect (mh-msg-filename message))
@@ -649,15 +657,16 @@ Original message has headers FROM and SUBJECT."
(format mh-forward-subject-format from subject))
;;;###mh-autoload
-(defun mh-redistribute (to cc &optional message)
+(defun mh-redistribute (to cc identity &optional message)
"Redistribute a message.
This command is similar in function to forwarding mail, but it
does not allow you to edit the message, nor does it add your name
to the \"From\" header field. It appears to the recipient as if
the message had come from the original sender. When you run this
-command, you are prompted for the TO and CC recipients. The
-default MESSAGE is the current message.
+command, you are prompted for the TO and CC recipients. You are
+also prompted for the sending IDENTITY to use. The default
+MESSAGE is the current message.
Also investigate the command \\[mh-edit-again] for another way to
redistribute messages.
@@ -668,6 +677,9 @@ The hook `mh-annotate-msg-hook' is run after annotating the
message and scan line."
(interactive (list (mh-read-address "Redist-To: ")
(mh-read-address "Redist-Cc: ")
+ (if mh-identity-list
+ (mh-select-identity mh-identity-default)
+ nil)
(mh-get-msg-num t)))
(or message
(setq message (mh-get-msg-num t)))
@@ -677,14 +689,51 @@ message and scan line."
(if mh-redist-full-contents-flag
(mh-msg-filename message)
nil)
- nil)))
- (mh-goto-header-end 0)
- (insert "Resent-To: " to "\n")
- (if (not (equal cc "")) (insert "Resent-cc: " cc "\n"))
- (mh-clean-msg-header
- (point-min)
- "^Message-Id:\\|^Received:\\|^Return-Path:\\|^Sender:\\|^Date:\\|^From:"
- nil)
+ nil))
+ (from (mh-identity-field identity "From"))
+ (fcc (mh-identity-field identity "Fcc"))
+ (bcc (mh-identity-field identity "Bcc"))
+ comp-fcc comp-to comp-cc comp-bcc)
+ (if mh-redist-full-contents-flag
+ (mh-clean-msg-header
+ (point-min)
+ "^Message-Id:\\|^Received:\\|^Return-Path:\\|^Date:\\|^Resent-.*:"
+ nil))
+ ;; Read fields from the distcomps file and put them in our
+ ;; draft. For "To", "Cc", "Bcc", and "Fcc", multiple headers are
+ ;; combined into a single header with comma-separated entries.
+ ;; For "From", the first value wins, with the identity's "From"
+ ;; trumping anything in the distcomps file.
+ (let ((components-file (mh-bare-components mh-dist-formfile)))
+ (mh-mapc
+ (function
+ (lambda (header-field)
+ (let ((field (car header-field))
+ (value (cdr header-field))
+ (case-fold-search t))
+ (cond
+ ((string-match field "^Resent-Fcc$")
+ (setq comp-fcc value))
+ ((string-match field "^Resent-From$")
+ (or from
+ (setq from value)))
+ ((string-match field "^Resent-To$")
+ (setq comp-to value))
+ ((string-match field "^Resent-Cc$")
+ (setq comp-cc value))
+ ((string-match field "^Resent-Bcc$")
+ (setq comp-bcc value))
+ ((string-match field "^Resent-.*$")
+ (mh-insert-fields field value))))))
+ (mh-components-to-list components-file))
+ (delete-file components-file))
+ (mh-insert-fields "Resent-To:" (mapconcat 'identity (list to comp-to) ",
")
+ "Resent-Cc:" (mapconcat 'identity (list cc comp-cc) ",
")
+ "Resent-Fcc:" (mapconcat 'identity (list fcc
+ comp-fcc) ",
")
+ "Resent-Bcc:" (mapconcat 'identity (list bcc
+ comp-bcc) ",
")
+ "Resent-From:" from)
(save-buffer)
(message "Redistributing...")
(let ((env "mhdist=1"))
@@ -702,7 +751,8 @@ message and scan line."
;; Annotate...
(mh-annotate-msg message folder mh-note-dist
"-component" "Resent:"
- "-text" (format "\"%s %s\"" to cc)))
+ "-text" (format "\"To: %s Cc: %s From: %s\""
+ to cc from)))
(kill-buffer draft)
(message "Redistributing...done"))))
@@ -898,7 +948,7 @@ CONFIG is the window configuration before sending mail."
(message "Composing a message...")
(let ((draft (mh-read-draft
"message"
- (mh-bare-components)
+ (mh-bare-components mh-comp-formfile)
t)))
(mh-insert-fields "To:" to "Subject:" subject "Cc:" cc)
(goto-char (point-max))
@@ -908,23 +958,25 @@ CONFIG is the window configuration before sending mail."
(mh-letter-mode-message)
(mh-letter-adjust-point))))
-(defun mh-bare-components ()
- "Generate a temporary, clean components file and return its path."
- ;; Let comp(1) create the skeleton for us. This is particularly
+(defun mh-bare-components (formfile)
+ "Generate a temporary, clean components file from FORMFILE.
+Return the path to the temporary file."
+ ;; Let comp(1) create the skeleton for us. This is particularly
;; important with nmh-1.5, because its default "components" needs
- ;; some processing before it can be used. Unfortunately, comp(1)
- ;; doesn't have a -build option. So, to avoid the possibility of
- ;; clobbering an existing draft, create a temporary directory and
- ;; use it as the drafts folder. Then copy the skeleton to a regular
- ;; temp file, and return the regular temp file.
+ ;; some processing before it can be used. Unfortunately, comp(1)
+ ;; didn't have a -build option until later versions of nmh. So, to
+ ;; avoid the possibility of clobbering an existing draft, create
+ ;; a temporary directory and use it as the drafts folder. Then
+ ;; copy the skeleton to a regular temp file, and return the
+ ;; regular temp file.
(let (new
(temp-folder (make-temp-file
(concat mh-user-path "draftfolder.") t)))
(mh-exec-cmd "comp" "-nowhatnowproc"
"-draftfolder" (format "+%s"
(file-name-nondirectory temp-folder))
- (if (stringp mh-comp-formfile)
- (list "-form" mh-comp-formfile)))
+ (if (stringp formfile)
+ (list "-form" formfile)))
(setq new (make-temp-file "comp."))
(rename-file (concat temp-folder "/" "1") new t)
;; The temp folder could contain various metadata files. Rather
diff --git a/lisp/mh-e/mh-funcs.el b/lisp/mh-e/mh-funcs.el
index 661d0ec..3574f8c 100644
--- a/lisp/mh-e/mh-funcs.el
+++ b/lisp/mh-e/mh-funcs.el
@@ -357,6 +357,8 @@ Arguments are IGNORED (for `revert-buffer')."
(yes-or-no-p "Undo all commands in folder? "))
(setq mh-delete-list nil
mh-refile-list nil
+ mh-blacklist nil
+ mh-whitelist nil
mh-seq-list nil
mh-next-direction 'forward)
(with-mh-folder-updating (nil)
diff --git a/lisp/mh-e/mh-identity.el b/lisp/mh-e/mh-identity.el
index fd7c2b8..a1eb22f 100644
--- a/lisp/mh-e/mh-identity.el
+++ b/lisp/mh-e/mh-identity.el
@@ -132,6 +132,33 @@ valid header field."
'mh-identity-handler-default))
;;;###mh-autoload
+(defun mh-select-identity (default)
+ "Prompt for and return an identity.
+If DEFAULT is non-nil, it will be used if the user doesn't enter a
+different identity.
+
+See `mh-identity-list'."
+ (let (identity)
+ (setq identity
+ (completing-read
+ "Identity: "
+ (cons '("None")
+ (mapcar 'list (mapcar 'car mh-identity-list)))
+ nil t default nil default))
+ (if (eq identity "None")
+ nil
+ identity)))
+
+;;;###mh-autoload
+(defun mh-identity-field (identity field)
+ "Return the specified FIELD of the given IDENTITY.
+
+See `mh-identity-list'."
+ (let* ((pers-list (cadr (assoc identity mh-identity-list)))
+ (value (cdr (assoc field pers-list))))
+ value))
+
+;;;###mh-autoload
(defun mh-insert-identity (identity &optional maybe-insert)
"Insert fields specified by given IDENTITY.
diff --git a/lisp/net/rcirc.el b/lisp/net/rcirc.el
index 5b63e0c..108e368 100644
--- a/lisp/net/rcirc.el
+++ b/lisp/net/rcirc.el
@@ -583,7 +583,7 @@ If ARG is non-nil, instead prompt for connection
parameters."
(setq-local rcirc-connection-info
(list server port nick user-name full-name startup-channels
- password encryption))
+ password encryption server-alias))
(setq-local rcirc-process process)
(setq-local rcirc-server server)
(setq-local rcirc-server-name
diff --git a/lisp/net/rlogin.el b/lisp/net/rlogin.el
index 3bfc4d7..015e04f 100644
--- a/lisp/net/rlogin.el
+++ b/lisp/net/rlogin.el
@@ -1,4 +1,4 @@
-;;; rlogin.el --- remote login interface
+;;; rlogin.el --- remote login interface -*- lexical-binding:t -*-
;; Copyright (C) 1992-1995, 1997-1998, 2001-2018 Free Software
;; Foundation, Inc.
@@ -30,9 +30,9 @@
;; tracking and the sending of some special characters.
;; If you wish for rlogin mode to prompt you in the minibuffer for
-;; passwords when a password prompt appears, just enter m-x send-invisible
-;; and type in your line, or add `comint-watch-for-password-prompt' to
-;; `comint-output-filter-functions'.
+;; passwords when a password prompt appears, just enter
+;; M-x comint-send-invisible and type in your line (or tweak
+;; `comint-password-prompt-regexp' to match your password prompt).
;;; Code:
diff --git a/lisp/net/soap-client.el b/lisp/net/soap-client.el
index 17f8308..f5de05d 100644
--- a/lisp/net/soap-client.el
+++ b/lisp/net/soap-client.el
@@ -685,14 +685,17 @@ This is a specialization of `soap-decode-type' for
(anyType (soap-decode-any-type node))
(Array (soap-decode-array node))))))
-(defun soap-type-of (element)
- "Return the type of ELEMENT."
- ;; Support Emacs < 26 byte-code running in Emacs >= 26 sessions
- ;; (Bug#31742).
- (let ((type (type-of element)))
- (if (eq type 'vector)
- (aref element 0) ; For Emacs 25 and earlier.
- type)))
+(defalias 'soap-type-of
+ (if (eq 'soap-xs-basic-type (type-of (make-soap-xs-basic-type)))
+ ;; `type-of' in Emacs ≥ 26 already does what we need.
+ #'type-of
+ ;; For Emacs < 26, use our own function.
+ (lambda (element)
+ "Return the type of ELEMENT."
+ (if (vectorp element)
+ (aref element 0) ;Assume this vector is actually a struct!
+ ;; This should never happen.
+ (type-of element)))))
;; Register methods for `soap-xs-basic-type'
(let ((tag (soap-type-of (make-soap-xs-basic-type))))
@@ -2881,6 +2884,8 @@ reference multiRef parts which are external to
RESPONSE-NODE."
;;;; SOAP type encoding
+;; FIXME: Use `cl-defmethod' (but this requires Emacs-25).
+
(defun soap-encode-attributes (value type)
"Encode XML attributes for VALUE according to TYPE.
This is a generic function which determines the attribute encoder
diff --git a/lisp/net/tramp-gvfs.el b/lisp/net/tramp-gvfs.el
index 1f40339..84af410 100644
--- a/lisp/net/tramp-gvfs.el
+++ b/lisp/net/tramp-gvfs.el
@@ -49,9 +49,9 @@
;; The user option `tramp-gvfs-methods' contains the list of supported
;; connection methods. Per default, these are "afp", "dav", "davs",
-;; "gdrive", "owncloud" and "sftp".
+;; "gdrive", "nextcloud" and "sftp".
-;; "gdrive" and "owncloud" connection methods require a respective
+;; "gdrive" and "nextcloud" connection methods require a respective
;; account in GNOME Online Accounts, with enabled "Files" service.
;; Other possible connection methods are "ftp", "http", "https" and
@@ -121,7 +121,7 @@
;;;###tramp-autoload
(defcustom tramp-gvfs-methods
- '("afp" "dav" "davs" "gdrive" "owncloud" "sftp")
+ '("afp" "dav" "davs" "gdrive" "nextcloud" "sftp")
"List of methods for remote files, accessed with GVFS."
:group 'tramp
:version "27.1"
@@ -132,11 +132,11 @@
(const "gdrive")
(const "http")
(const "https")
- (const "owncloud")
+ (const "nextcloud")
(const "sftp")
(const "smb"))))
-(defconst tramp-goa-methods '("gdrive" "owncloud")
+(defconst tramp-goa-methods '("gdrive" "nextcloud")
"List of methods which require registration at GNOME Online Accounts.")
;; Remove GNOME Online Accounts methods if not supported.
@@ -511,11 +511,11 @@ It has been changed in GVFS 1.14.")
":[[:blank:]]+\\(.*\\)$")
"Regexp to parse GVFS file system attributes with `gvfs-info'.")
-(defconst tramp-gvfs-owncloud-default-prefix "/remote.php/webdav"
+(defconst tramp-gvfs-nextcloud-default-prefix "/remote.php/webdav"
"Default prefix for owncloud / nextcloud methods.")
-(defconst tramp-gvfs-owncloud-default-prefix-regexp
- (concat (regexp-quote tramp-gvfs-owncloud-default-prefix) "$")
+(defconst tramp-gvfs-nextcloud-default-prefix-regexp
+ (concat (regexp-quote tramp-gvfs-nextcloud-default-prefix) "$")
"Regexp of default prefix for owncloud / nextcloud methods.")
@@ -1380,7 +1380,7 @@ file-notify events."
(with-parsed-tramp-file-name filename nil
(when (string-equal "gdrive" method)
(setq method "google-drive"))
- (when (string-equal "owncloud" method)
+ (when (string-equal "nextcloud" method)
(setq method "davs"
localname
(concat (tramp-gvfs-get-remote-prefix v) localname)))
@@ -1543,8 +1543,8 @@ file-notify events."
(setq method "davs"))
(when (and (string-equal "davs" method)
(string-match
- tramp-gvfs-owncloud-default-prefix-regexp prefix))
- (setq method "owncloud"))
+ tramp-gvfs-nextcloud-default-prefix-regexp prefix))
+ (setq method "nextcloud"))
(when (string-equal "google-drive" method)
(setq method "gdrive"))
(when (and (string-equal "http" method) (stringp uri))
@@ -1633,8 +1633,8 @@ file-notify events."
(setq method "davs"))
(when (and (string-equal "davs" method)
(string-match
- tramp-gvfs-owncloud-default-prefix-regexp prefix))
- (setq method "owncloud"))
+ tramp-gvfs-nextcloud-default-prefix-regexp prefix))
+ (setq method "nextcloud"))
(when (string-equal "google-drive" method)
(setq method "gdrive"))
(when (and (string-equal "http" method) (stringp uri))
@@ -1688,7 +1688,7 @@ It was \"a(say)\", but has changed to \"a{sv})\"."
(localname (tramp-file-name-unquote-localname vec))
(share (when (string-match "^/?\\([^/]+\\)" localname)
(match-string 1 localname)))
- (ssl (if (string-match "^davs\\|^owncloud" method) "true" "false"))
+ (ssl (if (string-match "^davs\\|^nextcloud" method) "true" "false"))
(mount-spec
`(:array
,@(cond
@@ -1696,7 +1696,7 @@ It was \"a(say)\", but has changed to \"a{sv})\"."
(list (tramp-gvfs-mount-spec-entry "type" "smb-share")
(tramp-gvfs-mount-spec-entry "server" host)
(tramp-gvfs-mount-spec-entry "share" share)))
- ((string-match "^dav\\|^owncloud" method)
+ ((string-match "^dav\\|^nextcloud" method)
(list (tramp-gvfs-mount-spec-entry "type" "dav")
(tramp-gvfs-mount-spec-entry "host" host)
(tramp-gvfs-mount-spec-entry "ssl" ssl)))
@@ -1707,6 +1707,9 @@ It was \"a(say)\", but has changed to \"a{sv})\"."
((string-equal "gdrive" method)
(list (tramp-gvfs-mount-spec-entry "type" "google-drive")
(tramp-gvfs-mount-spec-entry "host" host)))
+ ((string-equal "nextcloud" method)
+ (list (tramp-gvfs-mount-spec-entry "type" "owncloud")
+ (tramp-gvfs-mount-spec-entry "host" host)))
((string-match "^http" method)
(list (tramp-gvfs-mount-spec-entry "type" "http")
(tramp-gvfs-mount-spec-entry
@@ -1980,6 +1983,8 @@ VEC is used only for traces."
:port (match-string 3 identity)))
(when (string-equal (tramp-goa-name-method key) "google")
(setf (tramp-goa-name-method key) "gdrive"))
+ (when (string-equal (tramp-goa-name-method key) "owncloud")
+ (setf (tramp-goa-name-method key) "nextcloud"))
;; Cache all properties.
(dolist (prop (nconc account-properties files-properties))
(tramp-set-connection-property key (car prop) (cdr prop)))
@@ -2086,7 +2091,7 @@ This uses \"avahi-browse\" in case D-Bus is not enabled
in Avahi."
;; * (Customizable) unmount when exiting Emacs. See tramp-archive.el.
;; * Host name completion for existing mount points (afp-server,
-;; smb-server, google-drive, owncloud) or via smb-network or network.
+;; smb-server, google-drive, nextcloud) or via smb-network or network.
;;
;; * Check, how two shares of the same SMB server can be mounted in
;; parallel.
diff --git a/lisp/net/tramp-sh.el b/lisp/net/tramp-sh.el
index 2d25350..86e82d4 100644
--- a/lisp/net/tramp-sh.el
+++ b/lisp/net/tramp-sh.el
@@ -2547,7 +2547,11 @@ The method used must be an out-of-band method."
"Like `make-directory' for Tramp files."
(setq dir (expand-file-name dir))
(with-parsed-tramp-file-name dir nil
- (tramp-flush-directory-properties v (file-name-directory localname))
+ ;; When PARENTS is non-nil, DIR could be a chain of non-existent
+ ;; directories a/b/c/... Instead of checking, we simply flush the
+ ;; whole cache.
+ (tramp-flush-directory-properties
+ v (if parents "/" (file-name-directory localname)))
(save-excursion
(tramp-barf-unless-okay
v (format "%s %s"
diff --git a/lisp/net/tramp.el b/lisp/net/tramp.el
index d56b09a..1af2def 100644
--- a/lisp/net/tramp.el
+++ b/lisp/net/tramp.el
@@ -3567,16 +3567,20 @@ support symbolic links."
;; First, we must replace environment variables.
(setq filename (tramp-replace-environment-variables filename))
(with-parsed-tramp-file-name filename nil
- ;; We do not want to replace environment variables, again.
+ ;; We do not want to replace environment variables, again. "//"
+ ;; has a special meaning at the beginning of a file name on
+ ;; Cygwin and MS-Windows, we must remove it.
(let (process-environment)
;; Ignore in LOCALNAME everything before "//" or "/~".
(when (stringp localname)
(if (string-match "//\\(/\\|~\\)" localname)
- (setq filename (substitute-in-file-name localname))
+ (setq filename
+ (replace-regexp-in-string
+ "\\`/+" "/" (substitute-in-file-name localname)))
(setq filename
(concat (file-remote-p filename)
- (tramp-run-real-handler
- 'substitute-in-file-name (list localname)))))))
+ (replace-regexp-in-string
+ "\\`/+" "/" (substitute-in-file-name
localname)))))))
;; "/m:h:~" does not work for completion. We use "/m:h:~/".
(if (and (stringp localname) (string-equal "~" localname))
(concat filename "/")
diff --git a/lisp/net/trampver.el b/lisp/net/trampver.el
index 0b83afc..9bc8768 100644
--- a/lisp/net/trampver.el
+++ b/lisp/net/trampver.el
@@ -69,7 +69,7 @@
("2.2.9-24.4" . "24.4") ("2.2.11-24.5" . "24.5")
("2.2.13.25.1" . "25.1") ("2.2.13.25.2" . "25.2")
("2.2.13.25.2" . "25.3")
- ("2.3.3.26.1" . "26.1")))
+ ("2.3.3.26.1" . "26.1") ("2.3.4.26.2" . "26.2")))
(add-hook 'tramp-unload-hook
(lambda ()
diff --git a/lisp/profiler.el b/lisp/profiler.el
index eaeb697..41dea68 100644
--- a/lisp/profiler.el
+++ b/lisp/profiler.el
@@ -105,13 +105,13 @@
"Format ENTRY in human readable string. ENTRY would be a
function name of a function itself."
(cond ((memq (car-safe entry) '(closure lambda))
- (format "#<lambda 0x%x>" (sxhash entry)))
+ (format "#<lambda %#x>" (sxhash entry)))
((byte-code-function-p entry)
- (format "#<compiled 0x%x>" (sxhash entry)))
+ (format "#<compiled %#x>" (sxhash entry)))
((or (subrp entry) (symbolp entry) (stringp entry))
(format "%s" entry))
(t
- (format "#<unknown 0x%x>" (sxhash entry)))))
+ (format "#<unknown %#x>" (sxhash entry)))))
(defun profiler-fixup-entry (entry)
(if (symbolp entry)
diff --git a/lisp/progmodes/cc-engine.el b/lisp/progmodes/cc-engine.el
index 3961ea6..d1eb3c3 100644
--- a/lisp/progmodes/cc-engine.el
+++ b/lisp/progmodes/cc-engine.el
@@ -12607,7 +12607,11 @@ comment at the start of cc-engine.el for more info."
(= (point) containing-sexp)))
(if (eq (point) (c-point 'boi))
(c-add-syntax 'brace-list-close (point))
- (setq lim (c-most-enclosing-brace state-cache (point)))
+ (setq lim (or (save-excursion
+ (and
+ (c-back-over-member-initializers)
+ (point)))
+ (c-most-enclosing-brace state-cache (point))))
(c-beginning-of-statement-1 lim nil nil t)
(c-add-stmt-syntax 'brace-list-close nil t lim paren-state)))
@@ -12636,7 +12640,11 @@ comment at the start of cc-engine.el for more info."
(goto-char containing-sexp))
(if (eq (point) (c-point 'boi))
(c-add-syntax 'brace-list-intro (point))
- (setq lim (c-most-enclosing-brace state-cache (point)))
+ (setq lim (or (save-excursion
+ (and
+ (c-back-over-member-initializers)
+ (point)))
+ (c-most-enclosing-brace state-cache (point))))
(c-beginning-of-statement-1 lim)
(c-add-stmt-syntax 'brace-list-intro nil t lim paren-state)))
diff --git a/lisp/progmodes/elisp-mode.el b/lisp/progmodes/elisp-mode.el
index 58a58b4..f694252 100644
--- a/lisp/progmodes/elisp-mode.el
+++ b/lisp/progmodes/elisp-mode.el
@@ -45,7 +45,7 @@ It has `lisp-mode-abbrev-table' as its parent."
"Syntax table used in `emacs-lisp-mode'.")
(defvar emacs-lisp-mode-map
- (let ((map (make-sparse-keymap "Emacs-Lisp"))
+ (let ((map (make-sparse-keymap))
(menu-map (make-sparse-keymap "Emacs-Lisp"))
(lint-map (make-sparse-keymap))
(prof-map (make-sparse-keymap))
diff --git a/lisp/progmodes/grep.el b/lisp/progmodes/grep.el
index 519b768..0ededb1 100644
--- a/lisp/progmodes/grep.el
+++ b/lisp/progmodes/grep.el
@@ -374,7 +374,9 @@ Notice that using \\[next-error] or \\[compile-goto-error]
modifies
;; to handle weird file names (with colons in them) as
;; well as possible. E.g., use [1-9][0-9]* rather than
;; [0-9]+ so as to accept ":034:" in file names.
- "\\(?1:[^\n:]+?[^\n/:]\\):[\t ]*\\(?2:[1-9][0-9]*\\)[\t ]*:"
+ "\\(?1:"
+ "\\(?:[a-zA-Z]:\\)?" ; Allow "C:..." for w32.
+ "[^\n:]+?[^\n/:]\\):[\t ]*\\(?2:[1-9][0-9]*\\)[\t ]*:"
"\\)")
1 2
;; Calculate column positions (col . end-col) of first grep match on a
line
diff --git a/lisp/progmodes/hideif.el b/lisp/progmodes/hideif.el
index ce7127a..24ad2ff 100644
--- a/lisp/progmodes/hideif.el
+++ b/lisp/progmodes/hideif.el
@@ -1625,7 +1625,7 @@ not be expanded."
((integerp result)
(if (or (= 0 result) (= 1 result))
(message "%S <= `%s'" result exprstring)
- (message "%S (0x%x) <= `%s'" result result exprstring)))
+ (message "%S (%#x) <= `%s'" result result exprstring)))
((null result) (message "%S <= `%s'" 'false exprstring))
((eq t result) (message "%S <= `%s'" 'true exprstring))
(t (message "%S <= `%s'" result exprstring)))
diff --git a/lisp/progmodes/python.el b/lisp/progmodes/python.el
index e39ff08..c55b69e 100644
--- a/lisp/progmodes/python.el
+++ b/lisp/progmodes/python.el
@@ -526,9 +526,19 @@ The type returned can be `comment', `string' or `paren'."
font-lock-string-face)
font-lock-comment-face))
-(defvar python-font-lock-keywords
- ;; Keywords
- `(,(rx symbol-start
+(defvar python-font-lock-keywords-level-1
+ `((,(rx symbol-start "def" (1+ space) (group (1+ (or word ?_))))
+ (1 font-lock-function-name-face))
+ (,(rx symbol-start "class" (1+ space) (group (1+ (or word ?_))))
+ (1 font-lock-type-face)))
+ "Font lock keywords to use in python-mode for level 1 decoration.
+
+This is the minimum decoration level, including function and
+class declarations.")
+
+(defvar python-font-lock-keywords-level-2
+ `(,@python-font-lock-keywords-level-1
+ ,(rx symbol-start
(or
"and" "del" "from" "not" "while" "as" "elif" "global" "or" "with"
"assert" "else" "if" "pass" "yield" "break" "except" "import" "class"
@@ -548,12 +558,35 @@ The type returned can be `comment', `string' or `paren'."
;; Extra:
"self")
symbol-end)
- ;; functions
- (,(rx symbol-start "def" (1+ space) (group (1+ (or word ?_))))
- (1 font-lock-function-name-face))
- ;; classes
- (,(rx symbol-start "class" (1+ space) (group (1+ (or word ?_))))
- (1 font-lock-type-face))
+ ;; Builtins
+ (,(rx symbol-start
+ (or
+ "abs" "all" "any" "bin" "bool" "callable" "chr" "classmethod"
+ "compile" "complex" "delattr" "dict" "dir" "divmod" "enumerate"
+ "eval" "filter" "float" "format" "frozenset" "getattr" "globals"
+ "hasattr" "hash" "help" "hex" "id" "input" "int" "isinstance"
+ "issubclass" "iter" "len" "list" "locals" "map" "max" "memoryview"
+ "min" "next" "object" "oct" "open" "ord" "pow" "print" "property"
+ "range" "repr" "reversed" "round" "set" "setattr" "slice" "sorted"
+ "staticmethod" "str" "sum" "super" "tuple" "type" "vars" "zip"
+ "__import__"
+ ;; Python 2:
+ "basestring" "cmp" "execfile" "file" "long" "raw_input" "reduce"
+ "reload" "unichr" "unicode" "xrange" "apply" "buffer" "coerce"
+ "intern"
+ ;; Python 3:
+ "ascii" "bytearray" "bytes" "exec"
+ ;; Extra:
+ "__all__" "__doc__" "__name__" "__package__")
+ symbol-end) . font-lock-builtin-face))
+ "Font lock keywords to use in python-mode for level 2 decoration.
+
+This is the medium decoration level, including everything in
+`python-font-lock-keywords-level-1', as well as keywords and
+builtins.")
+
+(defvar python-font-lock-keywords-maximum-decoration
+ `(,@python-font-lock-keywords-level-2
;; Constants
(,(rx symbol-start
(or
@@ -596,27 +629,6 @@ The type returned can be `comment', `string' or `paren'."
"VMSError" "WindowsError"
)
symbol-end) . font-lock-type-face)
- ;; Builtins
- (,(rx symbol-start
- (or
- "abs" "all" "any" "bin" "bool" "callable" "chr" "classmethod"
- "compile" "complex" "delattr" "dict" "dir" "divmod" "enumerate"
- "eval" "filter" "float" "format" "frozenset" "getattr" "globals"
- "hasattr" "hash" "help" "hex" "id" "input" "int" "isinstance"
- "issubclass" "iter" "len" "list" "locals" "map" "max" "memoryview"
- "min" "next" "object" "oct" "open" "ord" "pow" "print" "property"
- "range" "repr" "reversed" "round" "set" "setattr" "slice" "sorted"
- "staticmethod" "str" "sum" "super" "tuple" "type" "vars" "zip"
- "__import__"
- ;; Python 2:
- "basestring" "cmp" "execfile" "file" "long" "raw_input" "reduce"
- "reload" "unichr" "unicode" "xrange" "apply" "buffer" "coerce"
- "intern"
- ;; Python 3:
- "ascii" "bytearray" "bytes" "exec"
- ;; Extra:
- "__all__" "__doc__" "__name__" "__package__")
- symbol-end) . font-lock-builtin-face)
;; assignments
;; support for a = b = c = 5
(,(lambda (limit)
@@ -640,7 +652,26 @@ The type returned can be `comment', `string' or `paren'."
(goto-char (match-end 1))
(python-syntax-context 'paren)))
res))
- (1 font-lock-variable-name-face nil nil))))
+ (1 font-lock-variable-name-face nil nil)))
+ "Font lock keywords to use in python-mode for maximum decoration.
+
+This decoration level includes everything in
+`python-font-lock-keywords-level-2', as well as constants,
+decorators, exceptions, and assignments.")
+
+(defvar python-font-lock-keywords
+ '(python-font-lock-keywords-level-1 ; When `font-lock-maximum-decoration'
is nil.
+ python-font-lock-keywords-level-1 ; When `font-lock-maximum-decoration'
is 1.
+ python-font-lock-keywords-level-2 ; When `font-lock-maximum-decoration'
is 2.
+ python-font-lock-keywords-maximum-decoration ; When
`font-lock-maximum-decoration'
+ ; is more than 1, or t (which
it is,
+ ; by default).
+ )
+ "List of font lock keyword specifications to use in python-mode.
+
+Which one will be chosen depends on the value of
+`font-lock-maximum-decoration'.")
+
(defconst python-syntax-propertize-function
(syntax-propertize-rules
@@ -5325,7 +5356,7 @@ REPORT-FN is Flymake's callback function."
'python-nav-forward-sexp)
(set (make-local-variable 'font-lock-defaults)
- '(python-font-lock-keywords
+ `(,python-font-lock-keywords
nil nil nil nil
(font-lock-syntactic-face-function
. python-font-lock-syntactic-face-function)))
diff --git a/lisp/progmodes/subword.el b/lisp/progmodes/subword.el
index ed71b86..685e171 100644
--- a/lisp/progmodes/subword.el
+++ b/lisp/progmodes/subword.el
@@ -110,9 +110,7 @@ called a `subword'. Here are some examples:
NSGraphicsContext => \"NS\", \"Graphics\" and \"Context\"
This mode changes the definition of a word so that word commands
-treat nomenclature boundaries as word boundaries.
-
-\\{subword-mode-map}"
+treat nomenclature boundaries as word boundaries."
:lighter " ,"
(when subword-mode (superword-mode -1))
(subword-setup-buffer))
diff --git a/lisp/register.el b/lisp/register.el
index 77d84c0..e25f9fd 100644
--- a/lisp/register.el
+++ b/lisp/register.el
@@ -231,6 +231,7 @@ Interactively, reads the register using
`register-read-with-preview'."
(defalias 'register-to-point 'jump-to-register)
(defun jump-to-register (register &optional delete)
"Move point to location stored in a register.
+Push the mark if jumping moves point, unless called in succession.
If the register contains a file name, find that file.
\(To put a file name in a register, you must use `set-register'.)
If the register contains a window configuration (one frame) or a frameset
@@ -390,7 +391,20 @@ Interactively, reads the register using
`register-read-with-preview'."
(cl-defmethod register-val-describe ((val cons) verbose)
(cond
((window-configuration-p (car val))
- (princ "a window configuration."))
+ (let* ((stored-window-config (car val))
+ (window-config-frame (window-configuration-frame
stored-window-config))
+ (current-frame (selected-frame)))
+ (princ (format "a window configuration: %s."
+ (if (frame-live-p window-config-frame)
+ (with-selected-frame window-config-frame
+ (save-window-excursion
+ (set-window-configuration stored-window-config)
+ (concat
+ (mapconcat (lambda (w) (buffer-name
(window-buffer w)))
+ (window-list (selected-frame)) ", ")
+ (unless (eq current-frame window-config-frame)
+ " in another frame"))))
+ "dead frame")))))
((frame-configuration-p (car val))
(princ "a frame configuration."))
diff --git a/lisp/scroll-bar.el b/lisp/scroll-bar.el
index 4d1ad03..7efbfc7 100644
--- a/lisp/scroll-bar.el
+++ b/lisp/scroll-bar.el
@@ -254,14 +254,22 @@ EVENT should be a scroll bar click or drag event."
(let* ((start-position (event-start event))
(window (nth 0 start-position))
(portion-whole (nth 2 start-position)))
- (save-excursion
- (with-current-buffer (window-buffer window)
- ;; Calculate position relative to the accessible part of the buffer.
- (goto-char (+ (point-min)
- (scroll-bar-scale portion-whole
- (- (point-max) (point-min)))))
- (vertical-motion 0 window)
- (set-window-start window (point))))))
+ ;; With 'scroll-bar-adjust-thumb-portion' nil and 'portion-whole'
+ ;; indicating that the buffer is fully visible, do not scroll the
+ ;; window since that might make it impossible to scroll it back
+ ;; with GTK's thumb (Bug#32002).
+ (when (or scroll-bar-adjust-thumb-portion
+ (not (numberp (car portion-whole)))
+ (not (numberp (cdr portion-whole)))
+ (/= (car portion-whole) (cdr portion-whole)))
+ (save-excursion
+ (with-current-buffer (window-buffer window)
+ ;; Calculate position relative to the accessible part of the buffer.
+ (goto-char (+ (point-min)
+ (scroll-bar-scale portion-whole
+ (- (point-max) (point-min)))))
+ (vertical-motion 0 window)
+ (set-window-start window (point)))))))
(defun scroll-bar-drag (event)
"Scroll the window by dragging the scroll bar slider.
diff --git a/lisp/shadowfile.el b/lisp/shadowfile.el
index 0095d69..86280c3 100644
--- a/lisp/shadowfile.el
+++ b/lisp/shadowfile.el
@@ -25,37 +25,38 @@
;; This package helps you to keep identical copies of files in more than one
;; place - possibly on different machines. When you save a file, it checks
;; whether it is on the list of files with "shadows", and if so, it tries to
-;; copy it when you exit Emacs (or use the shadow-copy-files command).
+;; copy it when you exit Emacs (or use the `shadow-copy-files' command).
;; Installation & Use:
-;; Add clusters (if necessary) and file groups with shadow-define-cluster,
-;; shadow-define-literal-group, and shadow-define-regexp-group (see the
+;; Add clusters (if necessary) and file groups with `shadow-define-cluster',
+;; `shadow-define-literal-group', and `shadow-define-regexp-group' (see the
;; documentation for these functions for information on how and when to use
;; them). After doing this once, everything should be automatic.
-;; The lists of clusters and shadows are saved in a ~/.emacs.d/shadows
-;; (`shadow-info-file') file, so that they can be remembered from one
-;; Emacs session to another, even (as much as possible) if the Emacs
-;; session terminates abnormally. The files needing to be copied are
-;; stored in `shadow-todo-file'; if a file cannot be copied for any
-;; reason, it will stay on the list to be tried again next time. The
-;; `shadow-info-file' file should itself have shadows on all your accounts
-;; so that the information in it is consistent everywhere, but
-;; `shadow-todo-file' is local information and should have no shadows.
+;; The lists of clusters and shadows are saved in `shadow-info-file',
+;; so that they can be remembered from one Emacs session to another,
+;; even (as much as possible) if the Emacs session terminates
+;; abnormally. The files needing to be copied are stored in
+;; `shadow-todo-file'; if a file cannot be copied for any reason, it
+;; will stay on the list to be tried again next time. The
+;; `shadow-info-file' file should itself have shadows on all your
+;; accounts so that the information in it is consistent everywhere,
+;; but `shadow-todo-file' is local information and should have no
+;; shadows.
;; If you do not want to copy a particular file, you can answer "no" and
-;; be asked again next time you hit C-x 4 s or exit Emacs. If you do not
-;; want to be asked again, use shadow-cancel, and you will not be asked
+;; be asked again next time you hit "C-x 4 s" or exit Emacs. If you do not
+;; want to be asked again, use "M-x shadow-cancel", and you will not be asked
;; until you change the file and save it again. If you do not want to
;; shadow that file ever again, you can edit it out of the shadows
-;; buffer. Anytime you edit the shadows buffer, you must type M-x
-;; shadow-read-files to load in the new information, or your changes will
+;; buffer. Anytime you edit the shadows buffer, you must type "M-x
+;; shadow-read-files" to load in the new information, or your changes will
;; be overwritten!
;; Bugs & Warnings:
;;
-;; - It is bad to have two emacses both running shadowfile at the same
+;; - It is bad to have two Emacsen both running shadowfile at the same
;; time. It tries to detect this condition, but is not always successful.
;;
;; - You have to be careful not to edit a file in two locations
@@ -64,19 +65,16 @@
;;
;; - It ought to check modification times of both files to make sure
;; it is doing the right thing. This will have to wait until
-;; file-newer-than-file-p works between machines.
+;; `file-newer-than-file-p' works between machines.
;;
;; - It will not make directories for you, it just fails to copy files
;; that belong in non-existent directories.
-;;
-;; Please report any bugs to me (address@hidden). Also let me know
-;; if you have suggestions or would like to be informed of updates.
;;; Code:
(require 'cl-lib)
-(require 'ange-ftp)
+(require 'tramp)
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;; Variables
@@ -107,35 +105,35 @@ files that have been changed and need to be copied to
other systems."
:type 'boolean
:group 'shadow)
-;; FIXME in a sense, this changed in 24.4 (addition of locate-user-emacs-file),
-;; but due to the weird way this variable is initialized to nil, it didn't
-;; literally change. Same for shadow-todo-file.
-(defcustom shadow-info-file nil
+(defcustom shadow-info-file (locate-user-emacs-file "shadows" ".shadows")
"File to keep shadow information in.
The `shadow-info-file' should be shadowed to all your accounts to
ensure consistency. Default: ~/.emacs.d/shadows"
- :type '(choice (const nil) file)
- :group 'shadow)
+ :type 'file
+ :group 'shadow
+ :version "26.2")
-(defcustom shadow-todo-file nil
+(defcustom shadow-todo-file
+ (locate-user-emacs-file "shadow_todo" ".shadow_todo")
"File to store the list of uncopied shadows in.
This means that if a remote system is down, or for any reason you cannot or
decide not to copy your shadow files at the end of one Emacs session, it will
remember and ask you again in your next Emacs session.
This file must NOT be shadowed to any other system, it is host-specific.
Default: ~/.emacs.d/shadow_todo"
- :type '(choice (const nil) file)
- :group 'shadow)
+ :type 'file
+ :group 'shadow
+ :version "26.2")
;;; The following two variables should in most cases initialize themselves
;;; correctly. They are provided as variables in case the defaults are wrong
;;; on your machine (and for efficiency).
-(defvar shadow-system-name (system-name)
- "The complete hostname of this machine.")
+(defvar shadow-system-name (concat "/" (system-name) ":")
+ "The identification for local files on this machine.")
-(defvar shadow-homedir nil
+(defvar shadow-homedir "~"
"Your home directory on this machine.")
;;;
@@ -186,12 +184,12 @@ created by `shadow-define-regexp-group'.")
(car list))
(defun shadow-regexp-superquote (string)
- "Like `regexp-quote', but includes the ^ and $.
+ "Like `regexp-quote', but includes the \\` and \\'.
This makes sure regexp matches nothing but STRING."
- (concat "^" (regexp-quote string) "$"))
+ (concat "\\`" (regexp-quote string) "\\'"))
(defun shadow-suffix (prefix string)
- "If PREFIX begins STRING, return the rest.
+ "If PREFIX begins with STRING, return the rest.
Return value is non-nil if PREFIX and STRING are `string=' up to the length of
PREFIX."
(let ((lp (length prefix))
@@ -204,70 +202,66 @@ PREFIX."
;;; Clusters and sites
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-;;; I use the term `site' to refer to a string which may be the name of a
-;;; cluster or a literal hostname. All user-level commands should accept
-;;; either.
-
-(defun shadow-make-cluster (name primary regexp)
- "Create a shadow cluster.
-It is called NAME, uses the PRIMARY hostname and REGEXP matching all
-hosts in the cluster. The variable `shadow-clusters' associates the
-names of clusters to these structures. This function is for program
-use: to create clusters interactively, use `shadow-define-cluster'
-instead."
- (list name primary regexp))
-
-(defmacro shadow-cluster-name (cluster)
- "Return the name of the CLUSTER."
- (list 'elt cluster 0))
+;;; I use the term `site' to refer to a string which may be the
+;;; cluster identification "/name:", a remote identification
+;;; "/method:address@hidden:", or "/system-name:' (the value of
+;;; `shadow-system-name') for the location of local files. All
+;;; user-level commands should accept either.
-(defmacro shadow-cluster-primary (cluster)
- "Return the primary hostname of a CLUSTER."
- (list 'elt cluster 1))
-
-(defmacro shadow-cluster-regexp (cluster)
- "Return the regexp matching hosts in a CLUSTER."
- (list 'elt cluster 2))
+(cl-defstruct (shadow-cluster (:type list) :named) name primary regexp)
(defun shadow-set-cluster (name primary regexp)
"Put cluster NAME on the list of clusters.
Replace old definition, if any. PRIMARY and REGEXP are the
information defining the cluster. For interactive use, call
`shadow-define-cluster' instead."
- (let ((rest (cl-remove-if (lambda (x) (equal name (car x)))
+ (let ((rest (cl-remove-if (lambda (x) (equal name (shadow-cluster-name x)))
shadow-clusters)))
(setq shadow-clusters
- (cons (shadow-make-cluster name primary regexp)
+ (cons (make-shadow-cluster :name name :primary primary :regexp regexp)
rest))))
-(defmacro shadow-get-cluster (name)
+(defun shadow-get-cluster (name)
"Return cluster named NAME, or nil."
- (list 'assoc name 'shadow-clusters))
+ (shadow-find
+ (lambda (x) (string-equal (shadow-cluster-name x) name))
+ shadow-clusters))
+
+;;; SITES
+
+(defun shadow-site-name (site)
+ "Return name if SITE has the form \"/name:\", otherwise SITE."
+ (if (string-match "\\`/\\([-.[:word:]]+\\):\\'" site)
+ (match-string 1 site) site))
+
+(defun shadow-name-site (name)
+ "Return \"/name:\" if NAME has word syntax, otherwise NAME."
+ (if (string-match "\\`[-.[:word:]]+\\'" name)
+ (format "/%s:"name) name))
(defun shadow-site-primary (site)
- "If SITE is a cluster, return primary host, otherwise return SITE."
- (let ((c (shadow-get-cluster site)))
- (if c
- (shadow-cluster-primary c)
+ "If SITE is a cluster, return primary identification, otherwise return SITE."
+ (let ((cluster (shadow-get-cluster (shadow-site-name site))))
+ (if cluster
+ (shadow-cluster-primary cluster)
site)))
-;;; SITES
-
(defun shadow-site-cluster (site)
- "Given a SITE (hostname or cluster name), return cluster it is in, or nil."
- (or (assoc site shadow-clusters)
+ "Given a SITE, return cluster it is in, or nil."
+ (or (shadow-get-cluster (shadow-site-name site))
(shadow-find
- (function (lambda (x)
- (string-match (shadow-cluster-regexp x)
- site)))
+ (lambda (x)
+ (string-match (shadow-cluster-regexp x) (shadow-name-site site)))
shadow-clusters)))
(defun shadow-read-site ()
- "Read a cluster name or hostname from the minibuffer."
- (let ((ans (completing-read "Host or cluster name [RET when done]: "
+ "Read a cluster name or host identification from the minibuffer."
+ (let ((ans (completing-read "Host identification or cluster name: "
shadow-clusters)))
- (if (equal "" ans)
- nil
+ (when (or (shadow-get-cluster (shadow-site-name ans))
+ (string-equal ans shadow-system-name)
+ (string-equal ans (shadow-site-name shadow-system-name))
+ (setq ans (file-remote-p ans)))
ans)))
(defun shadow-site-match (site1 site2)
@@ -281,63 +275,88 @@ be matched against the primary of SITE2."
(string-match (shadow-cluster-regexp cluster1) primary2)
(string-equal site1 primary2)))))
-(defun shadow-get-user (site)
- "Return the default username for a SITE."
- (ange-ftp-get-user (shadow-site-primary site)))
-
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;; Filename manipulation
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-(defun shadow-parse-fullname (fullname)
- "Parse FULLNAME into (site user path) list.
-Leave it alone if it already is one. Return nil if the argument is
-not a full ange-ftp pathname."
- (if (listp fullname)
- fullname
- (ange-ftp-ftp-name fullname)))
-
(defun shadow-parse-name (name)
- "Parse any NAME into (site user name) list.
-Argument can be a simple name, full ange-ftp name, or already a hup list."
- (or (shadow-parse-fullname name)
- (list shadow-system-name
- (user-login-name)
- name)))
-
-(defsubst shadow-make-fullname (host user name)
- "Make an ange-ftp style fullname out of HOST, USER (optional), and NAME.
-This is probably not as general as it ought to be."
- (concat "/"
- (if user (concat user "@"))
- host ":"
- name))
+ "Parse any NAME into a `tramp-file-name' structure.
+Argument can be a simple name, remote file name, or already a
+`tramp-file-name' structure."
+ (cond
+ ((null name) nil)
+ ((tramp-file-name-p name) name)
+ ((file-remote-p name) (tramp-dissect-file-name name))
+ ((shadow-local-file name)
+ (make-tramp-file-name
+ :host (shadow-site-name shadow-system-name)
+ :localname (shadow-local-file name)))
+ ;; Cluster name.
+ ((string-match "^/\\([^:/]+\\):\\([^:]*\\)$" name)
+ (let ((name (match-string 1 name))
+ (file (match-string 2 name)))
+ (when (shadow-get-cluster name)
+ (make-tramp-file-name :host name :localname file))))))
+
+(defsubst shadow-make-fullname (hup &optional host name)
+ "Make a Tramp style fullname out of HUP, a `tramp-file-name' structure.
+Replace HOST, and NAME when non-nil."
+ (let ((hup (copy-tramp-file-name hup)))
+ (when host (setf (tramp-file-name-host hup) host))
+ (when name (setf (tramp-file-name-localname hup) name))
+ (if (null (tramp-file-name-method hup))
+ (format
+ "/%s:%s" (tramp-file-name-host hup) (tramp-file-name-localname hup))
+ (tramp-make-tramp-file-name hup))))
(defun shadow-replace-name-component (fullname newname)
"Return FULLNAME with the name component changed to NEWNAME."
- (let ((hup (shadow-parse-fullname fullname)))
- (shadow-make-fullname (nth 0 hup) (nth 1 hup) newname)))
+ (concat (file-remote-p fullname) newname))
(defun shadow-local-file (file)
- "If FILE is at this site, remove /address@hidden part.
-If refers to a different system or a different user on this system,
-return nil."
- (let ((hup (shadow-parse-fullname file)))
- (cond ((null hup) file)
- ((and (shadow-site-match (nth 0 hup) shadow-system-name)
- (string-equal (nth 1 hup) (user-login-name)))
- (nth 2 hup))
- (t nil))))
+ "If FILE is not remote, return it.
+If it refers to a different system, return nil."
+ (cond
+ ((null file) nil)
+ ;; `tramp-file-name' structure.
+ ((and (tramp-file-name-p file) (null (tramp-file-name-method file)))
+ (tramp-file-name-localname file))
+ ((tramp-file-name-p file) nil)
+ ;; Local host name.
+ ((string-match
+ (format "^%s\\([^:]*\\)$" (regexp-quote shadow-system-name)) file)
+ (match-string 1 file))
+ ;; Cluster name.
+ ((and (string-match "^/\\([^:/]+\\):\\([^:]*\\)$" file)
+ (shadow-get-cluster (match-string 1 file)))
+ (let ((file (match-string 2 file))
+ (primary
+ (shadow-cluster-primary
+ (shadow-get-cluster (match-string 1 file)))))
+ (when (string-equal primary shadow-system-name) (setq primary nil))
+ (shadow-local-file (concat primary file))))
+ ;; Local name.
+ ((null (file-remote-p file)) file)))
(defun shadow-expand-cluster-in-file-name (file)
"If hostname part of FILE is a cluster, expand it to cluster's primary
hostname.
Will return the name bare if it is a local file."
- (let ((hup (shadow-parse-name file)))
- (cond ((null hup) file)
- ((shadow-local-file hup))
- ((shadow-make-fullname (shadow-site-primary (nth 0 hup))
- (nth 1 hup)
- (nth 2 hup))))))
+ (when (stringp file)
+ (cond
+ ;; Local file.
+ ((shadow-local-file file))
+ ;; Cluster name.
+ ((string-match "^\\(/[^:/]+:\\)[^:]*$" file)
+ (let ((primary
+ (save-match-data
+ (shadow-cluster-primary
+ (shadow-get-cluster
+ (shadow-site-name (match-string 1 file)))))))
+ (if (not primary)
+ file
+ (setq file (replace-match primary nil nil file 1))
+ (or (shadow-local-file file) file))))
+ (t file))))
(defun shadow-expand-file-name (file &optional default)
"Expand file name and get FILE's true name."
@@ -352,46 +371,50 @@ true."
(homedir (if (shadow-local-file hup)
shadow-homedir
(file-name-as-directory
- (nth 2 (shadow-parse-fullname
- (expand-file-name
- (shadow-make-fullname
- (nth 0 hup) (nth 1 hup) "~")))))))
- (suffix (shadow-suffix homedir (nth 2 hup)))
- (cluster (shadow-site-cluster (nth 0 hup))))
+ (file-local-name
+ (expand-file-name (shadow-make-fullname hup nil "~"))))))
+ (suffix (shadow-suffix homedir (tramp-file-name-localname hup)))
+ (cluster (shadow-site-cluster (shadow-make-fullname hup nil ""))))
+ (when cluster
+ (setf (tramp-file-name-method hup) nil
+ (tramp-file-name-host hup) (shadow-cluster-name cluster)))
(shadow-make-fullname
- (if cluster
- (shadow-cluster-name cluster)
- (nth 0 hup))
- (nth 1 hup)
+ hup nil
(if suffix
- (concat "~/" suffix)
- (nth 2 hup)))))
+ (concat "~/" suffix)
+ (tramp-file-name-localname hup)))))
(defun shadow-same-site (pattern file)
"True if the site of PATTERN and of FILE are on the same site.
-If usernames are supplied, they must also match exactly. PATTERN and FILE may
-be lists of host, user, name, or ange-ftp file names. FILE may also be just a
-local filename."
- (let ((pattern-sup (shadow-parse-fullname pattern))
+PATTERN and FILE may be Tramp vectors, or remote file names.
+FILE may also be just a local filename."
+ (let ((pattern-sup (shadow-parse-name pattern))
(file-sup (shadow-parse-name file)))
(and
- (shadow-site-match (nth 0 pattern-sup) (nth 0 file-sup))
- (or (null (nth 1 pattern-sup))
- (string-equal (nth 1 pattern-sup) (nth 1 file-sup))))))
+ (shadow-site-match
+ (tramp-file-name-host pattern-sup) (tramp-file-name-host file-sup))
+ (or (null (tramp-file-name-user pattern-sup))
+ (string-equal
+ (tramp-file-name-user pattern-sup)
+ (tramp-file-name-user file-sup))))))
(defun shadow-file-match (pattern file &optional regexp)
"Return t if PATTERN matches FILE.
If REGEXP is supplied and non-nil, the file part of the pattern is a regular
-expression, otherwise it must match exactly. The sites and usernames must
-match---see `shadow-same-site'. The pattern must be in full ange-ftp format,
+expression, otherwise it must match exactly. The sites must
+match---see `shadow-same-site'. The pattern must be in full Tramp format,
but the file can be any valid filename. This function does not do any
filename expansion or contraction, you must do that yourself first."
- (let* ((pattern-sup (shadow-parse-fullname pattern))
+ (let* ((pattern-sup (shadow-parse-name pattern))
(file-sup (shadow-parse-name file)))
(and (shadow-same-site pattern-sup file-sup)
(if regexp
- (string-match (nth 2 pattern-sup) (nth 2 file-sup))
- (string-equal (nth 2 pattern-sup) (nth 2 file-sup))))))
+ (string-match
+ (tramp-file-name-localname pattern-sup)
+ (tramp-file-name-localname file-sup))
+ (string-equal
+ (tramp-file-name-localname pattern-sup)
+ (tramp-file-name-localname file-sup))))))
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;; User-level Commands
@@ -405,30 +428,34 @@ one of them is sufficient to update the file on all of
them. Clusters are
defined by a name, the network address of a primary host (the one we copy
files to), and a regular expression that matches the hostnames of all the
sites in the cluster."
- (interactive (list (completing-read "Cluster name: " shadow-clusters () ())))
+ (interactive (list (completing-read "Cluster name: " shadow-clusters)))
(let* ((old (shadow-get-cluster name))
- (primary (read-string "Primary host: "
- (if old (shadow-cluster-primary old)
- name)))
- (regexp (let (try-regexp)
- (while (not
- (string-match
- (setq try-regexp
+ (primary (let (try-primary)
+ (while (not
+ (or
+ (string-equal
+ (setq try-primary
(read-string
- "Regexp matching all host names: "
- (if old (shadow-cluster-regexp old)
- (shadow-regexp-superquote primary))))
- primary))
- (message "Regexp doesn't include the primary host!")
- (sit-for 2))
- try-regexp))
-; (username (read-no-blanks-input
-; (format "Username (default %s): "
-; (shadow-get-user primary))
-; (if old (or (shadow-cluster-username old) "")
-; (user-login-name))))
- )
-; (if (string-equal "" username) (setq username nil))
+ "Primary host: "
+ (if old (shadow-cluster-primary old)
+ name)))
+ shadow-system-name)
+ (file-remote-p try-primary)))
+ (message "Not a valid primary!")
+ (sit-for 2))
+ try-primary))
+ (regexp (let (try-regexp)
+ (while (not
+ (string-match
+ (setq try-regexp
+ (read-string
+ "Regexp matching all host names: "
+ (if old (shadow-cluster-regexp old)
+ (shadow-regexp-superquote primary))))
+ primary))
+ (message "Regexp doesn't include the primary host!")
+ (sit-for 2))
+ try-regexp)))
(shadow-set-cluster name primary regexp)))
;;;###autoload
@@ -438,20 +465,14 @@ It may have different filenames on each site. When this
file is edited, the
new version will be copied to each of the other locations. Sites can be
specific hostnames, or names of clusters (see `shadow-define-cluster')."
(interactive)
- (let* ((hup (shadow-parse-fullname
+ (let* ((hup (shadow-parse-name
(shadow-contract-file-name (buffer-file-name))))
- (name (nth 2 hup))
- user site group)
+ (name (tramp-file-name-localname hup))
+ site group)
(while (setq site (shadow-read-site))
- (setq user (read-string (format "Username (default %s): "
- (shadow-get-user site)))
- name (read-string "Filename: " name))
- (setq group (cons (shadow-make-fullname site
- (if (string-equal "" user)
- (shadow-get-user site)
- user)
- name)
- group)))
+ (setq name (read-string "Filename: " name)
+ hup (shadow-parse-name (shadow-contract-file-name name))
+ group (cons (shadow-make-fullname hup site) group)))
(setq shadow-literal-groups (cons group shadow-literal-groups)))
(shadow-write-info-file))
@@ -468,19 +489,12 @@ function). Each site can be either a hostname or the
name of a cluster (see
"Filename regexp: "
(if (buffer-file-name)
(shadow-regexp-superquote
- (nth 2
- (shadow-parse-name
- (shadow-contract-file-name
- (buffer-file-name))))))))
- site sites usernames)
+ (file-local-name (buffer-file-name))))))
+ site sites)
(while (setq site (shadow-read-site))
- (setq sites (cons site sites))
- (setq usernames
- (cons (read-string (format "Username for %s: " site)
- (shadow-get-user site))
- usernames)))
+ (setq sites (cons site sites)))
(setq shadow-regexp-groups
- (cons (shadow-make-group regexp sites usernames)
+ (cons (shadow-make-group regexp sites)
shadow-regexp-groups))
(shadow-write-info-file)))
@@ -537,14 +551,14 @@ permanently, remove the group from
`shadow-literal-groups' or
;;; Internal functions
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-(defun shadow-make-group (regexp sites usernames)
+(defun shadow-make-group (regexp sites)
"Make a description of a file group---
-actually a list of regexp ange-ftp file names---from REGEXP (name of file to
-be shadowed), list of SITES, and corresponding list of USERNAMES for each
-site."
+actually a list of regexp Tramp file names---from REGEXP (name of file to
+be shadowed), and list of SITES"
(if sites
- (cons (shadow-make-fullname (car sites) (car usernames) regexp)
- (shadow-make-group regexp (cdr sites) (cdr usernames)))
+ (cons (shadow-make-fullname
+ (shadow-parse-name (shadow-site-primary (car sites))) nil regexp)
+ (shadow-make-group regexp (cdr sites)))
nil))
(defun shadow-copy-file (s)
@@ -601,7 +615,9 @@ Consider them as regular expressions if third arg REGEXP is
true."
(car groups))))
(append (cond ((equal nonmatching (car groups)) nil)
(regexp
- (let ((realname (nth 2 (shadow-parse-fullname file))))
+ (let ((realname
+ (tramp-file-name-localname
+ (shadow-parse-name file))))
(mapcar
(function
(lambda (x)
@@ -612,17 +628,26 @@ Consider them as regular expressions if third arg REGEXP
is true."
(defun shadow-add-to-todo ()
"If current buffer has shadows, add them to the list needing to be copied."
+ (message "shadow-add-to-todo 1 %s" (current-buffer))
+ (message "shadow-add-to-todo 2 %s" (buffer-file-name))
+ (message "shadow-add-to-todo 3 %s" (shadow-expand-file-name
(buffer-file-name (current-buffer))))
+ (message "shadow-add-to-todo 4 %s" (shadow-shadows-of
(shadow-expand-file-name (buffer-file-name (current-buffer)))))
(let ((shadows (shadow-shadows-of
(shadow-expand-file-name
(buffer-file-name (current-buffer))))))
(when shadows
+ (message "shadow-add-to-todo 5 %s" shadows)
+ (message "shadow-add-to-todo 6 %s" shadow-files-to-copy)
+ (message "shadow-add-to-todo 7 %s" (shadow-union shadows
shadow-files-to-copy))
(setq shadow-files-to-copy
(shadow-union shadows shadow-files-to-copy))
(when (not shadow-inhibit-message)
(message "%s" (substitute-command-keys
"Use \\[shadow-copy-files] to update shadows."))
(sit-for 1))
- (shadow-write-todo-file)))
+ (message "shadow-add-to-todo 8")
+ (shadow-write-todo-file)
+ (message "shadow-add-to-todo 9")))
nil) ; Return nil for write-file-functions
(defun shadow-remove-from-todo (pair)
@@ -636,9 +661,8 @@ PAIR must be `eq' to one of the elements of that list."
Thus restores shadowfile's state from your last Emacs session.
Return t unless files were locked; then return nil."
(interactive)
- (if (and (fboundp 'file-locked-p)
- (or (stringp (file-locked-p shadow-info-file))
- (stringp (file-locked-p shadow-todo-file))))
+ (if (or (stringp (file-locked-p shadow-info-file))
+ (stringp (file-locked-p shadow-todo-file)))
(progn
(message "Shadowfile is running in another Emacs; can't have two.")
(beep)
@@ -647,7 +671,7 @@ Return t unless files were locked; then return nil."
(save-current-buffer
(when shadow-info-file
(set-buffer (setq shadow-info-buffer
- (find-file-noselect shadow-info-file)))
+ (find-file-noselect shadow-info-file 'nowarn)))
(when (and (not (buffer-modified-p))
(file-newer-than-file-p (make-auto-save-file-name)
shadow-info-file))
@@ -680,6 +704,7 @@ defined, the old hashtable info is invalid."
(if (not shadow-info-buffer)
(setq shadow-info-buffer (find-file-noselect shadow-info-file)))
(set-buffer shadow-info-buffer)
+ (setq buffer-read-only nil)
(delete-region (point-min) (point-max))
(shadow-insert-var 'shadow-clusters)
(shadow-insert-var 'shadow-literal-groups)
@@ -689,17 +714,26 @@ defined, the old hashtable info is invalid."
"Write out information to `shadow-todo-file'.
With non-nil argument also saves the buffer."
(save-excursion
+ (message "shadow-write-todo-file 1 %s" shadow-todo-buffer)
(if (not shadow-todo-buffer)
(setq shadow-todo-buffer (find-file-noselect shadow-todo-file)))
+ (message "shadow-write-todo-file 2 %s" shadow-todo-buffer)
(set-buffer shadow-todo-buffer)
+ (message "shadow-write-todo-file 3 %s" shadow-todo-buffer)
+ (setq buffer-read-only nil)
(delete-region (point-min) (point-max))
+ (message "shadow-write-todo-file 4 %s" shadow-todo-buffer)
(shadow-insert-var 'shadow-files-to-copy)
- (if save (shadow-save-todo-file))))
+ (message "shadow-write-todo-file 5 %s" save)
+ (if save (shadow-save-todo-file))
+ (message "shadow-write-todo-file 6 %s" save)))
(defun shadow-save-todo-file ()
+ (message "shadow-save-todo-file 1 %s" shadow-todo-buffer)
(if (and shadow-todo-buffer (buffer-modified-p shadow-todo-buffer))
(with-current-buffer shadow-todo-buffer
- (condition-case nil ; have to continue even in case of
+ (message "shadow-save-todo-file 2 %s" shadow-todo-buffer)
+ (condition-case nil ; have to continue even in case of
(basic-save-buffer) ; error, otherwise kill-emacs might
(error ; not work!
(message "WARNING: Can't save shadow todo file; it is locked!")
@@ -765,24 +799,6 @@ look for files that have been changed and need to be
copied to other systems."
(kill-emacs)))
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-;;; Lucid Emacs compatibility
-;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
-
-;; This is on hold until someone tells me about a working version of
-;; map-ynp for Lucid Emacs.
-
-;(when (string-match "Lucid" emacs-version)
-; (require 'symlink-fix)
-; (require 'ange-ftp)
-; (require 'map-ynp)
-; (if (not (fboundp 'file-truename))
-; (fset 'shadow-expand-file-name
-; (symbol-function 'symlink-expand-file-name)))
-; (if (not (fboundp 'ange-ftp-ftp-name))
-; (fset 'ange-ftp-ftp-name
-; (symbol-function 'ange-ftp-ftp-name))))
-
-;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;; Hook us up
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
@@ -790,18 +806,10 @@ look for files that have been changed and need to be
copied to other systems."
(defun shadow-initialize ()
"Set up file shadowing."
(interactive)
- (if (null shadow-homedir)
- (setq shadow-homedir
- (file-name-as-directory (shadow-expand-file-name "~"))))
- (if (null shadow-info-file)
- (setq shadow-info-file
- ;; FIXME: Move defaults to their defcustom.
- (shadow-expand-file-name
- (locate-user-emacs-file "shadows" ".shadows"))))
- (if (null shadow-todo-file)
- (setq shadow-todo-file
- (shadow-expand-file-name
- (locate-user-emacs-file "shadow_todo" ".shadow_todo"))))
+ (setq shadow-homedir
+ (file-name-as-directory (shadow-expand-file-name shadow-homedir))
+ shadow-info-file (shadow-expand-file-name shadow-info-file)
+ shadow-todo-file (shadow-expand-file-name shadow-todo-file))
(if (not (shadow-read-files))
(progn
(message "Shadowfile information files not found - aborting")
diff --git a/lisp/shell.el b/lisp/shell.el
index fa6eee0..ac6f11a 100644
--- a/lisp/shell.el
+++ b/lisp/shell.el
@@ -73,7 +73,7 @@
;; c-c c-o comint-delete-output Delete last batch of process
output
;; c-c c-r comint-show-output Show last batch of process output
;; c-c c-l comint-dynamic-list-input-ring List input history
-;; send-invisible Read line w/o echo & send to proc
+;; comint-send-invisible Read line w/o echo & send to proc
;; comint-continue-subjob Useful if you accidentally suspend
;; top-level job
;; comint-mode-hook is the comint mode hook.
@@ -500,7 +500,7 @@ Shell buffers. It implements `shell-completion-execonly'
for
the end of process to the end of the current line.
\\[comint-send-input] before end of process output copies the current line
minus the prompt to
the end of the buffer and sends it (\\[comint-copy-old-input] just copies
the current line).
-\\[send-invisible] reads a line of text without echoing it, and sends it to
+\\[comint-send-invisible] reads a line of text without echoing it, and sends
it to
the shell. This is useful for entering passwords. Or, add the function
`comint-watch-for-password-prompt' to `comint-output-filter-functions'.
diff --git a/lisp/simple.el b/lisp/simple.el
index 6459531..8d77047 100644
--- a/lisp/simple.el
+++ b/lisp/simple.el
@@ -8346,14 +8346,12 @@ LSHIFTBY is the numeric value of this modifier, in
keyboard events.
PREFIX is the string that represents this modifier in an event type symbol."
(if (numberp event)
(cond ((eq symbol 'control)
- (if (and (<= (downcase event) ?z)
- (>= (downcase event) ?a))
- (- (downcase event) ?a -1)
- (if (and (<= (downcase event) ?Z)
- (>= (downcase event) ?A))
- (- (downcase event) ?A -1)
- (logior (lsh 1 lshiftby) event))))
+ (if (<= 64 (upcase event) 95)
+ (- (upcase event) 64)
+ (logior (lsh 1 lshiftby) event)))
((eq symbol 'shift)
+ ;; FIXME: Should we also apply this "upcase" behavior of shift
+ ;; to non-ascii letters?
(if (and (<= (downcase event) ?z)
(>= (downcase event) ?a))
(upcase event)
diff --git a/lisp/subr.el b/lisp/subr.el
index 10343e6..fbb3e49 100644
--- a/lisp/subr.el
+++ b/lisp/subr.el
@@ -555,12 +555,6 @@ If N is omitted or nil, remove the last element."
(if (> n 0) (setcdr (nthcdr (- (1- m) n) list) nil))
list))))
-(defun proper-list-p (object)
- "Return OBJECT's length if it is a proper list, nil otherwise.
-A proper list is neither circular nor dotted (i.e., its last cdr
-is nil)."
- (and (listp object) (ignore-errors (length object))))
-
(defun delete-dups (list)
"Destructively remove `equal' duplicates from LIST.
Store the result in LIST and return it. LIST must be a proper list.
@@ -2305,7 +2299,7 @@ some sort of escape sequence, the ambiguity is resolved
via `read-key-delay'."
If optional CONFIRM is non-nil, read the password twice to make sure.
Optional DEFAULT is a default password to use instead of empty input.
-This function echoes `.' for each character that the user types.
+This function echoes `*' for each character that the user types.
You could let-bind `read-hide-char' to another hiding character, though.
Once the caller uses the password, it can erase the password
@@ -2331,7 +2325,7 @@ by doing (clear-string STRING)."
beg)))
(dotimes (i (- end beg))
(put-text-property (+ i beg) (+ 1 i beg)
- 'display (string (or read-hide-char ?.))))))
+ 'display (string (or read-hide-char ?*))))))
minibuf)
(minibuffer-with-setup-hook
(lambda ()
@@ -2346,7 +2340,7 @@ by doing (clear-string STRING)."
(add-hook 'after-change-functions hide-chars-fun nil 'local))
(unwind-protect
(let ((enable-recursive-minibuffers t)
- (read-hide-char (or read-hide-char ?.)))
+ (read-hide-char (or read-hide-char ?*)))
(read-string prompt nil t default)) ; t = "no history"
(when (buffer-live-p minibuf)
(with-current-buffer minibuf
@@ -4693,25 +4687,6 @@ The properties used on SYMBOL are `composefunc',
`sendfunc',
(put symbol 'hookvar (or hookvar 'mail-send-hook)))
-(defun backtrace--print-frame (evald func args flags)
- "Print a trace of a single stack frame to `standard-output'.
-EVALD, FUNC, ARGS, FLAGS are as in `mapbacktrace'."
- (princ (if (plist-get flags :debug-on-exit) "* " " "))
- (cond
- ((and evald (not debugger-stack-frame-as-list))
- (cl-prin1 func)
- (if args (cl-prin1 args) (princ "()")))
- (t
- (cl-prin1 (cons func args))))
- (princ "\n"))
-
-(defun backtrace ()
- "Print a trace of Lisp function calls currently active.
-Output stream used is value of `standard-output'."
- (let ((print-level (or print-level 8))
- (print-escape-control-characters t))
- (mapbacktrace #'backtrace--print-frame 'backtrace)))
-
(defun backtrace-frames (&optional base)
"Collect all frames of current backtrace into a list.
If non-nil, BASE should be a function, and frames before its
diff --git a/lisp/term.el b/lisp/term.el
index 121a22e..9f8f1f7 100644
--- a/lisp/term.el
+++ b/lisp/term.el
@@ -343,6 +343,7 @@
(eval-when-compile (require 'cl-lib))
(require 'ring)
(require 'ehelp)
+(require 'comint) ; Password regexp.
(declare-function ring-empty-p "ring" (ring))
(declare-function ring-ref "ring" (ring index))
@@ -2215,6 +2216,7 @@ filter and C-g is pressed, this function returns nil
rather than a string).
Note that the keystrokes comprising the text can still be recovered
\(temporarily) with \\[view-lossage]. This may be a security bug for some
applications."
+ (declare (obsolete read-passwd "27.1"))
(let ((ans "")
(c 0)
(echo-keystrokes 0)
@@ -2255,12 +2257,10 @@ applications."
(defun term-send-invisible (str &optional proc)
"Read a string without echoing.
Then send it to the process running in the current buffer. A new-line
-is additionally sent. String is not saved on term input history list.
-Security bug: your string can still be temporarily recovered with
-\\[view-lossage]."
+is additionally sent. String is not saved on term input history list."
(interactive "P") ; Defeat snooping via C-x esc
(when (not (stringp str))
- (setq str (term-read-noecho "Non-echoed text: " t)))
+ (setq str (read-passwd "Non-echoed text: ")))
(when (not proc)
(setq proc (get-buffer-process (current-buffer))))
(if (not proc) (error "Current buffer has no process")
@@ -2269,6 +2269,16 @@ Security bug: your string can still be temporarily
recovered with
(term-send-string proc str)
(term-send-string proc "\n")))
+;; TODO: Maybe combine this with `comint-watch-for-password-prompt'.
+(defun term-watch-for-password-prompt (string)
+ "Prompt in the minibuffer for password and send without echoing.
+Checks if STRING contains a password prompt as defined by
+`comint-password-prompt-regexp'."
+ (when (term-in-line-mode)
+ (when (let ((case-fold-search t))
+ (string-match comint-password-prompt-regexp string))
+ (term-send-invisible (read-passwd string)))))
+
;;; Low-level process communication
@@ -3054,6 +3064,8 @@ See `term-prompt-regexp'."
(term-handle-deferred-scroll))
(set-marker (process-mark proc) (point))
+ (when (stringp decoded-substring)
+ (term-watch-for-password-prompt decoded-substring))
(when save-point
(goto-char save-point)
(set-marker save-point nil))
diff --git a/lisp/term/tty-colors.el b/lisp/term/tty-colors.el
index ab9149e..a776c83 100644
--- a/lisp/term/tty-colors.el
+++ b/lisp/term/tty-colors.el
@@ -824,10 +824,12 @@ A canonicalized color name is all-lower case, with any
blanks removed."
(replace-regexp-in-string " +" "" (downcase color))
color)))
-(defun tty-color-24bit (rgb)
- "Return pixel value on 24-bit terminals. Return nil if RGB is
-nil or not on 24-bit terminal."
- (when (and rgb (= (display-color-cells) 16777216))
+(defun tty-color-24bit (rgb &optional display)
+ "Return 24-bit color pixel value for RGB value on DISPLAY.
+DISPLAY can be a display name or a frame, and defaults to the
+selected frame's display.
+If DISPLAY is not on a 24-but TTY terminal, return nil."
+ (when (and rgb (= (display-color-cells display) 16777216))
(let ((r (lsh (car rgb) -8))
(g (lsh (cadr rgb) -8))
(b (lsh (nth 2 rgb) -8)))
@@ -850,7 +852,7 @@ If FRAME is not specified or is nil, it defaults to the
selected frame."
(error "Invalid specification for tty color \"%s\"" name))
(tty-modify-color-alist
(append (list (tty-color-canonicalize name)
- (or (tty-color-24bit rgb) index))
+ (or (tty-color-24bit rgb frame) index))
rgb)
frame))
@@ -1026,7 +1028,7 @@ might need to be approximated if it is not supported
directly."
(or (assoc color (tty-color-alist frame))
(let ((rgb (tty-color-standard-values color)))
(and rgb
- (let ((pixel (tty-color-24bit rgb)))
+ (let ((pixel (tty-color-24bit rgb frame)))
(or (and pixel (cons color (cons pixel rgb)))
(tty-color-approximate rgb frame)))))))))
diff --git a/lisp/textmodes/flyspell.el b/lisp/textmodes/flyspell.el
index 8ad6832..69bba10 100644
--- a/lisp/textmodes/flyspell.el
+++ b/lisp/textmodes/flyspell.el
@@ -31,10 +31,10 @@
;;
;; To enable Flyspell in text representing computer programs, type
;; M-x flyspell-prog-mode.
-;; In that mode only text inside comments is checked.
+;; In that mode only text inside comments and strings is checked.
;;
;; Some user variables control the behavior of flyspell. They are
-;; those defined under the `User variables' comment.
+;; those defined under the `User configuration' comment.
;;; Code:
@@ -137,7 +137,10 @@ This variable specifies how far to search to find such a
duplicate.
(defcustom flyspell-persistent-highlight t
"Non-nil means misspelled words remain highlighted until corrected.
If this variable is nil, only the most recently detected misspelled word
-is highlighted."
+is highlighted, and the highlight is turned off as soon as point moves
+off the misspelled word.
+
+Make sure this variable is non-nil if you use `flyspell-region'."
:group 'flyspell
:type 'boolean)
@@ -1371,7 +1374,10 @@ language."
;;* flyspell-small-region ... */
;;*---------------------------------------------------------------------*/
(defun flyspell-small-region (beg end)
- "Flyspell text between BEG and END."
+ "Flyspell text between BEG and END.
+
+This function is intended to work on small regions, as
+determined by `flyspell-large-region'."
(save-excursion
(if (> beg end)
(let ((old beg))
@@ -1642,7 +1648,10 @@ The buffer to mark them in is
`flyspell-large-region-buffer'."
;;*---------------------------------------------------------------------*/
;;;###autoload
(defun flyspell-region (beg end)
- "Flyspell text between BEG and END."
+ "Flyspell text between BEG and END.
+
+Make sure `flyspell-mode' is turned on if you want the highlight
+of a misspelled word removed when you've corrected it."
(interactive "r")
(ispell-set-spellchecker-params) ; Initialize variables and dicts alists
(if (= beg end)
diff --git a/lisp/textmodes/ispell.el b/lisp/textmodes/ispell.el
index d80447e..e6f436f 100644
--- a/lisp/textmodes/ispell.el
+++ b/lisp/textmodes/ispell.el
@@ -2262,8 +2262,9 @@ Global `ispell-quit' set to start location to continue
spell session."
(ispell-pdict-save ispell-silently-savep)
(message "%s"
(substitute-command-keys
- (concat "Spell-checking suspended;"
- " use C-u \\[ispell-word] to resume")))
+ (concat
+ "Spell-checking suspended; use "
+ "\\[universal-argument] \\[ispell-word] to resume")))
(setq ispell-quit start)
nil)
((= char ?q)
diff --git a/lisp/textmodes/reftex-vars.el b/lisp/textmodes/reftex-vars.el
index 11dbb8d..e7fe8ff 100644
--- a/lisp/textmodes/reftex-vars.el
+++ b/lisp/textmodes/reftex-vars.el
@@ -1030,7 +1030,9 @@ This is used to string together whole reference sets, like
("Hyperref" "hyperref"
(("\\autoref" ?a) ("\\autopageref" ?u)))
("Cleveref" "cleveref"
- (("\\cref" ?c) ("\\Cref" ?C) ("\\cpageref" ?d) ("\\Cpageref" ?D))))
+ (("\\cref" ?c) ("\\Cref" ?C) ("\\cpageref" ?d) ("\\Cpageref" ?D)))
+ ("AMSmath" "amsmath"
+ (("\\eqref" ?e))))
"Alist of reference styles.
Each element is a list of the style name, the name of the LaTeX
package associated with the style or t for any package, and an
@@ -1040,7 +1042,7 @@ the macro type is being prompted for. (See also
`reftex-ref-macro-prompt'.) The keys, represented as characters,
have to be unique."
:group 'reftex-referencing-labels
- :version "24.3"
+ :version "27.1"
:type '(alist :key-type (string :tag "Style name")
:value-type (group (choice :tag "Package"
(const :tag "Any package" t)
diff --git a/lisp/thingatpt.el b/lisp/thingatpt.el
index 4612e95..7fcb3bc 100644
--- a/lisp/thingatpt.el
+++ b/lisp/thingatpt.el
@@ -484,7 +484,7 @@ looks like an email address, \"ftp://\"; if it starts with
(put 'url 'end-op (lambda () (end-of-thing 'url)))
-(put 'url 'beginning-op (lambda () (end-of-thing 'url)))
+(put 'url 'beginning-op (lambda () (beginning-of-thing 'url)))
;; The normal thingatpt mechanism doesn't work for complex regexps.
;; This should work for almost any regexp wherever we are in the
diff --git a/lisp/vc/add-log.el b/lisp/vc/add-log.el
index 4d69aac..d6e8540 100644
--- a/lisp/vc/add-log.el
+++ b/lisp/vc/add-log.el
@@ -744,6 +744,7 @@ Optional arg BUFFER-FILE overrides `buffer-file-name'."
file-name)
(defun add-log-file-name (buffer-file log-file)
+ "Compute file-name of BUFFER-FILE to be used in entries in LOG-FILE."
;; Never want to add a change log entry for the ChangeLog file itself.
(unless (or (null buffer-file) (string= buffer-file log-file))
(if add-log-file-name-function
@@ -767,15 +768,57 @@ Optional arg BUFFER-FILE overrides `buffer-file-name'."
(file-name-sans-versions buffer-file)
buffer-file))))
+(defcustom add-log-dont-create-changelog-file t
+ "If non-nil, don't create ChangeLog files for log entries.
+If a ChangeLog file does not already exist, a non-nil value
+means to put log entries in a suitably named buffer."
+ :type :boolean
+ :version "27.1")
+
+(put 'add-log-dont-create-changelog-file 'safe-local-variable 'booleanp)
+
+(defun add-log--pseudo-changelog-buffer-name (changelog-file-name)
+ "Compute a suitable name for a non-file visiting ChangeLog buffer.
+CHANGELOG-FILE-NAME is the file name of the actual ChangeLog file
+if it were to exist."
+ (format "*changes to %s*"
+ (abbreviate-file-name
+ (file-name-directory changelog-file-name))))
+
+(defun add-log--changelog-buffer-p (changelog-file-name buffer)
+ "Return non-nil if BUFFER holds a change log for CHANGELOG-FILE-NAME."
+ (with-current-buffer buffer
+ (if buffer-file-name
+ (equal buffer-file-name changelog-file-name)
+ (equal (add-log--pseudo-changelog-buffer-name changelog-file-name)
+ (buffer-name)))))
+
+(defun add-log-find-changelog-buffer (changelog-file-name)
+ "Find a ChangeLog buffer for CHANGELOG-FILE-NAME.
+Respect `add-log-use-pseudo-changelog', which see."
+ (if (or (file-exists-p changelog-file-name)
+ (not add-log-dont-create-changelog-file))
+ (find-file-noselect changelog-file-name)
+ (get-buffer-create
+ (add-log--pseudo-changelog-buffer-name changelog-file-name))))
+
;;;###autoload
-(defun add-change-log-entry (&optional whoami file-name other-window new-entry
+(defun add-change-log-entry (&optional whoami
+ changelog-file-name
+ other-window new-entry
put-new-entry-on-new-line)
- "Find change log file, and add an entry for today and an item for this file.
-Optional arg WHOAMI (interactive prefix) non-nil means prompt for user
-name and email (stored in `add-log-full-name' and `add-log-mailing-address').
-
-Second arg FILE-NAME is file name of the change log.
-If nil, use the value of `change-log-default-name'.
+ "Find ChangeLog buffer, add an entry for today and an item for this file.
+Optional arg WHOAMI (interactive prefix) non-nil means prompt for
+user name and email (stored in `add-log-full-name'
+and `add-log-mailing-address').
+
+Second arg CHANGELOG-FILE-NAME is the file name of the change log.
+If nil, use the value of `change-log-default-name'. If the file
+thus named exists, it is used for the new entry. If it doesn't
+exist, it is created, unless `add-log-dont-create-changelog-file' is t,
+in which case a suitably named buffer that doesn't visit any file
+is used for keeping entries pertaining to CHANGELOG-FILE-NAME's
+directory.
Third arg OTHER-WINDOW non-nil means visit in other window.
@@ -804,20 +847,28 @@ non-nil, otherwise in local time."
(change-log-version-number-search)))
(buf-file-name (funcall add-log-buffer-file-name-function))
(buffer-file (if buf-file-name (expand-file-name buf-file-name)))
- (file-name (expand-file-name (find-change-log file-name buffer-file)))
+ (changelog-file-name (expand-file-name (find-change-log
+ changelog-file-name
+ buffer-file)))
;; Set ITEM to the file name to use in the new item.
- (item (add-log-file-name buffer-file file-name)))
+ (item (add-log-file-name buffer-file changelog-file-name)))
- (unless (equal file-name buffer-file-name)
+ ;; don't add entries from the ChangeLog file/buffer to itself.
+ (unless (equal changelog-file-name buffer-file-name)
(cond
- ((equal file-name (buffer-file-name (window-buffer)))
+ ((add-log--changelog-buffer-p
+ changelog-file-name
+ (window-buffer))
;; If the selected window already shows the desired buffer don't show
;; it again (particularly important if other-window is true).
;; This is important for diff-add-change-log-entries-other-window.
(set-buffer (window-buffer)))
((or other-window (window-dedicated-p))
- (find-file-other-window file-name))
- (t (find-file file-name))))
+ (switch-to-buffer-other-window
+ (add-log-find-changelog-buffer changelog-file-name)))
+ (t
+ (switch-to-buffer
+ (add-log-find-changelog-buffer changelog-file-name)))))
(or (derived-mode-p 'change-log-mode)
(change-log-mode))
(undo-boundary)
diff --git a/lisp/vc/diff-mode.el b/lisp/vc/diff-mode.el
index ffbd9e5..b91a2ba 100644
--- a/lisp/vc/diff-mode.el
+++ b/lisp/vc/diff-mode.el
@@ -96,6 +96,11 @@ when editing big diffs)."
:version "27.1"
:type 'boolean)
+(defcustom diff-font-lock-prettify nil
+ "If non-nil, font-lock will try and make the format prettier."
+ :version "27.1"
+ :type 'boolean)
+
(defvar diff-vc-backend nil
"The VC backend that created the current Diff buffer, if any.")
@@ -396,6 +401,7 @@ and the face `diff-added' for added lines.")
(1 font-lock-comment-delimiter-face)
(2 font-lock-comment-face))
("^[^-=+*!<>#].*\n" (0 'diff-context))
+ (,#'diff--font-lock-prettify)
(,#'diff--font-lock-refined)))
(defconst diff-font-lock-defaults
@@ -2195,6 +2201,35 @@ fixed, visit it in a buffer."
modified-buffers ", "))
(message "No trailing whitespace to delete.")))))
+
+;;; Prettifying from font-lock
+
+(defun diff--font-lock-prettify (limit)
+ ;; Mimicks the output of Magit's diff.
+ ;; FIXME: This has only been tested with Git's diff output.
+ (when diff-font-lock-prettify
+ (while (re-search-forward "^diff " limit t)
+ (when (save-excursion
+ (forward-line 0)
+ (looking-at (eval-when-compile
+ (concat "diff.*\n"
+ "\\(?:\\(?:new
file\\|deleted\\).*\n\\)?"
+ "\\(?:index.*\n\\)?"
+ "--- \\(?:/dev/null\\|a/\\(.*\\)\\)\n"
+ "\\+\\+\\+
\\(?:/dev/null\\|b/\\(.*\\)\\)\n"))))
+ (put-text-property (match-beginning 0)
+ (or (match-beginning 2) (match-beginning 1))
+ 'display (propertize
+ (cond
+ ((null (match-beginning 1)) "new file ")
+ ((null (match-beginning 2)) "deleted ")
+ (t "modified
"))
+ 'face '(diff-file-header diff-header)))
+ (unless (match-beginning 2)
+ (put-text-property (match-end 1) (1- (match-end 0))
+ 'display "")))))
+ nil)
+
;;; Support for converting a diff to diff3 markers via `wiggle'.
;; Wiggle can be found at http://neil.brown.name/wiggle/ or in your nearest
diff --git a/lisp/vc/diff.el b/lisp/vc/diff.el
index b850350..ac94586 100644
--- a/lisp/vc/diff.el
+++ b/lisp/vc/diff.el
@@ -226,8 +226,9 @@ With prefix arg, prompt for diff switches."
"View the differences between BUFFER and its associated file.
This requires the external program `diff' to be in your `exec-path'."
(interactive "bBuffer: ")
- (with-current-buffer (get-buffer (or buffer (current-buffer)))
- (diff buffer-file-name (current-buffer) nil 'noasync)))
+ (let ((buf (get-buffer (or buffer (current-buffer)))))
+ (with-current-buffer (or (buffer-base-buffer buf) buf)
+ (diff buffer-file-name (current-buffer) nil 'noasync))))
(provide 'diff)
diff --git a/lisp/vc/log-edit.el b/lisp/vc/log-edit.el
index 6ff782a..90860fb 100644
--- a/lisp/vc/log-edit.el
+++ b/lisp/vc/log-edit.el
@@ -913,8 +913,10 @@ where LOGBUFFER is the name of the ChangeLog buffer, and
each
(setq change-log-default-name nil)
(find-change-log)))))
(when (or (find-buffer-visiting changelog-file-name)
- (file-exists-p changelog-file-name))
- (with-current-buffer (find-file-noselect changelog-file-name)
+ (file-exists-p changelog-file-name)
+ add-log-dont-create-changelog-file)
+ (with-current-buffer
+ (add-log-find-changelog-buffer changelog-file-name)
(unless (eq major-mode 'change-log-mode) (change-log-mode))
(goto-char (point-min))
(if (looking-at "\\s-*\n") (goto-char (match-end 0)))
diff --git a/lisp/w32-fns.el b/lisp/w32-fns.el
index 825420c..bdba32c 100644
--- a/lisp/w32-fns.el
+++ b/lisp/w32-fns.el
@@ -280,7 +280,7 @@ bit output with no translation."
(w32-add-charset-info "iso8859-13" 'w32-charset-baltic 1257)
(w32-add-charset-info "koi8-r" 'w32-charset-russian 20866)
(w32-add-charset-info "iso8859-5" 'w32-charset-russian 28595)
- (w32-add-charset-info "tis620-2533" 'w32-charset-thai 874)
+ (w32-add-charset-info "iso8859-11" 'w32-charset-thai 874)
(w32-add-charset-info "windows-1258" 'w32-charset-vietnamese 1258)
(w32-add-charset-info "ksc5601.1992" 'w32-charset-johab 1361)
(w32-add-charset-info "mac-roman" 'w32-charset-mac 10000)
diff --git a/lisp/wdired.el b/lisp/wdired.el
index bb60e77..be0bde2 100644
--- a/lisp/wdired.el
+++ b/lisp/wdired.el
@@ -255,6 +255,7 @@ See `wdired-mode'."
(setq buffer-read-only nil)
(dired-unadvertise default-directory)
(add-hook 'kill-buffer-hook 'wdired-check-kill-buffer nil t)
+ (add-hook 'after-change-functions 'wdired--restore-dired-filename-prop nil t)
(setq major-mode 'wdired-mode)
(setq mode-name "Editable Dired")
(setq revert-buffer-function 'wdired-revert)
@@ -363,6 +364,7 @@ non-nil means return old filename."
(setq mode-name "Dired")
(dired-advertise)
(remove-hook 'kill-buffer-hook 'wdired-check-kill-buffer t)
+ (remove-hook 'after-change-functions 'wdired--restore-dired-filename-prop t)
(set (make-local-variable 'revert-buffer-function) 'dired-revert))
@@ -381,7 +383,6 @@ non-nil means return old filename."
(defun wdired-finish-edit ()
"Actually rename files based on your editing in the Dired buffer."
(interactive)
- (wdired-change-to-dired-mode)
(let ((changes nil)
(errors 0)
files-deleted
@@ -423,6 +424,11 @@ non-nil means return old filename."
(forward-line -1)))
(when files-renamed
(setq errors (+ errors (wdired-do-renames files-renamed))))
+ ;; We have to be in wdired-mode when wdired-do-renames is executed
+ ;; so that wdired--restore-dired-filename-prop runs, but we have
+ ;; to change back to dired-mode before reverting the buffer to
+ ;; avoid using wdired-revert, which changes back to wdired-mode.
+ (wdired-change-to-dired-mode)
(if changes
(progn
;; If we are displaying a single file (rather than the
@@ -543,19 +549,25 @@ and proceed depending on the answer."
(goto-char (point-max))
(forward-line -1)
(let ((done nil)
+ (failed t)
curr-filename)
(while (and (not done) (not (bobp)))
(setq curr-filename (wdired-get-filename nil t))
(if (equal curr-filename filename-ori)
- (progn
- (setq done t)
- (let ((inhibit-read-only t))
- (dired-move-to-filename)
- (search-forward (wdired-get-filename t) nil t)
- (replace-match (file-name-nondirectory filename-ori) t t))
- (dired-do-create-files-regexp
- (function dired-rename-file)
- "Move" 1 ".*" filename-new nil t))
+ (unwind-protect
+ (progn
+ (setq done t)
+ (let ((inhibit-read-only t))
+ (dired-move-to-filename)
+ (search-forward (wdired-get-filename t) nil t)
+ (replace-match (file-name-nondirectory filename-ori) t t))
+ (dired-do-create-files-regexp
+ (function dired-rename-file)
+ "Move" 1 ".*" filename-new nil t)
+ (setq failed nil))
+ ;; If user types C-g when prompted to change the file
+ ;; name, make sure we return to dired-mode.
+ (when failed (wdired-change-to-dired-mode)))
(forward-line -1))))))
;; marks a list of files for deletion
@@ -586,6 +598,25 @@ Optional arguments are ignored."
(not (y-or-n-p "Buffer changed. Discard changes and kill buffer? ")))
(error "Error")))
+;; Added to after-change-functions in wdired-change-to-wdired-mode to
+;; ensure that, on editing a file name, new characters get the
+;; dired-filename text property, which allows functions that look for
+;; this property (e.g. dired-isearch-filenames) to work in wdired-mode
+;; and also avoids an error with non-nil wdired-use-interactive-rename
+;; (bug#32173).
+(defun wdired--restore-dired-filename-prop (beg end _len)
+ (save-match-data
+ (save-excursion
+ (beginning-of-line)
+ (when (re-search-forward directory-listing-before-filename-regexp
+ (line-end-position) t)
+ (setq beg (point)
+ end (if (and (file-symlink-p (dired-get-filename))
+ (search-forward " -> " (line-end-position) t))
+ (goto-char (match-beginning 0))
+ (line-end-position)))
+ (put-text-property beg end 'dired-filename t)))))
+
(defun wdired-next-line (arg)
"Move down lines then position at filename or the current column.
See `wdired-use-dired-vertical-movement'. Optional prefix ARG
diff --git a/m4/builtin-expect.m4 b/m4/builtin-expect.m4
new file mode 100644
index 0000000..a1eaf96
--- /dev/null
+++ b/m4/builtin-expect.m4
@@ -0,0 +1,49 @@
+dnl Check for __builtin_expect.
+
+dnl Copyright 2016-2018 Free Software Foundation, Inc.
+dnl This file is free software; the Free Software Foundation
+dnl gives unlimited permission to copy and/or distribute it,
+dnl with or without modifications, as long as this notice is preserved.
+
+dnl Written by Paul Eggert.
+
+AC_DEFUN([gl___BUILTIN_EXPECT],
+[
+ AC_CACHE_CHECK([for __builtin_expect],
+ [gl_cv___builtin_expect],
+ [AC_LINK_IFELSE(
+ [AC_LANG_SOURCE([[
+ int
+ main (int argc, char **argv)
+ {
+ argc = __builtin_expect (argc, 100);
+ return argv[argc != 100][0];
+ }]])],
+ [gl_cv___builtin_expect=yes],
+ [AC_LINK_IFELSE(
+ [AC_LANG_SOURCE([[
+ #include <builtins.h>
+ int
+ main (int argc, char **argv)
+ {
+ argc = __builtin_expect (argc, 100);
+ return argv[argc != 100][0];
+ }]])],
+ [gl_cv___builtin_expect="in <builtins.h>"],
+ [gl_cv___builtin_expect=no])])])
+ if test "$gl_cv___builtin_expect" = yes; then
+ AC_DEFINE([HAVE___BUILTIN_EXPECT], [1])
+ elif test "$gl_cv___builtin_expect" = "in <builtins.h>"; then
+ AC_DEFINE([HAVE___BUILTIN_EXPECT], [2])
+ fi
+ AH_VERBATIM([HAVE___BUILTIN_EXPECT],
+ [/* Define to 1 if the compiler supports __builtin_expect,
+ and to 2 if <builtins.h> does. */
+#undef HAVE___BUILTIN_EXPECT
+#ifndef HAVE___BUILTIN_EXPECT
+# define __builtin_expect(e, c) (e)
+#elif HAVE___BUILTIN_EXPECT == 2
+# include <builtins.h>
+#endif
+ ])
+])
diff --git a/m4/eealloc.m4 b/m4/eealloc.m4
new file mode 100644
index 0000000..a5a4e26
--- /dev/null
+++ b/m4/eealloc.m4
@@ -0,0 +1,31 @@
+# eealloc.m4 serial 3
+dnl Copyright (C) 2003, 2009-2018 Free Software Foundation, Inc.
+dnl This file is free software; the Free Software Foundation
+dnl gives unlimited permission to copy and/or distribute it,
+dnl with or without modifications, as long as this notice is preserved.
+
+AC_DEFUN([gl_EEALLOC],
+[
+ AC_REQUIRE([gl_EEMALLOC])
+ AC_REQUIRE([gl_EEREALLOC])
+])
+
+AC_DEFUN([gl_EEMALLOC],
+[
+ _AC_FUNC_MALLOC_IF(
+ [gl_cv_func_malloc_0_nonnull=1],
+ [gl_cv_func_malloc_0_nonnull=0])
+ AC_DEFINE_UNQUOTED([MALLOC_0_IS_NONNULL], [$gl_cv_func_malloc_0_nonnull],
+ [If malloc(0) is != NULL, define this to 1. Otherwise define this
+ to 0.])
+])
+
+AC_DEFUN([gl_EEREALLOC],
+[
+ _AC_FUNC_REALLOC_IF(
+ [gl_cv_func_realloc_0_nonnull=1],
+ [gl_cv_func_realloc_0_nonnull=0])
+ AC_DEFINE_UNQUOTED([REALLOC_0_IS_NONNULL], [$gl_cv_func_realloc_0_nonnull],
+ [If realloc(NULL,0) is != NULL, define this to 1. Otherwise define this
+ to 0.])
+])
diff --git a/m4/extern-inline.m4 b/m4/extern-inline.m4
index da8a2cc..3661cbd 100644
--- a/m4/extern-inline.m4
+++ b/m4/extern-inline.m4
@@ -25,7 +25,8 @@ AC_DEFUN([gl_EXTERN_INLINE],
if isdigit is mistakenly implemented via a static inline function,
a program containing an extern inline function that calls isdigit
may not work since the C standard prohibits extern inline functions
- from calling static functions. This bug is known to occur on:
+ from calling static functions (ISO C 99 section 6.7.4.(3).
+ This bug is known to occur on:
OS X 10.8 and earlier; see:
https://lists.gnu.org/r/bug-gnulib/2012-12/msg00023.html
@@ -38,7 +39,18 @@ AC_DEFUN([gl_EXTERN_INLINE],
OS X 10.9 has a macro __header_inline indicating the bug is fixed for C and
for clang but remains for g++; see <https://trac.macports.org/ticket/41033>.
- Assume DragonFly and FreeBSD will be similar. */
+ Assume DragonFly and FreeBSD will be similar.
+
+ GCC 4.3 and above with -std=c99 or -std=gnu99 implements ISO C99
+ inline semantics, unless -fgnu89-inline is used. It defines a macro
+ __GNUC_STDC_INLINE__ to indicate this situation or a macro
+ __GNUC_GNU_INLINE__ to indicate the opposite situation.
+ GCC 4.2 with -std=c99 or -std=gnu99 implements the GNU C inline
+ semantics but warns, unless -fgnu89-inline is used:
+ warning: C99 inline functions are not supported; using GNU89
+ warning: to disable this warning use -fgnu89-inline or the gnu_inline
function attribute
+ It defines a macro __GNUC_GNU_INLINE__ to indicate this situation.
+ */
#if (((defined __APPLE__ && defined __MACH__) \
|| defined __DragonFly__ || defined __FreeBSD__) \
&& (defined __header_inline \
diff --git a/m4/glibc21.m4 b/m4/glibc21.m4
new file mode 100644
index 0000000..126aa1a
--- /dev/null
+++ b/m4/glibc21.m4
@@ -0,0 +1,34 @@
+# glibc21.m4 serial 5
+dnl Copyright (C) 2000-2002, 2004, 2008, 2010-2018 Free Software Foundation,
+dnl Inc.
+dnl This file is free software; the Free Software Foundation
+dnl gives unlimited permission to copy and/or distribute it,
+dnl with or without modifications, as long as this notice is preserved.
+
+# Test for the GNU C Library, version 2.1 or newer, or uClibc.
+# From Bruno Haible.
+
+AC_DEFUN([gl_GLIBC21],
+ [
+ AC_CACHE_CHECK([whether we are using the GNU C Library >= 2.1 or uClibc],
+ [ac_cv_gnu_library_2_1],
+ [AC_EGREP_CPP([Lucky],
+ [
+#include <features.h>
+#ifdef __GNU_LIBRARY__
+ #if (__GLIBC__ == 2 && __GLIBC_MINOR__ >= 1) || (__GLIBC__ > 2)
+ Lucky GNU user
+ #endif
+#endif
+#ifdef __UCLIBC__
+ Lucky user
+#endif
+ ],
+ [ac_cv_gnu_library_2_1=yes],
+ [ac_cv_gnu_library_2_1=no])
+ ]
+ )
+ AC_SUBST([GLIBC21])
+ GLIBC21="$ac_cv_gnu_library_2_1"
+ ]
+)
diff --git a/m4/gnulib-comp.m4 b/m4/gnulib-comp.m4
index a6e3be3..61aabaa 100644
--- a/m4/gnulib-comp.m4
+++ b/m4/gnulib-comp.m4
@@ -48,6 +48,7 @@ AC_DEFUN([gl_EARLY],
# Code from module allocator:
# Code from module at-internal:
# Code from module binary-io:
+ # Code from module builtin-expect:
# Code from module byteswap:
# Code from module c-ctype:
# Code from module c-strcase:
@@ -101,6 +102,7 @@ AC_DEFUN([gl_EARLY],
# Code from module gettimeofday:
# Code from module gitlog-to-changelog:
# Code from module group-member:
+ # Code from module ieee754-h:
# Code from module ignore-value:
# Code from module include_next:
# Code from module intprops:
@@ -128,6 +130,7 @@ AC_DEFUN([gl_EARLY],
# Code from module qcopy-acl:
# Code from module readlink:
# Code from module readlinkat:
+ # Code from module regex:
# Code from module root-uid:
# Code from module sig2str:
# Code from module signal-h:
@@ -295,6 +298,7 @@ AC_DEFUN([gl_INIT],
gl_PREREQ_GETTIMEOFDAY
fi
gl_SYS_TIME_MODULE_INDICATOR([gettimeofday])
+ gl_IEEE754_H
gl_INTTYPES_INCOMPLETE
AC_REQUIRE([gl_LARGEFILE])
gl_LIMITS_H
@@ -356,6 +360,11 @@ AC_DEFUN([gl_INIT],
AC_LIBOBJ([readlinkat])
fi
gl_UNISTD_MODULE_INDICATOR([readlinkat])
+ gl_REGEX
+ if test $ac_use_included_regex = yes; then
+ AC_LIBOBJ([regex])
+ gl_PREREQ_REGEX
+ fi
gl_FUNC_SIG2STR
if test $ac_cv_func_sig2str = no; then
AC_LIBOBJ([sig2str])
@@ -423,6 +432,7 @@ AC_DEFUN([gl_INIT],
gl_UTIMENS
AC_C_VARARRAYS
gl_gnulib_enabled_260941c0e5dc67ec9e87d1fb321c300b=false
+ gl_gnulib_enabled_37f71b604aa9c54446783d80f42fe547=false
gl_gnulib_enabled_cloexec=false
gl_gnulib_enabled_dirfd=false
gl_gnulib_enabled_dosname=false
@@ -446,6 +456,13 @@ AC_DEFUN([gl_INIT],
func_gl_gnulib_m4code_open
fi
}
+ func_gl_gnulib_m4code_37f71b604aa9c54446783d80f42fe547 ()
+ {
+ if ! $gl_gnulib_enabled_37f71b604aa9c54446783d80f42fe547; then
+ gl___BUILTIN_EXPECT
+ gl_gnulib_enabled_37f71b604aa9c54446783d80f42fe547=true
+ fi
+ }
func_gl_gnulib_m4code_cloexec ()
{
if ! $gl_gnulib_enabled_cloexec; then
@@ -649,6 +666,9 @@ AC_DEFUN([gl_INIT],
if test $HAVE_READLINKAT = 0; then
func_gl_gnulib_m4code_03e0aaad4cb89ca757653bd367a6ccb7
fi
+ if test $ac_use_included_regex = yes; then
+ func_gl_gnulib_m4code_37f71b604aa9c54446783d80f42fe547
+ fi
if { test $HAVE_DECL_STRTOIMAX = 0 || test $REPLACE_STRTOIMAX = 1; } && test
$ac_cv_type_long_long_int = yes; then
func_gl_gnulib_m4code_strtoll
fi
@@ -657,6 +677,7 @@ AC_DEFUN([gl_INIT],
fi
m4_pattern_allow([^gl_GNULIB_ENABLED_])
AM_CONDITIONAL([gl_GNULIB_ENABLED_260941c0e5dc67ec9e87d1fb321c300b],
[$gl_gnulib_enabled_260941c0e5dc67ec9e87d1fb321c300b])
+ AM_CONDITIONAL([gl_GNULIB_ENABLED_37f71b604aa9c54446783d80f42fe547],
[$gl_gnulib_enabled_37f71b604aa9c54446783d80f42fe547])
AM_CONDITIONAL([gl_GNULIB_ENABLED_cloexec], [$gl_gnulib_enabled_cloexec])
AM_CONDITIONAL([gl_GNULIB_ENABLED_dirfd], [$gl_gnulib_enabled_dirfd])
AM_CONDITIONAL([gl_GNULIB_ENABLED_dosname], [$gl_gnulib_enabled_dosname])
@@ -895,6 +916,7 @@ AC_DEFUN([gl_FILE_LIST], [
lib/gettimeofday.c
lib/gl_openssl.h
lib/group-member.c
+ lib/ieee754.in.h
lib/ignore-value.h
lib/intprops.h
lib/inttypes.in.h
@@ -921,6 +943,12 @@ AC_DEFUN([gl_FILE_LIST], [
lib/qcopy-acl.c
lib/readlink.c
lib/readlinkat.c
+ lib/regcomp.c
+ lib/regex.c
+ lib/regex.h
+ lib/regex_internal.c
+ lib/regex_internal.h
+ lib/regexec.c
lib/root-uid.h
lib/set-permissions.c
lib/sha1.c
@@ -977,6 +1005,7 @@ AC_DEFUN([gl_FILE_LIST], [
m4/absolute-header.m4
m4/acl.m4
m4/alloca.m4
+ m4/builtin-expect.m4
m4/byteswap.m4
m4/c-strtod.m4
m4/clock_time.m4
@@ -988,6 +1017,7 @@ AC_DEFUN([gl_FILE_LIST], [
m4/dirent_h.m4
m4/dirfd.m4
m4/dup2.m4
+ m4/eealloc.m4
m4/environ.m4
m4/errno_h.m4
m4/euidaccess.m4
@@ -1015,8 +1045,10 @@ AC_DEFUN([gl_FILE_LIST], [
m4/gettime.m4
m4/gettimeofday.m4
m4/gl-openssl.m4
+ m4/glibc21.m4
m4/gnulib-common.m4
m4/group-member.m4
+ m4/ieee754-h.m4
m4/include_next.m4
m4/inttypes.m4
m4/largefile.m4
@@ -1026,6 +1058,7 @@ AC_DEFUN([gl_FILE_LIST], [
m4/lstat.m4
m4/manywarnings-c++.m4
m4/manywarnings.m4
+ m4/mbstate_t.m4
m4/md5.m4
m4/memrchr.m4
m4/minmax.m4
@@ -1044,6 +1077,7 @@ AC_DEFUN([gl_FILE_LIST], [
m4/putenv.m4
m4/readlink.m4
m4/readlinkat.m4
+ m4/regex.m4
m4/sha1.m4
m4/sha256.m4
m4/sha512.m4
diff --git a/m4/ieee754-h.m4 b/m4/ieee754-h.m4
new file mode 100644
index 0000000..bf7c332
--- /dev/null
+++ b/m4/ieee754-h.m4
@@ -0,0 +1,21 @@
+# Configure ieee754-h module
+
+dnl Copyright 2018 Free Software Foundation, Inc.
+dnl This file is free software; the Free Software Foundation
+dnl gives unlimited permission to copy and/or distribute it,
+dnl with or without modifications, as long as this notice is preserved.
+
+AC_DEFUN([gl_IEEE754_H],
+[
+ AC_REQUIRE([AC_C_BIGENDIAN])
+ AC_CHECK_HEADERS_ONCE([ieee754.h])
+ if test $ac_cv_header_ieee754_h = yes; then
+ IEEE754_H=
+ else
+ IEEE754_H=ieee754.h
+ AC_DEFINE([_GL_REPLACE_IEEE754_H], 1,
+ [Define to 1 if <ieee754.h> is missing.])
+ fi
+ AC_SUBST([IEEE754_H])
+ AM_CONDITIONAL([GL_GENERATE_IEEE754_H], [test -n "$IEEE754_H"])
+])
diff --git a/m4/mbstate_t.m4 b/m4/mbstate_t.m4
new file mode 100644
index 0000000..004aa0d
--- /dev/null
+++ b/m4/mbstate_t.m4
@@ -0,0 +1,41 @@
+# mbstate_t.m4 serial 13
+dnl Copyright (C) 2000-2002, 2008-2018 Free Software Foundation, Inc.
+dnl This file is free software; the Free Software Foundation
+dnl gives unlimited permission to copy and/or distribute it,
+dnl with or without modifications, as long as this notice is preserved.
+
+# From Paul Eggert.
+
+# BeOS 5 has <wchar.h> but does not define mbstate_t,
+# so you can't declare an object of that type.
+# Check for this incompatibility with Standard C.
+
+# AC_TYPE_MBSTATE_T
+# -----------------
+AC_DEFUN([AC_TYPE_MBSTATE_T],
+[
+ AC_REQUIRE([AC_USE_SYSTEM_EXTENSIONS]) dnl for HP-UX 11.11
+
+ AC_CACHE_CHECK([for mbstate_t], [ac_cv_type_mbstate_t],
+ [AC_COMPILE_IFELSE(
+ [AC_LANG_PROGRAM(
+ [AC_INCLUDES_DEFAULT[
+/* Tru64 with Desktop Toolkit C has a bug: <stdio.h> must be included before
+ <wchar.h>.
+ BSD/OS 4.0.1 has a bug: <stddef.h>, <stdio.h> and <time.h> must be
+ included before <wchar.h>. */
+#include <stddef.h>
+#include <stdio.h>
+#include <time.h>
+#include <wchar.h>]],
+ [[mbstate_t x; return sizeof x;]])],
+ [ac_cv_type_mbstate_t=yes],
+ [ac_cv_type_mbstate_t=no])])
+ if test $ac_cv_type_mbstate_t = yes; then
+ AC_DEFINE([HAVE_MBSTATE_T], [1],
+ [Define to 1 if <wchar.h> declares mbstate_t.])
+ else
+ AC_DEFINE([mbstate_t], [int],
+ [Define to a type if <wchar.h> does not define.])
+ fi
+])
diff --git a/m4/regex.m4 b/m4/regex.m4
new file mode 100644
index 0000000..055d71b
--- /dev/null
+++ b/m4/regex.m4
@@ -0,0 +1,300 @@
+# serial 67
+
+# Copyright (C) 1996-2001, 2003-2018 Free Software Foundation, Inc.
+#
+# This file is free software; the Free Software Foundation
+# gives unlimited permission to copy and/or distribute it,
+# with or without modifications, as long as this notice is preserved.
+
+dnl Initially derived from code in GNU grep.
+dnl Mostly written by Jim Meyering.
+
+AC_PREREQ([2.50])
+
+AC_DEFUN([gl_REGEX],
+[
+ AC_REQUIRE([AC_CANONICAL_HOST]) dnl for cross-compiles
+ AC_ARG_WITH([included-regex],
+ [AS_HELP_STRING([--without-included-regex],
+ [don't compile regex; this is the default on systems
+ with recent-enough versions of the GNU C Library
+ (use with caution on other systems).])])
+
+ case $with_included_regex in #(
+ yes|no) ac_use_included_regex=$with_included_regex
+ ;;
+ '')
+ # If the system regex support is good enough that it passes the
+ # following run test, then default to *not* using the included regex.c.
+ # If cross compiling, assume the test would fail and use the included
+ # regex.c.
+ AC_CHECK_DECLS_ONCE([alarm])
+ AC_CHECK_HEADERS_ONCE([malloc.h])
+ AC_CACHE_CHECK([for working re_compile_pattern],
+ [gl_cv_func_re_compile_pattern_working],
+ [AC_RUN_IFELSE(
+ [AC_LANG_PROGRAM(
+ [[#include <regex.h>
+
+ #include <locale.h>
+ #include <limits.h>
+ #include <string.h>
+
+ #if defined M_CHECK_ACTION || HAVE_DECL_ALARM
+ # include <signal.h>
+ # include <unistd.h>
+ #endif
+
+ #if HAVE_MALLOC_H
+ # include <malloc.h>
+ #endif
+
+ #ifdef M_CHECK_ACTION
+ /* Exit with distinguishable exit code. */
+ static void sigabrt_no_core (int sig) { raise (SIGTERM); }
+ #endif
+ ]],
+ [[int result = 0;
+ static struct re_pattern_buffer regex;
+ unsigned char folded_chars[UCHAR_MAX + 1];
+ int i;
+ const char *s;
+ struct re_registers regs;
+
+ /* Some builds of glibc go into an infinite loop on this
+ test. Use alarm to force death, and mallopt to avoid
+ malloc recursion in diagnosing the corrupted heap. */
+#if HAVE_DECL_ALARM
+ signal (SIGALRM, SIG_DFL);
+ alarm (2);
+#endif
+#ifdef M_CHECK_ACTION
+ signal (SIGABRT, sigabrt_no_core);
+ mallopt (M_CHECK_ACTION, 2);
+#endif
+
+ if (setlocale (LC_ALL, "en_US.UTF-8"))
+ {
+ {
+ /*
https://sourceware.org/ml/libc-hacker/2006-09/msg00008.html
+ This test needs valgrind to catch the bug on Debian
+ GNU/Linux 3.1 x86, but it might catch the bug better
+ on other platforms and it shouldn't hurt to try the
+ test here. */
+ static char const pat[] = "insert into";
+ static char const data[] =
+ "\xFF\0\x12\xA2\xAA\xC4\xB1,K\x12\xC4\xB1*\xACK";
+ re_set_syntax (RE_SYNTAX_GREP | RE_HAT_LISTS_NOT_NEWLINE
+ | RE_ICASE);
+ memset (®ex, 0, sizeof regex);
+ s = re_compile_pattern (pat, sizeof pat - 1, ®ex);
+ if (s)
+ result |= 1;
+ else if (re_search (®ex, data, sizeof data - 1,
+ 0, sizeof data - 1, ®s)
+ != -1)
+ result |= 1;
+ regfree (®ex);
+ }
+
+ {
+ /* This test is from glibc bug 15078.
+ The test case is from Andreas Schwab in
+
<https://sourceware.org/ml/libc-alpha/2013-01/msg00967.html>.
+ */
+ static char const pat[] = "[^x]x";
+ static char const data[] =
+ /*
<U1000><U103B><U103D><U1014><U103A><U102F><U1015><U103A> */
+ "\xe1\x80\x80"
+ "\xe1\x80\xbb"
+ "\xe1\x80\xbd"
+ "\xe1\x80\x94"
+ "\xe1\x80\xba"
+ "\xe1\x80\xaf"
+ "\xe1\x80\x95"
+ "\xe1\x80\xba"
+ "x";
+ re_set_syntax (0);
+ memset (®ex, 0, sizeof regex);
+ s = re_compile_pattern (pat, sizeof pat - 1, ®ex);
+ if (s)
+ result |= 1;
+ else
+ {
+ i = re_search (®ex, data, sizeof data - 1,
+ 0, sizeof data - 1, 0);
+ if (i != 0 && i != 21)
+ result |= 1;
+ }
+ regfree (®ex);
+ }
+
+ if (! setlocale (LC_ALL, "C"))
+ return 1;
+ }
+
+ /* This test is from glibc bug 3957, reported by Andrew Mackey. */
+ re_set_syntax (RE_SYNTAX_EGREP | RE_HAT_LISTS_NOT_NEWLINE);
+ memset (®ex, 0, sizeof regex);
+ s = re_compile_pattern ("a[^x]b", 6, ®ex);
+ if (s)
+ result |= 2;
+ /* This should fail, but succeeds for glibc-2.5. */
+ else if (re_search (®ex, "a\nb", 3, 0, 3, ®s) != -1)
+ result |= 2;
+
+ /* This regular expression is from Spencer ere test number 75
+ in grep-2.3. */
+ re_set_syntax (RE_SYNTAX_POSIX_EGREP);
+ memset (®ex, 0, sizeof regex);
+ for (i = 0; i <= UCHAR_MAX; i++)
+ folded_chars[i] = i;
+ regex.translate = folded_chars;
+ s = re_compile_pattern ("a[[:@:>@:]]b\n", 11, ®ex);
+ /* This should fail with _Invalid character class name_ error. */
+ if (!s)
+ result |= 4;
+
+ /* Ensure that [b-a] is diagnosed as invalid, when
+ using RE_NO_EMPTY_RANGES. */
+ re_set_syntax (RE_SYNTAX_POSIX_EGREP | RE_NO_EMPTY_RANGES);
+ memset (®ex, 0, sizeof regex);
+ s = re_compile_pattern ("a[b-a]", 6, ®ex);
+ if (s == 0)
+ result |= 8;
+
+ /* This should succeed, but does not for glibc-2.1.3. */
+ memset (®ex, 0, sizeof regex);
+ s = re_compile_pattern ("{1", 2, ®ex);
+ if (s)
+ result |= 8;
+
+ /* The following example is derived from a problem report
+ against gawk from Jorge Stolfi <address@hidden>. */
+ memset (®ex, 0, sizeof regex);
+ s = re_compile_pattern ("[an\371]*n", 7, ®ex);
+ if (s)
+ result |= 8;
+ /* This should match, but does not for glibc-2.2.1. */
+ else if (re_match (®ex, "an", 2, 0, ®s) != 2)
+ result |= 8;
+
+ memset (®ex, 0, sizeof regex);
+ s = re_compile_pattern ("x", 1, ®ex);
+ if (s)
+ result |= 8;
+ /* glibc-2.2.93 does not work with a negative RANGE argument. */
+ else if (re_search (®ex, "wxy", 3, 2, -2, ®s) != 1)
+ result |= 8;
+
+ /* The version of regex.c in older versions of gnulib
+ ignored RE_ICASE. Detect that problem too. */
+ re_set_syntax (RE_SYNTAX_EMACS | RE_ICASE);
+ memset (®ex, 0, sizeof regex);
+ s = re_compile_pattern ("x", 1, ®ex);
+ if (s)
+ result |= 16;
+ else if (re_search (®ex, "WXY", 3, 0, 3, ®s) < 0)
+ result |= 16;
+
+ /* Catch a bug reported by Vin Shelton in
+ https://lists.gnu.org/r/bug-coreutils/2007-06/msg00089.html
+ */
+ re_set_syntax (RE_SYNTAX_POSIX_BASIC
+ & ~RE_CONTEXT_INVALID_DUP
+ & ~RE_NO_EMPTY_RANGES);
+ memset (®ex, 0, sizeof regex);
+ s = re_compile_pattern ("[[:alnum:]_-]\\\\+$", 16, ®ex);
+ if (s)
+ result |= 32;
+
+ /* REG_STARTEND was added to glibc on 2004-01-15.
+ Reject older versions. */
+ if (! REG_STARTEND)
+ result |= 64;
+
+#if 0
+ /* It would be nice to reject hosts whose regoff_t values are too
+ narrow (including glibc on hosts with 64-bit ptrdiff_t and
+ 32-bit int), but we should wait until glibc implements this
+ feature. Otherwise, support for equivalence classes and
+ multibyte collation symbols would always be broken except
+ when compiling --without-included-regex. */
+ if (sizeof (regoff_t) < sizeof (ptrdiff_t)
+ || sizeof (regoff_t) < sizeof (ssize_t))
+ result |= 64;
+#endif
+
+ return result;
+ ]])],
+ [gl_cv_func_re_compile_pattern_working=yes],
+ [gl_cv_func_re_compile_pattern_working=no],
+ [case "$host_os" in
+ # Guess no on native Windows.
+ mingw*) gl_cv_func_re_compile_pattern_working="guessing no" ;;
+ # Otherwise, assume it is not working.
+ *) gl_cv_func_re_compile_pattern_working="guessing no" ;;
+ esac
+ ])
+ ])
+ case "$gl_cv_func_re_compile_pattern_working" in #(
+ *yes) ac_use_included_regex=no;; #(
+ *no) ac_use_included_regex=yes;;
+ esac
+ ;;
+ *) AC_MSG_ERROR([Invalid value for --with-included-regex:
$with_included_regex])
+ ;;
+ esac
+
+ if test $ac_use_included_regex = yes; then
+ AC_DEFINE([_REGEX_INCLUDE_LIMITS_H], [1],
+ [Define if you want <regex.h> to include <limits.h>, so that it
+ consistently overrides <limits.h>'s RE_DUP_MAX.])
+ AC_DEFINE([_REGEX_LARGE_OFFSETS], [1],
+ [Define if you want regoff_t to be at least as wide POSIX requires.])
+ AC_DEFINE([re_syntax_options], [rpl_re_syntax_options],
+ [Define to rpl_re_syntax_options if the replacement should be used.])
+ AC_DEFINE([re_set_syntax], [rpl_re_set_syntax],
+ [Define to rpl_re_set_syntax if the replacement should be used.])
+ AC_DEFINE([re_compile_pattern], [rpl_re_compile_pattern],
+ [Define to rpl_re_compile_pattern if the replacement should be used.])
+ AC_DEFINE([re_compile_fastmap], [rpl_re_compile_fastmap],
+ [Define to rpl_re_compile_fastmap if the replacement should be used.])
+ AC_DEFINE([re_search], [rpl_re_search],
+ [Define to rpl_re_search if the replacement should be used.])
+ AC_DEFINE([re_search_2], [rpl_re_search_2],
+ [Define to rpl_re_search_2 if the replacement should be used.])
+ AC_DEFINE([re_match], [rpl_re_match],
+ [Define to rpl_re_match if the replacement should be used.])
+ AC_DEFINE([re_match_2], [rpl_re_match_2],
+ [Define to rpl_re_match_2 if the replacement should be used.])
+ AC_DEFINE([re_set_registers], [rpl_re_set_registers],
+ [Define to rpl_re_set_registers if the replacement should be used.])
+ AC_DEFINE([re_comp], [rpl_re_comp],
+ [Define to rpl_re_comp if the replacement should be used.])
+ AC_DEFINE([re_exec], [rpl_re_exec],
+ [Define to rpl_re_exec if the replacement should be used.])
+ AC_DEFINE([regcomp], [rpl_regcomp],
+ [Define to rpl_regcomp if the replacement should be used.])
+ AC_DEFINE([regexec], [rpl_regexec],
+ [Define to rpl_regexec if the replacement should be used.])
+ AC_DEFINE([regerror], [rpl_regerror],
+ [Define to rpl_regerror if the replacement should be used.])
+ AC_DEFINE([regfree], [rpl_regfree],
+ [Define to rpl_regfree if the replacement should be used.])
+ fi
+])
+
+# Prerequisites of lib/regex.c and lib/regex_internal.c.
+AC_DEFUN([gl_PREREQ_REGEX],
+[
+ AC_REQUIRE([AC_USE_SYSTEM_EXTENSIONS])
+ AC_REQUIRE([AC_C_INLINE])
+ AC_REQUIRE([AC_C_RESTRICT])
+ AC_REQUIRE([AC_TYPE_MBSTATE_T])
+ AC_REQUIRE([gl_EEMALLOC])
+ AC_REQUIRE([gl_GLIBC21])
+ AC_CHECK_HEADERS([libintl.h])
+ AC_CHECK_FUNCS_ONCE([isblank iswctype])
+ AC_CHECK_DECLS([isblank], [], [], [[#include <ctype.h>]])
+])
diff --git a/nt/inc/ms-w32.h b/nt/inc/ms-w32.h
index ff43178..e4dec04 100644
--- a/nt/inc/ms-w32.h
+++ b/nt/inc/ms-w32.h
@@ -34,6 +34,11 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
# ifdef __MINGW64_VERSION_MAJOR
# define MINGW_W64
# endif
+# if defined __MINGW32_VERSION && __MINGW32_VERSION >= 5001000L
+/* Avoid warnings about gettimeofday being deprecated. */
+# undef __POSIX_2008_DEPRECATED
+# define __POSIX_2008_DEPRECATED
+# endif
#endif
/* #undef const */
@@ -455,7 +460,12 @@ extern char *get_emacs_configuration_options (void);
windows.h. For this to have proper effect, config.h must always be
included before windows.h. */
#define _WINSOCKAPI_ 1
-#define _WINSOCK_H
+#if !(defined __MINGW32_VERSION && __MINGW32_VERSION >= 5000002L)
+/* mingw.org's MinGW 5.x changed how it includes winsock.h and time.h,
+ and now defining _WINSOCK_H skips the definition of struct timeval,
+ which we don't want. */
+# define _WINSOCK_H
+#endif
/* Defines size_t and alloca (). */
#include <stdlib.h>
diff --git a/nt/inc/sys/socket.h b/nt/inc/sys/socket.h
index 2582cbd..de282c4 100644
--- a/nt/inc/sys/socket.h
+++ b/nt/inc/sys/socket.h
@@ -49,6 +49,11 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#define timeval ws_timeval
#endif
+#if defined __MINGW32_VERSION && __MINGW32_VERSION >= 5000002L
+/* Need winerror.h before winsock2.h with mingw.org's MinGW 5.x,
+ otherwise some error codes are not defined. */
+# include <winerror.h>
+#endif
#include <winsock2.h>
#include <ws2tcpip.h>
/* process.c uses uint16_t (from C99) for IPv6, but
diff --git a/src/Makefile.in b/src/Makefile.in
index 05d24ac..52ce760 100644
--- a/src/Makefile.in
+++ b/src/Makefile.in
@@ -394,7 +394,7 @@ base_obj = dispnew.o frame.o scroll.o xdisp.o menu.o
$(XMENU_OBJ) window.o \
emacs.o keyboard.o macros.o keymap.o sysdep.o \
buffer.o filelock.o insdel.o marker.o \
minibuf.o fileio.o dired.o \
- cmds.o casetab.o casefiddle.o indent.o search.o regex.o undo.o \
+ cmds.o casetab.o casefiddle.o indent.o search.o regex-emacs.o undo.o \
alloc.o data.o doc.o editfns.o callint.o \
eval.o floatfns.o fns.o font.o print.o lread.o $(MODULES_OBJ) \
syntax.o $(UNEXEC_OBJ) bytecode.o \
diff --git a/src/alloc.c b/src/alloc.c
index a8bc55b..c3e02c2 100644
--- a/src/alloc.c
+++ b/src/alloc.c
@@ -2932,7 +2932,7 @@ set_next_vector (struct Lisp_Vector *v, struct
Lisp_Vector *p)
for the most common cases; it's not required to be a power of two, but
it's expected to be a mult-of-ROUNDUP_SIZE (see below). */
-#define VECTOR_BLOCK_SIZE 4096
+enum { VECTOR_BLOCK_SIZE = 4096 };
/* Vector size requests are a multiple of this. */
enum { roundup_size = COMMON_MULTIPLE (LISP_ALIGNMENT, word_size) };
@@ -2948,22 +2948,21 @@ verify (VECTOR_BLOCK_SIZE <= (1 <<
PSEUDOVECTOR_SIZE_BITS));
/* Rounding helps to maintain alignment constraints if USE_LSB_TAG. */
-#define VECTOR_BLOCK_BYTES (VECTOR_BLOCK_SIZE - vroundup_ct (sizeof (void *)))
+enum {VECTOR_BLOCK_BYTES = VECTOR_BLOCK_SIZE - vroundup_ct (sizeof (void *))};
/* Size of the minimal vector allocated from block. */
-#define VBLOCK_BYTES_MIN vroundup_ct (header_size + sizeof (Lisp_Object))
+enum { VBLOCK_BYTES_MIN = vroundup_ct (header_size + sizeof (Lisp_Object)) };
/* Size of the largest vector allocated from block. */
-#define VBLOCK_BYTES_MAX \
- vroundup ((VECTOR_BLOCK_BYTES / 2) - word_size)
+enum { VBLOCK_BYTES_MAX = vroundup_ct ((VECTOR_BLOCK_BYTES / 2) - word_size) };
/* We maintain one free list for each possible block-allocated
vector size, and this is the number of free lists we have. */
-#define VECTOR_MAX_FREE_LIST_INDEX \
- ((VECTOR_BLOCK_BYTES - VBLOCK_BYTES_MIN) / roundup_size + 1)
+enum { VECTOR_MAX_FREE_LIST_INDEX =
+ (VECTOR_BLOCK_BYTES - VBLOCK_BYTES_MIN) / roundup_size + 1 };
/* Common shortcut to advance vector pointer over a block data. */
@@ -3090,14 +3089,14 @@ init_vectors (void)
/* Allocate vector from a vector block. */
static struct Lisp_Vector *
-allocate_vector_from_block (size_t nbytes)
+allocate_vector_from_block (ptrdiff_t nbytes)
{
struct Lisp_Vector *vector;
struct vector_block *block;
size_t index, restbytes;
- eassert (VBLOCK_BYTES_MIN <= nbytes && nbytes <= VBLOCK_BYTES_MAX);
- eassert (nbytes % roundup_size == 0);
+ eassume (VBLOCK_BYTES_MIN <= nbytes && nbytes <= VBLOCK_BYTES_MAX);
+ eassume (nbytes % roundup_size == 0);
/* First, try to allocate from a free list
containing vectors of the requested size. */
@@ -3182,35 +3181,45 @@ vector_nbytes (struct Lisp_Vector *v)
return vroundup (header_size + word_size * nwords);
}
+/* Convert a pseudovector pointer P to its underlying struct T pointer.
+ Verify that the struct is small, since cleanup_vector is called
+ only on small vector-like objects. */
+
+#define PSEUDOVEC_STRUCT(p, t) \
+ verify_expr ((header_size + VECSIZE (struct t) * word_size \
+ <= VBLOCK_BYTES_MAX), \
+ (struct t *) (p))
+
/* Release extra resources still in use by VECTOR, which may be any
- vector-like object. */
+ small vector-like object. */
static void
cleanup_vector (struct Lisp_Vector *vector)
{
detect_suspicious_free (vector);
- if (PSEUDOVECTOR_TYPEP (&vector->header, PVEC_FONT)
- && ((vector->header.size & PSEUDOVECTOR_SIZE_MASK)
- == FONT_OBJECT_MAX))
+ if (PSEUDOVECTOR_TYPEP (&vector->header, PVEC_FONT))
{
- struct font_driver const *drv = ((struct font *) vector)->driver;
-
- /* The font driver might sometimes be NULL, e.g. if Emacs was
- interrupted before it had time to set it up. */
- if (drv)
+ if ((vector->header.size & PSEUDOVECTOR_SIZE_MASK) == FONT_OBJECT_MAX)
{
- /* Attempt to catch subtle bugs like Bug#16140. */
- eassert (valid_font_driver (drv));
- drv->close ((struct font *) vector);
+ struct font *font = PSEUDOVEC_STRUCT (vector, font);
+ struct font_driver const *drv = font->driver;
+
+ /* The font driver might sometimes be NULL, e.g. if Emacs was
+ interrupted before it had time to set it up. */
+ if (drv)
+ {
+ /* Attempt to catch subtle bugs like Bug#16140. */
+ eassert (valid_font_driver (drv));
+ drv->close (font);
+ }
}
}
-
- if (PSEUDOVECTOR_TYPEP (&vector->header, PVEC_THREAD))
- finalize_one_thread ((struct thread_state *) vector);
+ else if (PSEUDOVECTOR_TYPEP (&vector->header, PVEC_THREAD))
+ finalize_one_thread (PSEUDOVEC_STRUCT (vector, thread_state));
else if (PSEUDOVECTOR_TYPEP (&vector->header, PVEC_MUTEX))
- finalize_one_mutex ((struct Lisp_Mutex *) vector);
+ finalize_one_mutex (PSEUDOVEC_STRUCT (vector, Lisp_Mutex));
else if (PSEUDOVECTOR_TYPEP (&vector->header, PVEC_CONDVAR))
- finalize_one_condvar ((struct Lisp_CondVar *) vector);
+ finalize_one_condvar (PSEUDOVEC_STRUCT (vector, Lisp_CondVar));
}
/* Reclaim space used by unmarked vectors. */
@@ -6719,7 +6728,7 @@ mark_object (Lisp_Object arg)
CHECK_ALLOCATED_AND_LIVE (live_cons_p);
CONS_MARK (ptr);
/* If the cdr is nil, avoid recursion for the car. */
- if (EQ (ptr->u.s.u.cdr, Qnil))
+ if (NILP (ptr->u.s.u.cdr))
{
obj = ptr->u.s.car;
cdr_count = 0;
diff --git a/src/callint.c b/src/callint.c
index c8b7585..81efb26 100644
--- a/src/callint.c
+++ b/src/callint.c
@@ -200,7 +200,7 @@ fix_command (Lisp_Object input, Lisp_Object values)
carelt = XCAR (elt);
/* If it is (if X Y), look at Y. */
if (EQ (carelt, Qif)
- && EQ (Fnthcdr (make_fixnum (3), elt), Qnil))
+ && NILP (Fnthcdr (make_fixnum (3), elt)))
elt = Fnth (make_fixnum (2), elt);
/* If it is (when ... Y), look at Y. */
else if (EQ (carelt, Qwhen))
diff --git a/src/casetab.c b/src/casetab.c
index 6b1c64f..0e742de 100644
--- a/src/casetab.c
+++ b/src/casetab.c
@@ -144,7 +144,8 @@ set_case_table (Lisp_Object table, bool standard)
set_char_table_extras (table, 2, eqv);
}
- /* This is so set_image_of_range_1 in regex.c can find the EQV table. */
+ /* This is so set_image_of_range_1 in regex-emacs.c can find the EQV
+ table. */
set_char_table_extras (canon, 2, eqv);
if (standard)
diff --git a/src/character.c b/src/character.c
index 97698d7..851e61e 100644
--- a/src/character.c
+++ b/src/character.c
@@ -34,6 +34,7 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#include "lisp.h"
#include "character.h"
#include "buffer.h"
+#include "dispextern.h"
#include "composite.h"
#include "disptab.h"
@@ -288,10 +289,15 @@ char_width (int c, struct Lisp_Char_Table *dp)
if (VECTORP (disp))
for (i = 0, width = 0; i < ASIZE (disp); i++)
{
+ int c = -1;
ch = AREF (disp, i);
- if (CHARACTERP (ch))
+ if (GLYPH_CODE_P (ch))
+ c = GLYPH_CODE_CHAR (ch);
+ else if (CHARACTERP (ch))
+ c = XFASTINT (ch);
+ if (c >= 0)
{
- int w = CHARACTER_WIDTH (XFIXNAT (ch));
+ int w = CHARACTER_WIDTH (c);
if (INT_ADD_WRAPV (width, w, &width))
string_overflow ();
}
diff --git a/src/chartab.c b/src/chartab.c
index 0383a84..3d38b3c 100644
--- a/src/chartab.c
+++ b/src/chartab.c
@@ -605,7 +605,7 @@ a cons of character codes (for characters in the range), or
a character code. *
Lisp_Object val;
CHECK_CHAR_TABLE (char_table);
- if (EQ (range, Qnil))
+ if (NILP (range))
val = XCHAR_TABLE (char_table)->defalt;
else if (CHARACTERP (range))
val = CHAR_TABLE_REF (char_table, XFIXNAT (range));
@@ -642,7 +642,7 @@ or a character code. Return VALUE. */)
for (i = 0; i < chartab_size[0]; i++)
set_char_table_contents (char_table, i, value);
}
- else if (EQ (range, Qnil))
+ else if (NILP (range))
set_char_table_defalt (char_table, value);
else if (CHARACTERP (range))
char_table_set (char_table, XFIXNUM (range), value);
diff --git a/src/dbusbind.c b/src/dbusbind.c
index 4cf5604..fe92d39 100644
--- a/src/dbusbind.c
+++ b/src/dbusbind.c
@@ -200,7 +200,7 @@ xd_symbol_to_dbus_type (Lisp_Object object)
`dbus-send-signal', into corresponding C values appended as
arguments to a D-Bus message. */
#define XD_OBJECT_TO_DBUS_TYPE(object) \
- ((EQ (object, Qt) || EQ (object, Qnil)) ? DBUS_TYPE_BOOLEAN \
+ ((EQ (object, Qt) || NILP (object)) ? DBUS_TYPE_BOOLEAN \
: (FIXNATP (object)) ? DBUS_TYPE_UINT32 \
: (FIXNUMP (object)) ? DBUS_TYPE_INT32 \
: (FLOATP (object)) ? DBUS_TYPE_DOUBLE \
@@ -360,7 +360,7 @@ xd_signature (char *signature, int dtype, int parent_type,
Lisp_Object object)
break;
case DBUS_TYPE_BOOLEAN:
- if (!EQ (object, Qt) && !EQ (object, Qnil))
+ if (!EQ (object, Qt) && !NILP (object))
wrong_type_argument (intern ("booleanp"), object);
sprintf (signature, "%c", dtype);
break;
diff --git a/src/deps.mk b/src/deps.mk
index 7b6ae9c..f202d0e 100644
--- a/src/deps.mk
+++ b/src/deps.mk
@@ -71,7 +71,7 @@ cmds.o: cmds.c syntax.h buffer.h character.h commands.h
window.h lisp.h \
pre-crt0.o: pre-crt0.c
dbusbind.o: dbusbind.c termhooks.h frame.h keyboard.h lisp.h $(config_h)
dired.o: dired.c commands.h buffer.h lisp.h $(config_h) character.h charset.h \
- coding.h regex.h systime.h blockinput.h atimer.h composite.h \
+ coding.h regex-emacs.h systime.h blockinput.h atimer.h composite.h \
../lib/filemode.h ../lib/unistd.h globals.h
dispnew.o: dispnew.c systime.h commands.h process.h frame.h coding.h \
window.h buffer.h termchar.h termopts.h termhooks.h cm.h \
@@ -169,20 +169,21 @@ process.o: process.c process.h buffer.h window.h
termhooks.h termopts.h \
blockinput.h atimer.h coding.h msdos.h nsterm.h composite.h \
keyboard.h lisp.h globals.h $(config_h) character.h xgselect.h sysselect.h \
../lib/unistd.h gnutls.h
-regex.o: regex.c syntax.h buffer.h lisp.h globals.h $(config_h) regex.h \
+regex-emacs.o: regex-emacs.c syntax.h buffer.h lisp.h globals.h \
+ $(config_h) regex-emacs.h \
category.h character.h
region-cache.o: region-cache.c buffer.h region-cache.h \
lisp.h globals.h $(config_h)
scroll.o: scroll.c termchar.h dispextern.h frame.h msdos.h keyboard.h \
termhooks.h lisp.h globals.h $(config_h) systime.h coding.h composite.h \
window.h
-search.o: search.c regex.h commands.h buffer.h region-cache.h syntax.h \
+search.o: search.c regex-emacs.h commands.h buffer.h region-cache.h syntax.h \
blockinput.h atimer.h systime.h category.h character.h charset.h \
$(INTERVALS_H) lisp.h globals.h $(config_h)
sound.o: sound.c dispextern.h syssignal.h lisp.h globals.h $(config_h) \
atimer.h systime.h ../lib/unistd.h msdos.h
syntax.o: syntax.c syntax.h buffer.h commands.h category.h character.h \
- keymap.h regex.h $(INTERVALS_H) lisp.h globals.h $(config_h)
+ keymap.h regex-emacs.h $(INTERVALS_H) lisp.h globals.h $(config_h)
sysdep.o: sysdep.c syssignal.h systty.h systime.h syswait.h blockinput.h \
process.h dispextern.h termhooks.h termchar.h termopts.h coding.h \
frame.h atimer.h window.h msdos.h dosfns.h keyboard.h cm.h lisp.h \
diff --git a/src/dired.c b/src/dired.c
index 3bb5997..b92cd2b 100644
--- a/src/dired.c
+++ b/src/dired.c
@@ -1058,7 +1058,7 @@ return a list with one element, taken from
`user-real-login-name'. */)
endpwent ();
#endif
- if (EQ (users, Qnil))
+ if (NILP (users))
/* At least current user is always known. */
users = list1 (Vuser_real_login_name);
return users;
diff --git a/src/dispextern.h b/src/dispextern.h
index 944a37d..9cc65f6 100644
--- a/src/dispextern.h
+++ b/src/dispextern.h
@@ -2482,7 +2482,7 @@ struct it
If `what' is anything else, these two are undefined (will
probably hold values for the last IT_CHARACTER or IT_COMPOSITION
- traversed by the iterator.
+ traversed by the iterator).
The values are updated by get_next_display_element, so they are
out of sync with the value returned by IT_CHARPOS between the
diff --git a/src/editfns.c b/src/editfns.c
index f9157cd..b1b9eb6 100644
--- a/src/editfns.c
+++ b/src/editfns.c
@@ -3256,21 +3256,9 @@ differences between the two buffers. */)
Instead, we announce a single modification for the entire
modified region. But don't do that if the caller inhibited
modification hooks, because then they don't want that. */
- ptrdiff_t from, to;
if (!inhibit_modification_hooks)
{
- ptrdiff_t k, l;
-
- /* Find the first character position to be changed. */
- for (k = 0; k < size_a && !bit_is_set (ctx.deletions, k); k++)
- ;
- from = BEGV + k;
-
- /* Find the last character position to be changed. */
- for (l = size_a; l > 0 && !bit_is_set (ctx.deletions, l - 1); l--)
- ;
- to = BEGV + l;
- prepare_to_modify_buffer (from, to, NULL);
+ prepare_to_modify_buffer (BEGV, ZV, NULL);
specbind (Qinhibit_modification_hooks, Qt);
modification_hooks_inhibited = true;
}
@@ -3322,9 +3310,8 @@ differences between the two buffers. */)
if (modification_hooks_inhibited)
{
- ptrdiff_t updated_to = to + ZV - BEGV - size_a;
- signal_after_change (from, to - from, updated_to - from);
- update_compositions (from, updated_to, CHECK_INSIDE);
+ signal_after_change (BEGV, size_a, ZV - BEGV);
+ update_compositions (BEGV, ZV, CHECK_INSIDE);
}
return Qnil;
@@ -4195,14 +4182,14 @@ Nth argument is substituted instead of the next one. A
format can
contain either numbered or unnumbered %-sequences but not both, except
that %% can be mixed with numbered %-sequences.
-The + flag character inserts a + before any positive number, while a
-space inserts a space before any positive number; these flags only
-affect %d, %e, %f, and %g sequences, and the + flag takes precedence.
+The + flag character inserts a + before any nonnegative number, while a
+space inserts a space before any nonnegative number; these flags
+affect only numeric %-sequences, and the + flag takes precedence.
The - and 0 flags affect the width specifier, as described below.
The # flag means to use an alternate display form for %o, %x, %X, %e,
%f, and %g sequences: for %o, it ensures that the result begins with
-\"0\"; for %x and %X, it prefixes the result with \"0x\" or \"0X\";
+\"0\"; for %x and %X, it prefixes nonzero results with \"0x\" or \"0X\";
for %e and %f, it causes a decimal point to be included even if the
precision is zero; for %g, it causes a decimal point to be
included even if the precision is zero, and also forces trailing
@@ -4756,10 +4743,22 @@ styled_format (ptrdiff_t nargs, Lisp_Object *args, bool
message)
}
else
{
- /* Don't sign-extend for octal or hex printing. */
uprintmax_t x;
- if (FIXNUMP (arg))
- x = XUFIXNUM (arg);
+ bool negative;
+ if (INTEGERP (arg))
+ {
+ if (binary_as_unsigned)
+ {
+ x = XUINT (arg);
+ negative = false;
+ }
+ else
+ {
+ EMACS_INT i = XINT (arg);
+ negative = i < 0;
+ x = negative ? -i : i;
+ }
+ }
else
{
double d = XFLOAT_DATA (arg);
@@ -4767,8 +4766,13 @@ styled_format (ptrdiff_t nargs, Lisp_Object *args, bool
message)
if (! (0 <= d && d < uprintmax + 1))
xsignal1 (Qoverflow_error, arg);
x = d;
+ negative = false;
}
- sprintf_bytes = sprintf (sprintf_buf, convspec, prec, x);
+ sprintf_buf[0] = negative ? '-' : plus_flag ? '+' : ' ';
+ bool signedp = negative | plus_flag | space_flag;
+ sprintf_bytes = sprintf (sprintf_buf + signedp,
+ convspec, prec, x);
+ sprintf_bytes += signedp;
}
/* Now the length of the formatted item is known, except it omits
@@ -5578,6 +5582,22 @@ functions if all the text being accessed has this
property. */);
DEFVAR_LISP ("operating-system-release", Voperating_system_release,
doc: /* The release of the operating system Emacs is running on.
*/);
+ DEFVAR_BOOL ("binary-as-unsigned",
+ binary_as_unsigned,
+ doc: /* Non-nil means `format' %x and %o treat integers as
unsigned.
+This has machine-dependent results. Nil means to treat integers as
+signed, which is portable; for example, if N is a negative integer,
+(read (format "#x%x") N) returns N only when this variable is nil.
+
+This variable is experimental; email address@hidden if you need
+it to be non-nil. */);
+ /* For now, default to true if bignums exist, false in traditional Emacs. */
+#ifdef lisp_h_FIXNUMP
+ binary_as_unsigned = false;
+#else
+ binary_as_unsigned = true;
+#endif
+
defsubr (&Spropertize);
defsubr (&Schar_equal);
defsubr (&Sgoto_char);
diff --git a/src/emacs.c b/src/emacs.c
index dc62ce8..97205d2 100644
--- a/src/emacs.c
+++ b/src/emacs.c
@@ -84,7 +84,7 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#include "composite.h"
#include "dispextern.h"
#include "ptr-bounds.h"
-#include "regex.h"
+#include "regex-emacs.h"
#include "sheap.h"
#include "syntax.h"
#include "sysselect.h"
@@ -862,9 +862,9 @@ main (int argc, char **argv)
{
rlim_t lim = rlim.rlim_cur;
- /* Approximate the amount regex.c needs per unit of
+ /* Approximate the amount regex-emacs.c needs per unit of
emacs_re_max_failures, then add 33% to cover the size of the
- smaller stacks that regex.c successively allocates and
+ smaller stacks that regex-emacs.c successively allocates and
discards on its way to the maximum. */
int min_ratio = 20 * sizeof (char *);
int ratio = min_ratio + min_ratio / 3;
@@ -903,7 +903,7 @@ main (int argc, char **argv)
lim = newlim;
}
}
- /* If the stack is big enough, let regex.c more of it before
+ /* If the stack is big enough, let regex-emacs.c more of it before
falling back to heap allocation. */
emacs_re_safe_alloca = max
(min (lim - extra, SIZE_MAX) * (min_ratio / ratio),
@@ -2035,6 +2035,10 @@ all of which are called before Emacs is actually killed.
*/
{
int exit_code;
+#ifdef HAVE_LIBSYSTEMD
+ sd_notify(0, "STOPPING=1");
+#endif /* HAVE_LIBSYSTEMD */
+
/* Fsignal calls emacs_abort () if it sees that waiting_for_input is
set. */
waiting_for_input = 0;
@@ -2495,6 +2499,13 @@ from the parent process and its tty file descriptors.
*/)
error ("This function can only be called after loading the init files");
#ifndef WINDOWSNT
+ if (daemon_type == 1)
+ {
+#ifdef HAVE_LIBSYSTEMD
+ sd_notify(0, "READY=1");
+#endif /* HAVE_LIBSYSTEMD */
+ }
+
if (daemon_type == 2)
{
int nfd;
diff --git a/src/eval.c b/src/eval.c
index f7d4d5b..8745ba9 100644
--- a/src/eval.c
+++ b/src/eval.c
@@ -1732,28 +1732,12 @@ xsignal3 (Lisp_Object error_symbol, Lisp_Object arg1,
Lisp_Object arg2, Lisp_Obj
}
/* Signal `error' with message S, and additional arg ARG.
- If ARG is not a genuine list, make it a one-element list. */
+ If ARG is not a proper list, make it a one-element list. */
void
signal_error (const char *s, Lisp_Object arg)
{
- Lisp_Object tortoise, hare;
-
- hare = tortoise = arg;
- while (CONSP (hare))
- {
- hare = XCDR (hare);
- if (!CONSP (hare))
- break;
-
- hare = XCDR (hare);
- tortoise = XCDR (tortoise);
-
- if (EQ (hare, tortoise))
- break;
- }
-
- if (!NILP (hare))
+ if (NILP (Fproper_list_p (arg)))
arg = list1 (arg);
xsignal (Qerror, Fcons (build_string (s), arg));
diff --git a/src/fileio.c b/src/fileio.c
index 14089dc..04e763f 100644
--- a/src/fileio.c
+++ b/src/fileio.c
@@ -196,8 +196,8 @@ check_writable (const char *filename, int amode)
list before reporting it; this saves report_file_errno's caller the
trouble of preserving errno before calling list1. */
-void
-report_file_errno (char const *string, Lisp_Object name, int errorno)
+Lisp_Object
+get_file_errno_data (char const *string, Lisp_Object name, int errorno)
{
Lisp_Object data = CONSP (name) || NILP (name) ? name : list1 (name);
char *str = emacs_strerror (errorno);
@@ -207,10 +207,18 @@ report_file_errno (char const *string, Lisp_Object name,
int errorno)
Lisp_Object errdata = Fcons (errstring, data);
if (errorno == EEXIST)
- xsignal (Qfile_already_exists, errdata);
+ return Fcons (Qfile_already_exists, errdata);
else
- xsignal (errorno == ENOENT ? Qfile_missing : Qfile_error,
- Fcons (build_string (string), errdata));
+ return Fcons (errorno == ENOENT ? Qfile_missing : Qfile_error,
+ Fcons (build_string (string), errdata));
+}
+
+void
+report_file_errno (char const *string, Lisp_Object name, int errorno)
+{
+ Lisp_Object data = get_file_errno_data (string, name, errorno);
+
+ xsignal (Fcar (data), Fcdr (data));
}
/* Signal a file-access failure that set errno. STRING describes the
@@ -2288,6 +2296,21 @@ The arg must be a string. */)
if (!NILP (handler))
return call2 (handler, Qfile_name_case_insensitive_p, filename);
+ /* If the file doesn't exist, move up the filesystem tree until we
+ reach an existing directory or the root. */
+ if (NILP (Ffile_exists_p (filename)))
+ {
+ filename = Ffile_name_directory (filename);
+ while (NILP (Ffile_exists_p (filename)))
+ {
+ Lisp_Object newname = expand_and_dir_to_file (filename);
+ /* Avoid infinite loop if the root is reported as non-existing
+ (impossible?). */
+ if (!NILP (Fstring_equal (newname, filename)))
+ break;
+ filename = newname;
+ }
+ }
filename = ENCODE_FILE (filename);
return file_name_case_insensitive_p (SSDATA (filename)) ? Qt : Qnil;
}
@@ -5714,7 +5737,7 @@ A non-nil CURRENT-ONLY argument means save only current
buffer. */)
spare the user annoying messages. */
&& XFIXNAT (BVAR (b, save_length)) > 5000
/* These messages are frequent and annoying for `*mail*'. */
- && !EQ (BVAR (b, filename), Qnil)
+ && !NILP (BVAR (b, filename))
&& NILP (no_message))
{
/* It has shrunk too much; turn off auto-saving here. */
diff --git a/src/floatfns.c b/src/floatfns.c
index f8463f3..bbf7df4 100644
--- a/src/floatfns.c
+++ b/src/floatfns.c
@@ -47,13 +47,6 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#include <count-leading-zeros.h>
-#ifndef isfinite
-# define isfinite(x) ((x) - (x) == 0)
-#endif
-#ifndef isnan
-# define isnan(x) ((x) != (x))
-#endif
-
/* Check that X is a floating point number. */
static void
diff --git a/src/fns.c b/src/fns.c
index 92a853e..38b2d28 100644
--- a/src/fns.c
+++ b/src/fns.c
@@ -144,6 +144,28 @@ which is at least the number of distinct elements. */)
return make_fixnum_or_float (len);
}
+DEFUN ("proper-list-p", Fproper_list_p, Sproper_list_p, 1, 1, 0,
+ doc: /* Return OBJECT's length if it is a proper list, nil otherwise.
+A proper list is neither circular nor dotted (i.e., its last cdr is nil). */
+ attributes: const)
+ (Lisp_Object object)
+{
+ intptr_t len = 0;
+ Lisp_Object last_tail = object;
+ Lisp_Object tail = object;
+ FOR_EACH_TAIL_SAFE (tail)
+ {
+ len++;
+ rarely_quit (len);
+ last_tail = XCDR (tail);
+ }
+ if (!NILP (last_tail))
+ return Qnil;
+ if (MOST_POSITIVE_FIXNUM < len)
+ xsignal0 (Qoverflow_error);
+ return make_number (len);
+}
+
DEFUN ("string-bytes", Fstring_bytes, Sstring_bytes, 1, 1, 0,
doc: /* Return the number of bytes in STRING.
If STRING is multibyte, this may be greater than the length of STRING. */)
@@ -718,7 +740,7 @@ concat (ptrdiff_t nargs, Lisp_Object *args,
val = make_uninit_string (result_len);
/* In `append', if all but last arg are nil, return last arg. */
- if (target_type == Lisp_Cons && EQ (val, Qnil))
+ if (target_type == Lisp_Cons && NILP (val))
return last_tail;
/* Copy the contents of the args into the result. */
@@ -1419,6 +1441,29 @@ DEFUN ("elt", Felt, Selt, 2, 2, 0,
return Faref (sequence, n);
}
+enum { WORDS_PER_DOUBLE = (sizeof (double) / sizeof (EMACS_UINT)
+ + (sizeof (double) % sizeof (EMACS_UINT) != 0)) };
+union double_and_words
+{
+ double val;
+ EMACS_UINT word[WORDS_PER_DOUBLE];
+};
+
+/* Return true if X and Y are the same floating-point value.
+ This looks at X's and Y's representation, since (unlike '==')
+ it returns true if X and Y are the same NaN. */
+static bool
+same_float (Lisp_Object x, Lisp_Object y)
+{
+ union double_and_words
+ xu = { .val = XFLOAT_DATA (x) },
+ yu = { .val = XFLOAT_DATA (y) };
+ EMACS_UINT neql = 0;
+ for (int i = 0; i < WORDS_PER_DOUBLE; i++)
+ neql |= xu.word[i] ^ yu.word[i];
+ return !neql;
+}
+
DEFUN ("member", Fmember, Smember, 2, 2, 0,
doc: /* Return non-nil if ELT is an element of LIST. Comparison done
with `equal'.
The value is actually the tail of LIST whose car is ELT. */)
@@ -1457,7 +1502,7 @@ The value is actually the tail of LIST whose car is ELT.
*/)
FOR_EACH_TAIL (tail)
{
Lisp_Object tem = XCAR (tail);
- if (FLOATP (tem) && equal_no_quit (elt, tem))
+ if (FLOATP (tem) && same_float (elt, tem))
return tail;
}
CHECK_LIST_END (tail, list);
@@ -2170,11 +2215,15 @@ The PLIST is modified by side effects. */)
}
DEFUN ("eql", Feql, Seql, 2, 2, 0,
- doc: /* Return t if the two args are the same Lisp object.
-Floating-point numbers of equal value are `eql', but they may not be `eq'. */)
+ doc: /* Return t if the two args are `eq' or are indistinguishable
numbers.
+Floating-point values with the same sign, exponent and fraction are `eql'.
+This differs from numeric comparison: (eql 0.0 -0.0) returns nil and
+\(eql 0.0e+NaN 0.0e+NaN) returns t, whereas `=' does the opposite. */)
(Lisp_Object obj1, Lisp_Object obj2)
{
- if (FLOATP (obj1) || BIGNUMP (obj1))
+ if (FLOATP (obj1))
+ return FLOATP (obj2) && same_float (obj1, obj2) ? Qt : Qnil;
+ else if (BIGNUMP (obj1))
return equal_no_quit (obj1, obj2) ? Qt : Qnil;
else
return EQ (obj1, obj2) ? Qt : Qnil;
@@ -2185,8 +2234,8 @@ DEFUN ("equal", Fequal, Sequal, 2, 2, 0,
They must have the same data type.
Conses are compared by comparing the cars and the cdrs.
Vectors and strings are compared element by element.
-Numbers are compared by value, but integers cannot equal floats.
- (Use `=' if you want integers and floats to be able to be equal.)
+Numbers are compared via `eql', so integers do not equal floats.
+\(Use `=' if you want integers and floats to be able to be equal.)
Symbols must match exactly. */)
(Lisp_Object o1, Lisp_Object o2)
{
@@ -2266,13 +2315,7 @@ internal_equal (Lisp_Object o1, Lisp_Object o2, enum
equal_kind equal_kind,
switch (XTYPE (o1))
{
case Lisp_Float:
- {
- double d1 = XFLOAT_DATA (o1);
- double d2 = XFLOAT_DATA (o2);
- /* If d is a NaN, then d != d. Two NaNs should be `equal' even
- though they are not =. */
- return d1 == d2 || (d1 != d1 && d2 != d2);
- }
+ return same_float (o1, o2);
case Lisp_Cons:
if (equal_kind == EQUAL_NO_QUIT)
@@ -3708,9 +3751,8 @@ HASH_INDEX (struct Lisp_Hash_Table *h, ptrdiff_t idx)
return XFIXNUM (AREF (h->index, idx));
}
-/* Compare KEY1 which has hash code HASH1 and KEY2 with hash code
- HASH2 in hash table H using `eql'. Value is true if KEY1 and
- KEY2 are the same. */
+/* Compare KEY1 and KEY2 in hash table HT using `eql'. Value is true
+ if KEY1 and KEY2 are the same. KEY1 and KEY2 must not be eq. */
static bool
cmpfn_eql (struct hash_table_test *ht,
@@ -3719,7 +3761,7 @@ cmpfn_eql (struct hash_table_test *ht,
{
if (FLOATP (key1)
&& FLOATP (key2)
- && XFLOAT_DATA (key1) == XFLOAT_DATA (key2))
+ && same_float (key1, key2))
return true;
return (BIGNUMP (key1)
&& BIGNUMP (key2)
@@ -3727,9 +3769,8 @@ cmpfn_eql (struct hash_table_test *ht,
}
-/* Compare KEY1 which has hash code HASH1 and KEY2 with hash code
- HASH2 in hash table H using `equal'. Value is true if KEY1 and
- KEY2 are the same. */
+/* Compare KEY1 and KEY2 in hash table HT using `equal'. Value is
+ true if KEY1 and KEY2 are the same. */
static bool
cmpfn_equal (struct hash_table_test *ht,
@@ -3740,9 +3781,8 @@ cmpfn_equal (struct hash_table_test *ht,
}
-/* Compare KEY1 which has hash code HASH1, and KEY2 with hash code
- HASH2 in hash table H using H->user_cmp_function. Value is true
- if KEY1 and KEY2 are the same. */
+/* Compare KEY1 and KEY2 in hash table HT using HT->user_cmp_function.
+ Value is true if KEY1 and KEY2 are the same. */
static bool
cmpfn_user_defined (struct hash_table_test *ht,
@@ -4336,18 +4376,8 @@ static EMACS_UINT
sxhash_float (double val)
{
EMACS_UINT hash = 0;
- enum {
- WORDS_PER_DOUBLE = (sizeof val / sizeof hash
- + (sizeof val % sizeof hash != 0))
- };
- union {
- double val;
- EMACS_UINT word[WORDS_PER_DOUBLE];
- } u;
- int i;
- u.val = val;
- memset (&u.val + 1, 0, sizeof u - sizeof u.val);
- for (i = 0; i < WORDS_PER_DOUBLE; i++)
+ union double_and_words u = { .val = val };
+ for (int i = 0; i < WORDS_PER_DOUBLE; i++)
hash = sxhash_combine (hash, u.word[i]);
return SXHASH_REDUCE (hash);
}
@@ -5319,6 +5349,7 @@ this variable. */);
defsubr (&Srandom);
defsubr (&Slength);
defsubr (&Ssafe_length);
+ defsubr (&Sproper_list_p);
defsubr (&Sstring_bytes);
defsubr (&Sstring_distance);
defsubr (&Sstring_equal);
diff --git a/src/frame.c b/src/frame.c
index c2e3d4b..f9a73c8 100644
--- a/src/frame.c
+++ b/src/frame.c
@@ -139,14 +139,9 @@ check_window_system (struct frame *f)
/* Return the value of frame parameter PROP in frame FRAME. */
Lisp_Object
-get_frame_param (register struct frame *frame, Lisp_Object prop)
+get_frame_param (struct frame *frame, Lisp_Object prop)
{
- register Lisp_Object tem;
-
- tem = Fassq (prop, frame->param_alist);
- if (EQ (tem, Qnil))
- return tem;
- return Fcdr (tem);
+ return Fcdr (Fassq (prop, frame->param_alist));
}
@@ -189,9 +184,9 @@ frame_inhibit_resize (struct frame *f, bool horizontal,
Lisp_Object parameter)
|| (CONSP (frame_inhibit_implied_resize)
&& !NILP (Fmemq (parameter, frame_inhibit_implied_resize)))
|| (horizontal
- && !EQ (fullscreen, Qnil) && !EQ (fullscreen, Qfullheight))
+ && !NILP (fullscreen) && !EQ (fullscreen, Qfullheight))
|| (!horizontal
- && !EQ (fullscreen, Qnil) && !EQ (fullscreen, Qfullwidth))
+ && !NILP (fullscreen) && !EQ (fullscreen, Qfullwidth))
|| FRAME_TERMCAP_P (f) || FRAME_MSDOS_P (f))
: ((horizontal && f->inhibit_horizontal_resize)
|| (!horizontal && f->inhibit_vertical_resize)));
@@ -2808,10 +2803,8 @@ frames_discard_buffer (Lisp_Object buffer)
void
store_in_alist (Lisp_Object *alistptr, Lisp_Object prop, Lisp_Object val)
{
- register Lisp_Object tem;
-
- tem = Fassq (prop, *alistptr);
- if (EQ (tem, Qnil))
+ Lisp_Object tem = Fassq (prop, *alistptr);
+ if (NILP (tem))
*alistptr = Fcons (Fcons (prop, val), *alistptr);
else
Fsetcdr (tem, val);
@@ -2975,7 +2968,7 @@ store_frame_param (struct frame *f, Lisp_Object prop,
Lisp_Object val)
/* Update the frame parameter alist. */
old_alist_elt = Fassq (prop, f->param_alist);
- if (EQ (old_alist_elt, Qnil))
+ if (NILP (old_alist_elt))
fset_param_alist (f, Fcons (Fcons (prop, val), f->param_alist));
else
Fsetcdr (old_alist_elt, val);
@@ -4516,13 +4509,13 @@ x_set_visibility (struct frame *f, Lisp_Object value,
Lisp_Object oldval)
void
x_set_autoraise (struct frame *f, Lisp_Object arg, Lisp_Object oldval)
{
- f->auto_raise = !EQ (Qnil, arg);
+ f->auto_raise = !NILP (arg);
}
void
x_set_autolower (struct frame *f, Lisp_Object arg, Lisp_Object oldval)
{
- f->auto_lower = !EQ (Qnil, arg);
+ f->auto_lower = !NILP (arg);
}
void
@@ -4973,7 +4966,7 @@ x_get_arg (Display_Info *dpyinfo, Lisp_Object alist,
Lisp_Object param,
/* If it wasn't specified in ALIST or the Lisp-level defaults,
look in the X resources. */
- if (EQ (tem, Qnil))
+ if (NILP (tem))
{
if (attribute && dpyinfo)
{
diff --git a/src/gnutls.c b/src/gnutls.c
index 111dc61..2d455ea 100644
--- a/src/gnutls.c
+++ b/src/gnutls.c
@@ -2071,7 +2071,14 @@ gnutls_symmetric (bool encrypting, Lisp_Object cipher,
cipher = intern (SSDATA (cipher));
if (SYMBOLP (cipher))
- info = XCDR (Fassq (cipher, Fgnutls_ciphers ()));
+ {
+ info = Fassq (cipher, Fgnutls_ciphers ());
+ if (!CONSP (info))
+ xsignal2 (Qerror,
+ build_string ("GnuTLS cipher is invalid or not found"),
+ cipher);
+ info = XCDR (info);
+ }
else if (TYPE_RANGED_FIXNUMP (gnutls_cipher_algorithm_t, cipher))
gca = XFIXNUM (cipher);
else
@@ -2086,7 +2093,8 @@ gnutls_symmetric (bool encrypting, Lisp_Object cipher,
ptrdiff_t key_size = gnutls_cipher_get_key_size (gca);
if (key_size == 0)
- error ("GnuTLS cipher is invalid or not found");
+ xsignal2 (Qerror,
+ build_string ("GnuTLS cipher is invalid or not found"), cipher);
ptrdiff_t kstart_byte, kend_byte;
const char *kdata = extract_data_from_object (key, &kstart_byte, &kend_byte);
@@ -2342,7 +2350,14 @@ itself. */)
hash_method = intern (SSDATA (hash_method));
if (SYMBOLP (hash_method))
- info = XCDR (Fassq (hash_method, Fgnutls_macs ()));
+ {
+ info = Fassq (hash_method, Fgnutls_macs ());
+ if (!CONSP (info))
+ xsignal2 (Qerror,
+ build_string ("GnuTLS MAC-method is invalid or not found"),
+ hash_method);
+ info = XCDR (info);
+ }
else if (TYPE_RANGED_FIXNUMP (gnutls_mac_algorithm_t, hash_method))
gma = XFIXNUM (hash_method);
else
@@ -2357,7 +2372,9 @@ itself. */)
ptrdiff_t digest_length = gnutls_hmac_get_len (gma);
if (digest_length == 0)
- error ("GnuTLS MAC-method is invalid or not found");
+ xsignal2 (Qerror,
+ build_string ("GnuTLS MAC-method is invalid or not found"),
+ hash_method);
ptrdiff_t kstart_byte, kend_byte;
const char *kdata = extract_data_from_object (key, &kstart_byte, &kend_byte);
@@ -2423,7 +2440,14 @@ the number itself. */)
digest_method = intern (SSDATA (digest_method));
if (SYMBOLP (digest_method))
- info = XCDR (Fassq (digest_method, Fgnutls_digests ()));
+ {
+ info = Fassq (digest_method, Fgnutls_digests ());
+ if (!CONSP (info))
+ xsignal2 (Qerror,
+ build_string ("GnuTLS digest-method is invalid or not found"),
+ digest_method);
+ info = XCDR (info);
+ }
else if (TYPE_RANGED_FIXNUMP (gnutls_digest_algorithm_t, digest_method))
gda = XFIXNUM (digest_method);
else
@@ -2438,7 +2462,9 @@ the number itself. */)
ptrdiff_t digest_length = gnutls_hash_get_len (gda);
if (digest_length == 0)
- error ("GnuTLS digest-method is invalid or not found");
+ xsignal2 (Qerror,
+ build_string ("GnuTLS digest-method is invalid or not found"),
+ digest_method);
gnutls_hash_hd_t hash;
int ret = gnutls_hash_init (&hash, gda);
diff --git a/src/image.c b/src/image.c
index 78eaf8a..499c1b6 100644
--- a/src/image.c
+++ b/src/image.c
@@ -1610,7 +1610,7 @@ Anything else, means only clear those images which refer
to FILTER,
which is then usually a filename. */)
(Lisp_Object filter)
{
- if (!(EQ (filter, Qnil) || FRAMEP (filter)))
+ if (! (NILP (filter) || FRAMEP (filter)))
clear_image_caches (filter);
else
clear_image_cache (decode_window_system_frame (filter), Qt);
diff --git a/src/intervals.c b/src/intervals.c
index af27afe..7e77714 100644
--- a/src/intervals.c
+++ b/src/intervals.c
@@ -197,7 +197,7 @@ intervals_equal (INTERVAL i0, INTERVAL i1)
}
/* i0 has something i1 doesn't. */
- if (EQ (i1_val, Qnil))
+ if (NILP (i1_val))
return false;
/* i0 and i1 both have sym, but it has different values in each. */
diff --git a/src/intervals.h b/src/intervals.h
index 162c4ef..f37372a 100644
--- a/src/intervals.h
+++ b/src/intervals.h
@@ -116,7 +116,7 @@ struct interval
/* True if this is a default interval, which is the same as being null
or having no properties. */
-#define DEFAULT_INTERVAL_P(i) (!i || EQ ((i)->plist, Qnil))
+#define DEFAULT_INTERVAL_P(i) (!i || NILP ((i)->plist))
/* Test what type of parent we have. Three possibilities: another
interval, a buffer or string object, or NULL. */
diff --git a/src/json.c b/src/json.c
index 17edb41..4e413a2 100644
--- a/src/json.c
+++ b/src/json.c
@@ -7,7 +7,7 @@ This file is part of GNU Emacs.
GNU Emacs is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or (at
-nyour option) any later version.
+your option) any later version.
GNU Emacs is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
@@ -159,7 +159,12 @@ init_json_functions (void)
than PTRDIFF_MAX. Such objects wouldn't play well with the rest of
Emacs's codebase, which generally uses ptrdiff_t for sizes and
indices. The other functions in this file also generally assume
- that size_t values never exceed PTRDIFF_MAX. */
+ that size_t values never exceed PTRDIFF_MAX.
+
+ In addition, we need to use a custom allocator because on
+ MS-Windows we replace malloc/free with our own functions, see
+ w32heap.c, so we must force the library to use our allocator, or
+ else we won't be able to free storage allocated by the library. */
static void *
json_malloc (size_t size)
@@ -605,7 +610,7 @@ usage: (json-serialize OBJECT &rest ARGS) */)
char *string = json_dumps (json, JSON_COMPACT);
if (string == NULL)
json_out_of_memory ();
- record_unwind_protect_ptr (free, string);
+ record_unwind_protect_ptr (json_free, string);
return unbind_to (count, json_build_string (string));
}
diff --git a/src/keyboard.c b/src/keyboard.c
index 4c8807d..e596b28 100644
--- a/src/keyboard.c
+++ b/src/keyboard.c
@@ -3150,6 +3150,10 @@ help_char_p (Lisp_Object c)
static void
record_char (Lisp_Object c)
{
+ /* quail.el binds this to avoid recording keys twice. */
+ if (inhibit_record_char)
+ return;
+
int recorded = 0;
if (CONSP (c) && (EQ (XCAR (c), Qhelp_echo) || EQ (XCAR (c),
Qmouse_movement)))
@@ -3256,7 +3260,7 @@ record_char (Lisp_Object c)
/* Write c to the dribble file. If c is a lispy event, write
the event's symbol to the dribble file, in <brackets>. Bleaugh.
If you, dear reader, have a better idea, you've got the source. :-) */
- if (dribble)
+ if (dribble && NILP (Vexecuting_kbd_macro))
{
block_input ();
if (FIXNUMP (c))
@@ -10110,10 +10114,13 @@ DEFUN ("recursion-depth", Frecursion_depth,
Srecursion_depth, 0, 0, 0,
DEFUN ("open-dribble-file", Fopen_dribble_file, Sopen_dribble_file, 1, 1,
"FOpen dribble file: ",
- doc: /* Start writing all keyboard characters to a dribble file called
FILE.
+ doc: /* Start writing input events to a dribble file called FILE.
If FILE is nil, close any open dribble file.
The file will be closed when Emacs exits.
+The events written to the file include keyboard and mouse input
+events, but not events from executing keyboard macros.
+
Be aware that this records ALL characters you type!
This may include sensitive information such as passwords. */)
(Lisp_Object file)
@@ -11813,10 +11820,10 @@ if the command is in this list, the selection is not
updated. */);
DEFVAR_LISP ("debug-on-event",
Vdebug_on_event,
- doc: /* Enter debugger on this event. When Emacs
-receives the special event specified by this variable, it will try to
-break into the debugger as soon as possible instead of processing the
-event normally through `special-event-map'.
+ doc: /* Enter debugger on this event.
+When Emacs receives the special event specified by this variable,
+it will try to break into the debugger as soon as possible instead
+of processing the event normally through `special-event-map'.
Currently, the only supported values for this
variable are `sigusr1' and `sigusr2'. */);
@@ -11824,21 +11831,23 @@ variable are `sigusr1' and `sigusr2'. */);
DEFVAR_BOOL ("attempt-stack-overflow-recovery",
attempt_stack_overflow_recovery,
- doc: /* If non-nil, attempt to recover from C stack
-overflow. This recovery is unsafe and may lead to deadlocks or data
+ doc: /* If non-nil, attempt to recover from C stack overflows.
+This recovery is potentially unsafe and may lead to deadlocks or data
corruption, but it usually works and may preserve modified buffers
that would otherwise be lost. If nil, treat stack overflow like any
-other kind of crash. */);
+other kind of crash or fatal error. */);
attempt_stack_overflow_recovery = true;
DEFVAR_BOOL ("attempt-orderly-shutdown-on-fatal-signal",
attempt_orderly_shutdown_on_fatal_signal,
- doc: /* If non-nil, attempt to perform an orderly
-shutdown when Emacs receives a fatal signal (e.g., a crash).
-This cleanup is unsafe and may lead to deadlocks or data corruption,
-but it usually works and may preserve modified buffers that would
-otherwise be lost. If nil, crash immediately in response to fatal
-signals. */);
+ doc: /* If non-nil, attempt orderly shutdown on fatal signals.
+By default this variable is non-nil, and Emacs attempts to perform
+an orderly shutdown when it catches a fatal signal (e.g., a crash).
+The orderly shutdown includes an attempt to auto-save your unsaved edits
+and other useful cleanups. These cleanups are potentially unsafe and may
+lead to deadlocks or data corruption, but it usually works and may
+preserve data in modified buffers that would otherwise be lost.
+If nil, Emacs crashes immediately in response to fatal signals. */);
attempt_orderly_shutdown_on_fatal_signal = true;
/* Create the initial keyboard. Qt means 'unset'. */
@@ -11848,6 +11857,14 @@ signals. */);
Vwhile_no_input_ignore_events,
doc: /* Ignored events from while-no-input. */);
Vwhile_no_input_ignore_events = Qnil;
+
+ DEFVAR_BOOL ("inhibit--record-char",
+ inhibit_record_char,
+ doc: /* If non-nil, don't record input events.
+This inhibits recording input events for the purposes of keyboard
+macros, dribble file, and `recent-keys'.
+Internal use only. */);
+ inhibit_record_char = false;
}
void
diff --git a/src/lisp.h b/src/lisp.h
index 6726d69..6ca3416 100644
--- a/src/lisp.h
+++ b/src/lisp.h
@@ -2704,17 +2704,14 @@ XFLOAT_DATA (Lisp_Object f)
/* Most hosts nowadays use IEEE floating point, so they use IEC 60559
representations, have infinities and NaNs, and do not trap on
- exceptions. Define IEEE_FLOATING_POINT if this host is one of the
+ exceptions. Define IEEE_FLOATING_POINT to 1 if this host is one of the
typical ones. The C11 macro __STDC_IEC_559__ is close to what is
wanted here, but is not quite right because Emacs does not require
all the features of C11 Annex F (and does not require C11 at all,
for that matter). */
-enum
- {
- IEEE_FLOATING_POINT
- = (FLT_RADIX == 2 && FLT_MANT_DIG == 24
- && FLT_MIN_EXP == -125 && FLT_MAX_EXP == 128)
- };
+
+#define IEEE_FLOATING_POINT (FLT_RADIX == 2 && FLT_MANT_DIG == 24 \
+ && FLT_MIN_EXP == -125 && FLT_MAX_EXP == 128)
/* A character, declared with the following typedef, is a member
of some character set associated with the current buffer. */
@@ -4120,6 +4117,7 @@ extern Lisp_Object write_region (Lisp_Object,
Lisp_Object, Lisp_Object,
extern void close_file_unwind (int);
extern void fclose_unwind (void *);
extern void restore_point_unwind (Lisp_Object);
+extern Lisp_Object get_file_errno_data (const char *, Lisp_Object, int);
extern _Noreturn void report_file_errno (const char *, Lisp_Object, int);
extern _Noreturn void report_file_error (const char *, Lisp_Object);
extern _Noreturn void report_file_notify_error (const char *, Lisp_Object);
@@ -4803,7 +4801,7 @@ enum
#define FOR_EACH_TAIL(tail) \
FOR_EACH_TAIL_INTERNAL (tail, circular_list (tail), true)
-/* Like FOR_EACH_TAIL (LIST), except do not signal or quit.
+/* Like FOR_EACH_TAIL (TAIL), except do not signal or quit.
If the loop exits due to a cycle, TAIL’s value is undefined. */
#define FOR_EACH_TAIL_SAFE(tail) \
diff --git a/src/lread.c b/src/lread.c
index 3a2d9c8..df2fe58 100644
--- a/src/lread.c
+++ b/src/lread.c
@@ -72,6 +72,10 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#define file_tell ftell
#endif
+#if IEEE_FLOATING_POINT
+# include <ieee754.h>
+#endif
+
/* The objects or placeholders read with the #n=object form.
A hash table maps a number to either a placeholder (while the
@@ -2730,7 +2734,7 @@ read1 (Lisp_Object readcharfun, int *pch, bool
first_in_list)
int c;
bool uninterned_symbol = false;
bool multibyte;
- char stackbuf[MAX_ALLOCA];
+ char stackbuf[128]; /* Small, as read1 is recursive (Bug#31995). */
current_thread->stack_top = stackbuf;
*pch = 0;
@@ -3616,7 +3620,7 @@ substitute_object_recurse (struct subst *subst,
Lisp_Object subtree)
return subtree;
/* If we've been to this node before, don't explore it again. */
- if (!EQ (Qnil, Fmemq (subtree, subst->seen)))
+ if (!NILP (Fmemq (subtree, subst->seen)))
return subtree;
/* If this node can be the entry point to a cycle, remember that
@@ -3763,14 +3767,18 @@ string_to_number (char const *string, int base, int
flags)
cp += 3;
value = INFINITY;
}
+#if IEEE_FLOATING_POINT
else if (cp[-1] == '+'
&& cp[0] == 'N' && cp[1] == 'a' && cp[2] == 'N')
{
state |= E_EXP;
cp += 3;
- /* NAN is a "positive" NaN on all known Emacs hosts. */
- value = NAN;
+ union ieee754_double u
+ = { .ieee_nan = { .exponent = -1, .quiet_nan = 1,
+ .mantissa0 = n >> 31 >> 1, .mantissa1 = n }};
+ value = u.d;
}
+#endif
else
cp = ecp;
}
@@ -4236,7 +4244,7 @@ usage: (unintern NAME OBARRAY) */)
session if we unintern them, as well as even more ways to use
`setq' or `fset' or whatnot to make the Emacs session
unusable. Let's not go down this silly road. --Stef */
- /* if (EQ (tem, Qnil) || EQ (tem, Qt))
+ /* if (NILP (tem) || EQ (tem, Qt))
error ("Attempt to unintern t or nil"); */
XSYMBOL (tem)->u.s.interned = SYMBOL_UNINTERNED;
diff --git a/src/menu.c b/src/menu.c
index eac8201..d75a842 100644
--- a/src/menu.c
+++ b/src/menu.c
@@ -647,7 +647,7 @@ digest_single_submenu (int start, int end, bool
top_level_items)
i = start;
while (i < end)
{
- if (EQ (AREF (menu_items, i), Qnil))
+ if (NILP (AREF (menu_items, i)))
{
submenu_stack[submenu_depth++] = save_wv;
save_wv = prev_wv;
@@ -900,7 +900,7 @@ find_and_call_menu_selection (struct frame *f, int
menu_bar_items_used,
while (i < menu_bar_items_used)
{
- if (EQ (AREF (vector, i), Qnil))
+ if (NILP (AREF (vector, i)))
{
subprefix_stack[submenu_depth++] = prefix;
prefix = entry;
@@ -985,7 +985,7 @@ find_and_return_menu_selection (struct frame *f, bool
keymaps, void *client_data
while (i < menu_items_used)
{
- if (EQ (AREF (menu_items, i), Qnil))
+ if (NILP (AREF (menu_items, i)))
{
subprefix_stack[submenu_depth++] = prefix;
prefix = entry;
diff --git a/src/nsfns.m b/src/nsfns.m
index 527dd77..ece21c6 100644
--- a/src/nsfns.m
+++ b/src/nsfns.m
@@ -363,7 +363,7 @@ x_set_icon_name (struct frame *f, Lisp_Object arg,
Lisp_Object oldval)
if (STRINGP (oldval) && EQ (Fstring_equal (oldval, arg), Qt))
return;
}
- else if (!STRINGP (oldval) && EQ (oldval, Qnil) == EQ (arg, Qnil))
+ else if (!STRINGP (oldval) && NILP (oldval) == NILP (arg))
return;
fset_icon_name (f, arg);
@@ -1291,7 +1291,7 @@ DEFUN ("x-create-frame", Fx_create_frame, Sx_create_frame,
window_prompting = x_figure_window_size (f, parms, true, &x_width,
&x_height);
tem = x_get_arg (dpyinfo, parms, Qunsplittable, 0, 0, RES_TYPE_BOOLEAN);
- f->no_split = minibuffer_only || (!EQ (tem, Qunbound) && !EQ (tem, Qnil));
+ f->no_split = minibuffer_only || (!EQ (tem, Qunbound) && !NILP (tem));
/* NOTE: on other terms, this is done in set_mouse_color, however this
was not getting called under Nextstep. */
diff --git a/src/nsmenu.m b/src/nsmenu.m
index 4e22d7b..7010b77 100644
--- a/src/nsmenu.m
+++ b/src/nsmenu.m
@@ -828,7 +828,7 @@ ns_menu_show (struct frame *f, int x, int y, int menuflags,
i = 0;
while (i < menu_items_used)
{
- if (EQ (AREF (menu_items, i), Qnil))
+ if (NILP (AREF (menu_items, i)))
{
submenu_stack[submenu_depth++] = save_wv;
save_wv = prev_wv;
diff --git a/src/nsselect.m b/src/nsselect.m
index b7e134b..35705bf 100644
--- a/src/nsselect.m
+++ b/src/nsselect.m
@@ -164,7 +164,7 @@ ns_get_our_change_count_for (Lisp_Object selection)
static void
ns_string_to_pasteboard_internal (id pb, Lisp_Object str, NSString *gtype)
{
- if (EQ (str, Qnil))
+ if (NILP (str))
{
[pb declareTypes: [NSArray array] owner: nil];
}
@@ -399,7 +399,7 @@ these literal upper-case names.) The symbol nil is the
same as
return Qnil;
CHECK_SYMBOL (selection);
- if (EQ (selection, Qnil)) selection = QPRIMARY;
+ if (NILP (selection)) selection = QPRIMARY;
if (EQ (selection, Qt)) selection = QSECONDARY;
pb = ns_symbol_to_pb (selection);
if (pb == nil) return Qnil;
@@ -421,7 +421,7 @@ and t is the same as `SECONDARY'. */)
{
check_window_system (NULL);
CHECK_SYMBOL (selection);
- if (EQ (selection, Qnil)) selection = QPRIMARY;
+ if (NILP (selection)) selection = QPRIMARY;
if (EQ (selection, Qt)) selection = QSECONDARY;
return ns_get_pb_change_count (selection)
== ns_get_our_change_count_for (selection)
diff --git a/src/print.c b/src/print.c
index 998ff2d..3819c50 100644
--- a/src/print.c
+++ b/src/print.c
@@ -38,6 +38,11 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#include <c-ctype.h>
#include <float.h>
#include <ftoastr.h>
+#include <math.h>
+
+#if IEEE_FLOATING_POINT
+# include <ieee754.h>
+#endif
#ifdef WINDOWSNT
# include <sys/socket.h> /* for F_DUPFD_CLOEXEC */
@@ -1001,43 +1006,22 @@ float_to_string (char *buf, double data)
int width;
int len;
- /* Check for plus infinity in a way that won't lose
- if there is no plus infinity. */
- if (data == data / 2 && data > 1.0)
- {
- static char const infinity_string[] = "1.0e+INF";
- strcpy (buf, infinity_string);
- return sizeof infinity_string - 1;
- }
- /* Likewise for minus infinity. */
- if (data == data / 2 && data < -1.0)
+ if (isinf (data))
{
static char const minus_infinity_string[] = "-1.0e+INF";
- strcpy (buf, minus_infinity_string);
- return sizeof minus_infinity_string - 1;
+ bool positive = 0 < data;
+ strcpy (buf, minus_infinity_string + positive);
+ return sizeof minus_infinity_string - 1 - positive;
}
- /* Check for NaN in a way that won't fail if there are no NaNs. */
- if (! (data * 0.0 >= 0.0))
+#if IEEE_FLOATING_POINT
+ if (isnan (data))
{
- /* Prepend "-" if the NaN's sign bit is negative.
- The sign bit of a double is the bit that is 1 in -0.0. */
- static char const NaN_string[] = "0.0e+NaN";
- int i;
- union { double d; char c[sizeof (double)]; } u_data, u_minus_zero;
- bool negative = 0;
- u_data.d = data;
- u_minus_zero.d = - 0.0;
- for (i = 0; i < sizeof (double); i++)
- if (u_data.c[i] & u_minus_zero.c[i])
- {
- *buf = '-';
- negative = 1;
- break;
- }
-
- strcpy (buf + negative, NaN_string);
- return negative + sizeof NaN_string - 1;
+ union ieee754_double u = { .d = data };
+ uprintmax_t hi = u.ieee_nan.mantissa0;
+ return sprintf (buf, &"-%"pMu".0e+NaN"[!u.ieee_nan.negative],
+ (hi << 31 << 1) + u.ieee_nan.mantissa1);
}
+#endif
if (NILP (Vfloat_output_format)
|| !STRINGP (Vfloat_output_format))
diff --git a/src/process.c b/src/process.c
index 698a2c3..c8123be 100644
--- a/src/process.c
+++ b/src/process.c
@@ -3587,17 +3587,23 @@ connect_network_socket (Lisp_Object proc, Lisp_Object
addrinfos,
if (s < 0)
{
+ const char *err = (p->is_server
+ ? "make server process failed"
+ : "make client process failed");
+
/* If non-blocking got this far - and failed - assume non-blocking is
not supported after all. This is probably a wrong assumption, but
the normal blocking calls to open-network-stream handles this error
better. */
if (p->is_non_blocking_client)
- return;
+ {
+ Lisp_Object data = get_file_errno_data (err, contact, xerrno);
- report_file_errno ((p->is_server
- ? "make server process failed"
- : "make client process failed"),
- contact, xerrno);
+ pset_status (p, list2 (Fcar (data), Fcdr (data)));
+ return;
+ }
+
+ report_file_errno (err, contact, xerrno);
}
inch = s;
@@ -3905,7 +3911,7 @@ usage: (make-network-process &rest ARGS) */)
CHECK_STRING (name);
/* :local ADDRESS or :remote ADDRESS */
- if (!NILP (server))
+ if (NILP (server))
address = Fplist_get (contact, QCremote);
else
address = Fplist_get (contact, QClocal);
@@ -4608,16 +4614,15 @@ is nil, from any process) before the timeout expired.
*/)
/* Can't wait for a process that is dedicated to a different
thread. */
- if (!EQ (proc->thread, Qnil) && !EQ (proc->thread, Fcurrent_thread ()))
+ if (!NILP (proc->thread) && !EQ (proc->thread, Fcurrent_thread ()))
{
Lisp_Object proc_thread_name = XTHREAD (proc->thread)->name;
- if (STRINGP (proc_thread_name))
- error ("Attempt to accept output from process %s locked to thread
%s",
- SDATA (proc->name), SDATA (proc_thread_name));
- else
- error ("Attempt to accept output from process %s locked to thread
%p",
- SDATA (proc->name), XTHREAD (proc->thread));
+ error ("Attempt to accept output from process %s locked to thread %s",
+ SDATA (proc->name),
+ STRINGP (proc_thread_name)
+ ? SDATA (proc_thread_name)
+ : SDATA (Fprin1_to_string (proc->thread, Qt)));
}
}
else
@@ -5016,7 +5021,7 @@ wait_reading_process_output (intmax_t time_limit, int
nsecs, int read_kbd,
struct timespec now = invalid_timespec ();
eassert (wait_proc == NULL
- || EQ (wait_proc->thread, Qnil)
+ || NILP (wait_proc->thread)
|| XTHREAD (wait_proc->thread) == current_thread);
FD_ZERO (&Available);
diff --git a/src/regex.c b/src/regex-emacs.c
similarity index 65%
rename from src/regex.c
rename to src/regex-emacs.c
index 6ee13c4..d19838a 100644
--- a/src/regex.c
+++ b/src/regex-emacs.c
@@ -1,6 +1,4 @@
-/* Extended regular expression matching and search library, version
- 0.12. (Implements POSIX draft P1003.2/D11.2, except for some of the
- internationalization features.)
+/* Emacs regular expression matching and search
Copyright (C) 1993-2018 Free Software Foundation, Inc.
@@ -19,166 +17,64 @@
/* TODO:
- structure the opcode space into opcode+flag.
- - merge with glibc's regex.[ch].
- replace (succeed_n + jump_n + set_number_at) with something that doesn't
- need to modify the compiled regexp so that re_match can be reentrant.
+ need to modify the compiled regexp so that re_search can be reentrant.
- get rid of on_failure_jump_smart by doing the optimization in re_comp
- rather than at run-time, so that re_match can be reentrant.
+ rather than at run-time, so that re_search can be reentrant.
*/
-/* AIX requires this to be the first thing in the file. */
-#if defined _AIX && !defined REGEX_MALLOC
- #pragma alloca
-#endif
-
-/* Ignore some GCC warnings for now. This section should go away
- once the Emacs and Gnulib regex code is merged. */
-#if 4 < __GNUC__ + (5 <= __GNUC_MINOR__) || defined __clang__
-# pragma GCC diagnostic ignored "-Wstrict-overflow"
-# ifndef emacs
-# pragma GCC diagnostic ignored "-Wunused-function"
-# pragma GCC diagnostic ignored "-Wunused-macros"
-# pragma GCC diagnostic ignored "-Wunused-result"
-# pragma GCC diagnostic ignored "-Wunused-variable"
-# endif
-#endif
-
-#if 4 < __GNUC__ + (6 <= __GNUC_MINOR__) && ! defined __clang__
-# pragma GCC diagnostic ignored "-Wunused-but-set-variable"
-#endif
-
#include <config.h>
-#include <stddef.h>
-#include <stdlib.h>
-
-#ifdef emacs
-/* We need this for `regex.h', and perhaps for the Emacs include files. */
-# include <sys/types.h>
-#endif
-
-/* Whether to use ISO C Amendment 1 wide char functions.
- Those should not be used for Emacs since it uses its own. */
-#if defined _LIBC
-#define WIDE_CHAR_SUPPORT 1
-#else
-#define WIDE_CHAR_SUPPORT \
- (HAVE_WCTYPE_H && HAVE_WCHAR_H && HAVE_BTOWC && !emacs)
-#endif
+#include "regex-emacs.h"
-/* For platform which support the ISO C amendment 1 functionality we
- support user defined character classes. */
-#if WIDE_CHAR_SUPPORT
-/* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. */
-# include <wchar.h>
-# include <wctype.h>
-#endif
-
-#ifdef _LIBC
-/* We have to keep the namespace clean. */
-# define regfree(preg) __regfree (preg)
-# define regexec(pr, st, nm, pm, ef) __regexec (pr, st, nm, pm, ef)
-# define regcomp(preg, pattern, cflags) __regcomp (preg, pattern, cflags)
-# define regerror(err_code, preg, errbuf, errbuf_size) \
- __regerror (err_code, preg, errbuf, errbuf_size)
-# define re_set_registers(bu, re, nu, st, en) \
- __re_set_registers (bu, re, nu, st, en)
-# define re_match_2(bufp, string1, size1, string2, size2, pos, regs, stop) \
- __re_match_2 (bufp, string1, size1, string2, size2, pos, regs, stop)
-# define re_match(bufp, string, size, pos, regs) \
- __re_match (bufp, string, size, pos, regs)
-# define re_search(bufp, string, size, startpos, range, regs) \
- __re_search (bufp, string, size, startpos, range, regs)
-# define re_compile_pattern(pattern, length, bufp) \
- __re_compile_pattern (pattern, length, bufp)
-# define re_set_syntax(syntax) __re_set_syntax (syntax)
-# define re_search_2(bufp, st1, s1, st2, s2, startpos, range, regs, stop) \
- __re_search_2 (bufp, st1, s1, st2, s2, startpos, range, regs, stop)
-# define re_compile_fastmap(bufp) __re_compile_fastmap (bufp)
-
-/* Make sure we call libc's function even if the user overrides them. */
-# define btowc __btowc
-# define iswctype __iswctype
-# define wctype __wctype
-
-# define WEAK_ALIAS(a,b) weak_alias (a, b)
-
-/* We are also using some library internals. */
-# include <locale/localeinfo.h>
-# include <locale/elem-hash.h>
-# include <langinfo.h>
-#else
-# define WEAK_ALIAS(a,b)
-#endif
-
-/* This is for other GNU distributions with internationalized messages. */
-#if HAVE_LIBINTL_H || defined _LIBC
-# include <libintl.h>
-#else
-# define gettext(msgid) (msgid)
-#endif
+#include <stdlib.h>
-#ifndef gettext_noop
-/* This define is so xgettext can find the internationalizable
- strings. */
-# define gettext_noop(String) String
+#include "character.h"
+#include "buffer.h"
+#include "syntax.h"
+#include "category.h"
+
+/* Maximum number of duplicates an interval can allow. Some systems
+ define this in other header files, but we want our value, so remove
+ any previous define. Repeat counts are stored in opcodes as 2-byte
+ unsigned integers. */
+#ifdef RE_DUP_MAX
+# undef RE_DUP_MAX
#endif
-
-/* The `emacs' switch turns on certain matching commands
- that make sense only in Emacs. */
-#ifdef emacs
-
-# include "lisp.h"
-# include "character.h"
-# include "buffer.h"
-
-# include "syntax.h"
-# include "category.h"
+#define RE_DUP_MAX (0xffff)
/* Make syntax table lookup grant data in gl_state. */
-# define SYNTAX(c) syntax_property (c, 1)
-
-# ifdef malloc
-# undef malloc
-# endif
-# define malloc xmalloc
-# ifdef realloc
-# undef realloc
-# endif
-# define realloc xrealloc
-# ifdef free
-# undef free
-# endif
-# define free xfree
-
-/* Converts the pointer to the char to BEG-based offset from the start. */
-# define PTR_TO_OFFSET(d) POS_AS_IN_BUFFER (POINTER_TO_OFFSET (d))
-/* Strings are 0-indexed, buffers are 1-indexed; we pun on the boolean
+#define SYNTAX(c) syntax_property (c, 1)
+
+/* Convert the pointer to the char to BEG-based offset from the start. */
+#define PTR_TO_OFFSET(d) POS_AS_IN_BUFFER (POINTER_TO_OFFSET (d))
+/* Strings are 0-indexed, buffers are 1-indexed; pun on the boolean
result to get the right base index. */
-# define POS_AS_IN_BUFFER(p) \
+#define POS_AS_IN_BUFFER(p) \
((p) + (NILP (gl_state.object) || BUFFERP (gl_state.object)))
-# define RE_MULTIBYTE_P(bufp) ((bufp)->multibyte)
-# define RE_TARGET_MULTIBYTE_P(bufp) ((bufp)->target_multibyte)
-# define RE_STRING_CHAR(p, multibyte) \
- (multibyte ? (STRING_CHAR (p)) : (*(p)))
-# define RE_STRING_CHAR_AND_LENGTH(p, len, multibyte) \
- (multibyte ? (STRING_CHAR_AND_LENGTH (p, len)) : ((len) = 1, *(p)))
+#define RE_MULTIBYTE_P(bufp) ((bufp)->multibyte)
+#define RE_TARGET_MULTIBYTE_P(bufp) ((bufp)->target_multibyte)
+#define RE_STRING_CHAR(p, multibyte) \
+ (multibyte ? STRING_CHAR (p) : *(p))
+#define RE_STRING_CHAR_AND_LENGTH(p, len, multibyte) \
+ (multibyte ? STRING_CHAR_AND_LENGTH (p, len) : ((len) = 1, *(p)))
-# define RE_CHAR_TO_MULTIBYTE(c) UNIBYTE_TO_CHAR (c)
+#define RE_CHAR_TO_MULTIBYTE(c) UNIBYTE_TO_CHAR (c)
-# define RE_CHAR_TO_UNIBYTE(c) CHAR_TO_BYTE_SAFE (c)
+#define RE_CHAR_TO_UNIBYTE(c) CHAR_TO_BYTE_SAFE (c)
/* Set C a (possibly converted to multibyte) character before P. P
points into a string which is the virtual concatenation of STR1
(which ends at END1) or STR2 (which ends at END2). */
-# define GET_CHAR_BEFORE_2(c, p, str1, end1, str2, end2) \
+#define GET_CHAR_BEFORE_2(c, p, str1, end1, str2, end2)
\
do { \
if (target_multibyte) \
{
\
re_char *dtemp = (p) == (str2) ? (end1) : (p); \
- re_char *dlimit = ((p) > (str2) && (p) <= (end2)) ? (str2) : (str1); \
- while (dtemp-- > dlimit && !CHAR_HEAD_P (*dtemp)); \
+ re_char *dlimit = (p) > (str2) && (p) <= (end2) ? (str2) : (str1); \
+ while (dtemp-- > dlimit && !CHAR_HEAD_P (*dtemp)) \
+ continue; \
c = STRING_CHAR (dtemp); \
}
\
else \
@@ -186,11 +82,11 @@
(c = ((p) == (str2) ? (end1) : (p))[-1]); \
(c) = RE_CHAR_TO_MULTIBYTE (c); \
}
\
- } while (0)
+ } while (false)
/* Set C a (possibly converted to multibyte) character at P, and set
LEN to the byte length of that character. */
-# define GET_CHAR_AFTER(c, p, len) \
+#define GET_CHAR_AFTER(c, p, len) \
do { \
if (target_multibyte) \
(c) = STRING_CHAR_AND_LENGTH (p, len); \
@@ -200,336 +96,108 @@
len = 1; \
(c) = RE_CHAR_TO_MULTIBYTE (c); \
} \
- } while (0)
-
-#else /* not emacs */
-
-/* If we are not linking with Emacs proper,
- we can't use the relocating allocator
- even if config.h says that we can. */
-# undef REL_ALLOC
-
-# include <unistd.h>
-
-/* When used in Emacs's lib-src, we need xmalloc and xrealloc. */
-
-static ATTRIBUTE_MALLOC void *
-xmalloc (size_t size)
-{
- void *val = malloc (size);
- if (!val && size)
- {
- write (STDERR_FILENO, "virtual memory exhausted\n", 25);
- exit (1);
- }
- return val;
-}
-
-static void *
-xrealloc (void *block, size_t size)
-{
- void *val;
- /* We must call malloc explicitly when BLOCK is 0, since some
- reallocs don't do this. */
- if (! block)
- val = malloc (size);
- else
- val = realloc (block, size);
- if (!val && size)
- {
- write (STDERR_FILENO, "virtual memory exhausted\n", 25);
- exit (1);
- }
- return val;
-}
-
-# ifdef malloc
-# undef malloc
-# endif
-# define malloc xmalloc
-# ifdef realloc
-# undef realloc
-# endif
-# define realloc xrealloc
-
-# include <stdbool.h>
-# include <string.h>
-
-/* Define the syntax stuff for \<, \>, etc. */
-
-/* Sword must be nonzero for the wordchar pattern commands in re_match_2. */
-enum syntaxcode { Swhitespace = 0, Sword = 1, Ssymbol = 2 };
-
-/* Dummy macros for non-Emacs environments. */
-# define MAX_MULTIBYTE_LENGTH 1
-# define RE_MULTIBYTE_P(x) 0
-# define RE_TARGET_MULTIBYTE_P(x) 0
-# define WORD_BOUNDARY_P(c1, c2) (0)
-# define BYTES_BY_CHAR_HEAD(p) (1)
-# define PREV_CHAR_BOUNDARY(p, limit) ((p)--)
-# define STRING_CHAR(p) (*(p))
-# define RE_STRING_CHAR(p, multibyte) STRING_CHAR (p)
-# define CHAR_STRING(c, s) (*(s) = (c), 1)
-# define STRING_CHAR_AND_LENGTH(p, actual_len) ((actual_len) = 1, *(p))
-# define RE_STRING_CHAR_AND_LENGTH(p, len, multibyte) STRING_CHAR_AND_LENGTH
(p, len)
-# define RE_CHAR_TO_MULTIBYTE(c) (c)
-# define RE_CHAR_TO_UNIBYTE(c) (c)
-# define GET_CHAR_BEFORE_2(c, p, str1, end1, str2, end2) \
- (c = ((p) == (str2) ? *((end1) - 1) : *((p) - 1)))
-# define GET_CHAR_AFTER(c, p, len) \
- (c = *p, len = 1)
-# define CHAR_BYTE8_P(c) (0)
-# define CHAR_LEADING_CODE(c) (c)
-
-#endif /* not emacs */
-
-#ifndef RE_TRANSLATE
-# define RE_TRANSLATE(TBL, C) ((unsigned char)(TBL)[C])
-# define RE_TRANSLATE_P(TBL) (TBL)
-#endif
+ } while (false)
-/* Get the interface, including the syntax bits. */
-#include "regex.h"
-
-/* isalpha etc. are used for the character classes. */
-#include <ctype.h>
-
-#ifdef emacs
-
/* 1 if C is an ASCII character. */
-# define IS_REAL_ASCII(c) ((c) < 0200)
+#define IS_REAL_ASCII(c) ((c) < 0200)
/* 1 if C is a unibyte character. */
-# define ISUNIBYTE(c) (SINGLE_BYTE_CHAR_P ((c)))
+#define ISUNIBYTE(c) (SINGLE_BYTE_CHAR_P ((c)))
/* The Emacs definitions should not be directly affected by locales. */
/* In Emacs, these are only used for single-byte characters. */
-# define ISDIGIT(c) ((c) >= '0' && (c) <= '9')
-# define ISCNTRL(c) ((c) < ' ')
-# define ISXDIGIT(c) (0 <= char_hexdigit (c))
+#define ISDIGIT(c) ((c) >= '0' && (c) <= '9')
+#define ISCNTRL(c) ((c) < ' ')
+#define ISXDIGIT(c) (0 <= char_hexdigit (c))
/* The rest must handle multibyte characters. */
-# define ISBLANK(c) (IS_REAL_ASCII (c) \
+#define ISBLANK(c) (IS_REAL_ASCII (c) \
? ((c) == ' ' || (c) == '\t') \
: blankp (c))
-# define ISGRAPH(c) (SINGLE_BYTE_CHAR_P (c) \
+#define ISGRAPH(c) (SINGLE_BYTE_CHAR_P (c) \
? (c) > ' ' && !((c) >= 0177 && (c) <= 0240) \
: graphicp (c))
-# define ISPRINT(c) (SINGLE_BYTE_CHAR_P (c) \
+#define ISPRINT(c) (SINGLE_BYTE_CHAR_P (c) \
? (c) >= ' ' && !((c) >= 0177 && (c) <= 0237) \
: printablep (c))
-# define ISALNUM(c) (IS_REAL_ASCII (c) \
+#define ISALNUM(c) (IS_REAL_ASCII (c) \
? (((c) >= 'a' && (c) <= 'z') \
|| ((c) >= 'A' && (c) <= 'Z') \
|| ((c) >= '0' && (c) <= '9')) \
: alphanumericp (c))
-# define ISALPHA(c) (IS_REAL_ASCII (c) \
+#define ISALPHA(c) (IS_REAL_ASCII (c) \
? (((c) >= 'a' && (c) <= 'z') \
|| ((c) >= 'A' && (c) <= 'Z')) \
: alphabeticp (c))
-# define ISLOWER(c) lowercasep (c)
+#define ISLOWER(c) lowercasep (c)
-# define ISPUNCT(c) (IS_REAL_ASCII (c) \
+#define ISPUNCT(c) (IS_REAL_ASCII (c) \
? ((c) > ' ' && (c) < 0177 \
&& !(((c) >= 'a' && (c) <= 'z') \
|| ((c) >= 'A' && (c) <= 'Z') \
|| ((c) >= '0' && (c) <= '9'))) \
: SYNTAX (c) != Sword)
-# define ISSPACE(c) (SYNTAX (c) == Swhitespace)
+#define ISSPACE(c) (SYNTAX (c) == Swhitespace)
-# define ISUPPER(c) uppercasep (c)
+#define ISUPPER(c) uppercasep (c)
-# define ISWORD(c) (SYNTAX (c) == Sword)
-
-#else /* not emacs */
-
-/* 1 if C is an ASCII character. */
-# define IS_REAL_ASCII(c) ((c) < 0200)
-
-/* This distinction is not meaningful, except in Emacs. */
-# define ISUNIBYTE(c) 1
-
-# ifdef isblank
-# define ISBLANK(c) isblank (c)
-# else
-# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
-# endif
-# ifdef isgraph
-# define ISGRAPH(c) isgraph (c)
-# else
-# define ISGRAPH(c) (isprint (c) && !isspace (c))
-# endif
-
-/* Solaris defines ISPRINT so we must undefine it first. */
-# undef ISPRINT
-# define ISPRINT(c) isprint (c)
-# define ISDIGIT(c) isdigit (c)
-# define ISALNUM(c) isalnum (c)
-# define ISALPHA(c) isalpha (c)
-# define ISCNTRL(c) iscntrl (c)
-# define ISLOWER(c) islower (c)
-# define ISPUNCT(c) ispunct (c)
-# define ISSPACE(c) isspace (c)
-# define ISUPPER(c) isupper (c)
-# define ISXDIGIT(c) isxdigit (c)
-
-# define ISWORD(c) ISALPHA (c)
-
-# ifdef _tolower
-# define TOLOWER(c) _tolower (c)
-# else
-# define TOLOWER(c) tolower (c)
-# endif
-
-/* How many characters in the character set. */
-# define CHAR_SET_SIZE 256
-
-# ifdef SYNTAX_TABLE
-
-extern char *re_syntax_table;
-
-# else /* not SYNTAX_TABLE */
-
-static char re_syntax_table[CHAR_SET_SIZE];
-
-static void
-init_syntax_once (void)
-{
- register int c;
- static int done = 0;
-
- if (done)
- return;
-
- memset (re_syntax_table, 0, sizeof re_syntax_table);
-
- for (c = 0; c < CHAR_SET_SIZE; ++c)
- if (ISALNUM (c))
- re_syntax_table[c] = Sword;
-
- re_syntax_table['_'] = Ssymbol;
-
- done = 1;
-}
-
-# endif /* not SYNTAX_TABLE */
-
-# define SYNTAX(c) re_syntax_table[(c)]
-
-#endif /* not emacs */
+#define ISWORD(c) (SYNTAX (c) == Sword)
#define SIGN_EXTEND_CHAR(c) ((signed char) (c))
-/* Should we use malloc or alloca? If REGEX_MALLOC is not defined, we
- use `alloca' instead of `malloc'. This is because using malloc in
+/* Use alloca instead of malloc. This is because using malloc in
re_search* or re_match* could cause memory leaks when C-g is used
in Emacs (note that SAFE_ALLOCA could also call malloc, but does so
- via `record_xmalloc' which uses `unwind_protect' to ensure the
+ via 'record_xmalloc' which uses 'unwind_protect' to ensure the
memory is freed even in case of non-local exits); also, malloc is
slower and causes storage fragmentation. On the other hand, malloc
is more portable, and easier to debug.
Because we sometimes use alloca, some routines have to be macros,
- not functions -- `alloca'-allocated space disappears at the end of the
+ not functions -- 'alloca'-allocated space disappears at the end of the
function it is called in. */
-#ifdef REGEX_MALLOC
-
-# define REGEX_ALLOCATE malloc
-# define REGEX_REALLOCATE(source, osize, nsize) realloc (source, nsize)
-# define REGEX_FREE free
-
-#else /* not REGEX_MALLOC */
-
-# ifdef emacs
/* This may be adjusted in main(), if the stack is successfully grown. */
ptrdiff_t emacs_re_safe_alloca = MAX_ALLOCA;
/* Like USE_SAFE_ALLOCA, but use emacs_re_safe_alloca. */
-# define REGEX_USE_SAFE_ALLOCA \
- ptrdiff_t sa_avail = emacs_re_safe_alloca; \
- ptrdiff_t sa_count = SPECPDL_INDEX ()
-
-# define REGEX_SAFE_FREE() SAFE_FREE ()
-# define REGEX_ALLOCATE SAFE_ALLOCA
-# else
-# include <alloca.h>
-# define REGEX_ALLOCATE alloca
-# endif
-
-/* Assumes a `char *destination' variable. */
-# define REGEX_REALLOCATE(source, osize, nsize)
\
- (destination = REGEX_ALLOCATE (nsize), \
- memcpy (destination, source, osize))
-
-/* No need to do anything to free, after alloca. */
-# define REGEX_FREE(arg) ((void)0) /* Do nothing! But inhibit gcc warning. */
-
-#endif /* not REGEX_MALLOC */
-
-#ifndef REGEX_USE_SAFE_ALLOCA
-# define REGEX_USE_SAFE_ALLOCA ((void) 0)
-# define REGEX_SAFE_FREE() ((void) 0)
-#endif
-
-/* Define how to allocate the failure stack. */
-
-#if defined REL_ALLOC && defined REGEX_MALLOC
-
-# define REGEX_ALLOCATE_STACK(size) \
- r_alloc (&failure_stack_ptr, (size))
-# define REGEX_REALLOCATE_STACK(source, osize, nsize) \
- r_re_alloc (&failure_stack_ptr, (nsize))
-# define REGEX_FREE_STACK(ptr) \
- r_alloc_free (&failure_stack_ptr)
-
-#else /* not using relocating allocator */
-
-# define REGEX_ALLOCATE_STACK(size) REGEX_ALLOCATE (size)
-# define REGEX_REALLOCATE_STACK(source, o, n) REGEX_REALLOCATE (source, o, n)
-# define REGEX_FREE_STACK(ptr) REGEX_FREE (ptr)
-
-#endif /* not using relocating allocator */
+#define REGEX_USE_SAFE_ALLOCA \
+ USE_SAFE_ALLOCA; sa_avail = emacs_re_safe_alloca
+/* Assumes a 'char *destination' variable. */
+#define REGEX_REALLOCATE(source, osize, nsize) \
+ (destination = SAFE_ALLOCA (nsize), \
+ memcpy (destination, source, osize))
-/* True if `size1' is non-NULL and PTR is pointing anywhere inside
- `string1' or just past its end. This works if PTR is NULL, which is
+/* True if 'size1' is non-NULL and PTR is pointing anywhere inside
+ 'string1' or just past its end. This works if PTR is NULL, which is
a good thing. */
#define FIRST_STRING_P(ptr) \
(size1 && string1 <= (ptr) && (ptr) <= string1 + size1)
/* (Re)Allocate N items of type T using malloc, or fail. */
-#define TALLOC(n, t) ((t *) malloc ((n) * sizeof (t)))
-#define RETALLOC(addr, n, t) ((addr) = (t *) realloc (addr, (n) * sizeof (t)))
-#define REGEX_TALLOC(n, t) ((t *) REGEX_ALLOCATE ((n) * sizeof (t)))
+#define TALLOC(n, t) ((t *) xmalloc ((n) * sizeof (t)))
+#define RETALLOC(addr, n, t) ((addr) = (t *) xrealloc (addr, (n) * sizeof (t)))
#define BYTEWIDTH 8 /* In bits. */
-#ifndef emacs
-# undef max
-# undef min
-# define max(a, b) ((a) > (b) ? (a) : (b))
-# define min(a, b) ((a) < (b) ? (a) : (b))
-#endif
-
/* Type of source-pattern and string chars. */
typedef const unsigned char re_char;
-typedef char boolean;
-
-static regoff_t re_match_2_internal (struct re_pattern_buffer *bufp,
+static void re_compile_fastmap (struct re_pattern_buffer *);
+static ptrdiff_t re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1, size_t size1,
re_char *string2, size_t size2,
- ssize_t pos,
+ ptrdiff_t pos,
struct re_registers *regs,
- ssize_t stop);
+ ptrdiff_t stop);
/* These are the command codes that appear in compiled regular
expressions. Some opcodes are followed by argument bytes. A
@@ -577,7 +245,7 @@ typedef enum
/* Stop remembering the text that is matched and store it in a
memory register. Followed by one byte with the register
- number, in the range 0 to one less than `re_nsub' in the
+ number, in the range 0 to one less than 're_nsub' in the
pattern buffer. */
stop_memory,
@@ -591,8 +259,7 @@ typedef enum
/* Fail unless at end of line. */
endline,
- /* Succeeds if at beginning of buffer (if emacs) or at beginning
- of string to be matched (if not). */
+ /* Succeeds if at beginning of buffer. */
begbuf,
/* Analogously, for end of buffer/string. */
@@ -609,23 +276,23 @@ typedef enum
current string position when executed. */
on_failure_keep_string_jump,
- /* Just like `on_failure_jump', except that it checks that we
+ /* Just like 'on_failure_jump', except that it checks that we
don't get stuck in an infinite loop (matching an empty string
indefinitely). */
on_failure_jump_loop,
- /* Just like `on_failure_jump_loop', except that it checks for
+ /* Just like 'on_failure_jump_loop', except that it checks for
a different kind of loop (the kind that shows up with non-greedy
operators). This operation has to be immediately preceded
- by a `no_op'. */
+ by a 'no_op'. */
on_failure_jump_nastyloop,
- /* A smart `on_failure_jump' used for greedy * and + operators.
+ /* A smart 'on_failure_jump' used for greedy * and + operators.
It analyzes the loop before which it is put and if the
loop does not require backtracking, it changes itself to
- `on_failure_keep_string_jump' and short-circuits the loop,
- else it just defaults to changing itself into `on_failure_jump'.
- It assumes that it is pointing to just past a `jump'. */
+ 'on_failure_keep_string_jump' and short-circuits the loop,
+ else it just defaults to changing itself into 'on_failure_jump'.
+ It assumes that it is pointing to just past a 'jump'. */
on_failure_jump_smart,
/* Followed by two-byte relative address and two-byte number n.
@@ -657,10 +324,9 @@ typedef enum
syntaxspec,
/* Matches any character whose syntax is not that specified. */
- notsyntaxspec
+ notsyntaxspec,
-#ifdef emacs
- , at_dot, /* Succeeds if at point. */
+ at_dot, /* Succeeds if at point. */
/* Matches any character whose category-set contains the specified
category. The operator is followed by a byte which contains a
@@ -671,7 +337,6 @@ typedef enum
specified category. The operator is followed by a byte which
contains the category code (mnemonic ASCII character). */
notcategoryspec
-#endif /* emacs */
} re_opcode_t;
/* Common operations on the compiled pattern. */
@@ -682,7 +347,7 @@ typedef enum
do { \
(destination)[0] = (number) & 0377;
\
(destination)[1] = (number) >> 8; \
- } while (0)
+ } while (false)
/* Same as STORE_NUMBER, except increment DESTINATION to
the byte after where the number is stored. Therefore, DESTINATION
@@ -692,7 +357,7 @@ typedef enum
do { \
STORE_NUMBER (destination, number);
\
(destination) += 2;
\
- } while (0)
+ } while (false)
/* Put into DESTINATION a number stored in two contiguous bytes starting
at SOURCE. */
@@ -731,7 +396,7 @@ extract_number_and_incr (re_char **source)
(destination)[1] = ((character) >> 8) & 0377; \
(destination)[2] = (character) >> 16; \
(destination) += 3; \
- } while (0)
+ } while (false)
/* Put into DESTINATION a character stored in three contiguous bytes
starting at SOURCE. */
@@ -741,7 +406,7 @@ extract_number_and_incr (re_char **source)
(destination) = ((source)[0] \
| ((source)[1] << 8) \
| ((source)[2] << 16)); \
- } while (0)
+ } while (false)
/* Macros for charset. */
@@ -755,47 +420,39 @@ extract_number_and_incr (re_char **source)
/* Return the address of range table of charset P. But not the start
of table itself, but the before where the number of ranges is
- stored. `2 +' means to skip re_opcode_t and size of bitmap,
+ stored. '2 +' means to skip re_opcode_t and size of bitmap,
and the 2 bytes of flags at the start of the range table. */
#define CHARSET_RANGE_TABLE(p) (&(p)[4 + CHARSET_BITMAP_SIZE (p)])
-#ifdef emacs
/* Extract the bit flags that start a range table. */
#define CHARSET_RANGE_TABLE_BITS(p) \
((p)[2 + CHARSET_BITMAP_SIZE (p)] \
+ (p)[3 + CHARSET_BITMAP_SIZE (p)] * 0x100)
-#endif
/* Return the address of end of RANGE_TABLE. COUNT is number of
- ranges (which is a pair of (start, end)) in the RANGE_TABLE. `* 2'
- is start of range and end of range. `* 3' is size of each start
+ ranges (which is a pair of (start, end)) in the RANGE_TABLE. '* 2'
+ is start of range and end of range. '* 3' is size of each start
and end. */
#define CHARSET_RANGE_TABLE_END(range_table, count) \
((range_table) + (count) * 2 * 3)
-/* If DEBUG is defined, Regex prints many voluminous messages about what
- it is doing (if the variable `debug' is nonzero). If linked with the
- main program in `iregex.c', you can enter patterns and strings
- interactively. And if linked with the main program in `main.c' and
- the other test files, you can run the already-written tests. */
+/* If REGEX_EMACS_DEBUG is defined, print many voluminous messages
+ (if the variable regex_emacs_debug is positive). */
-#ifdef DEBUG
+#ifdef REGEX_EMACS_DEBUG
-/* We use standard I/O for debugging. */
+/* Use standard I/O for debugging. */
# include <stdio.h>
-/* It is useful to test things that ``must'' be true when debugging. */
-# include <assert.h>
-
-static int debug = -100000;
+static int regex_emacs_debug = -100000;
# define DEBUG_STATEMENT(e) e
-# define DEBUG_PRINT(...) if (debug > 0) printf (__VA_ARGS__)
+# define DEBUG_PRINT(...) if (regex_emacs_debug > 0) printf (__VA_ARGS__)
# define DEBUG_COMPILES_ARGUMENTS
# define DEBUG_PRINT_COMPILED_PATTERN(p, s, e) \
- if (debug > 0) print_partial_compiled_pattern (s, e)
+ if (regex_emacs_debug > 0) print_partial_compiled_pattern (s, e)
# define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2)
\
- if (debug > 0) print_double_string (w, s1, sz1, s2, sz2)
+ if (regex_emacs_debug > 0) print_double_string (w, s1, sz1, s2, sz2)
/* Print the fastmap in human-readable form. */
@@ -1084,7 +741,7 @@ print_compiled_pattern (struct re_pattern_buffer *bufp)
re_char *buffer = bufp->buffer;
print_partial_compiled_pattern (buffer, buffer + bufp->used);
- printf ("%ld bytes used/%ld bytes allocated.\n",
+ printf ("%zu bytes used/%zu bytes allocated.\n",
bufp->used, bufp->allocated);
if (bufp->fastmap_accurate && bufp->fastmap)
@@ -1096,9 +753,6 @@ print_compiled_pattern (struct re_pattern_buffer *bufp)
printf ("re_nsub: %zu\t", bufp->re_nsub);
printf ("regs_alloc: %d\t", bufp->regs_allocated);
printf ("can_be_null: %d\t", bufp->can_be_null);
- printf ("no_sub: %d\t", bufp->no_sub);
- printf ("not_bol: %d\t", bufp->not_bol);
- printf ("not_eol: %d\t", bufp->not_eol);
#ifndef emacs
printf ("syntax: %lx\n", bufp->syntax);
#endif
@@ -1130,151 +784,105 @@ print_double_string (re_char *where, re_char *string1,
ssize_t size1,
}
}
-#else /* not DEBUG */
-
-# undef assert
-# define assert(e)
+#else /* not REGEX_EMACS_DEBUG */
# define DEBUG_STATEMENT(e)
# define DEBUG_PRINT(...)
# define DEBUG_PRINT_COMPILED_PATTERN(p, s, e)
# define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2)
-#endif /* not DEBUG */
+#endif /* not REGEX_EMACS_DEBUG */
-#ifndef emacs
-
-/* Set by `re_set_syntax' to the current regexp syntax to recognize. Can
- also be assigned to arbitrarily: each pattern buffer stores its own
- syntax, so it can be changed between regex compilations. */
-/* This has no initializer because initialized variables in Emacs
- become read-only after dumping. */
-reg_syntax_t re_syntax_options;
-
-
-/* Specify the precise syntax of regexps for compilation. This provides
- for compatibility for various utilities which historically have
- different, incompatible syntaxes.
-
- The argument SYNTAX is a bit mask comprised of the various bits
- defined in regex.h. We return the old syntax. */
-
-reg_syntax_t
-re_set_syntax (reg_syntax_t syntax)
+typedef enum
{
- reg_syntax_t ret = re_syntax_options;
-
- re_syntax_options = syntax;
- return ret;
-}
-WEAK_ALIAS (__re_set_syntax, re_set_syntax)
-
-#endif
-
-/* This table gives an error message for each of the error codes listed
- in regex.h. Obviously the order here has to be same as there.
- POSIX doesn't require that we do anything for REG_NOERROR,
- but why not be nice? */
+ REG_NOERROR = 0, /* Success. */
+ REG_NOMATCH, /* Didn't find a match (for regexec). */
+
+ /* POSIX regcomp return error codes. (In the order listed in the
+ standard.) An older version of this code supported the POSIX
+ API; this version continues to use these names internally. */
+ REG_BADPAT, /* Invalid pattern. */
+ REG_ECOLLATE, /* Not implemented. */
+ REG_ECTYPE, /* Invalid character class name. */
+ REG_EESCAPE, /* Trailing backslash. */
+ REG_ESUBREG, /* Invalid back reference. */
+ REG_EBRACK, /* Unmatched left bracket. */
+ REG_EPAREN, /* Parenthesis imbalance. */
+ REG_EBRACE, /* Unmatched \{. */
+ REG_BADBR, /* Invalid contents of \{\}. */
+ REG_ERANGE, /* Invalid range end. */
+ REG_ESPACE, /* Ran out of memory. */
+ REG_BADRPT, /* No preceding re for repetition op. */
+
+ /* Error codes we've added. */
+ REG_EEND, /* Premature end. */
+ REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
+ REG_ERPAREN, /* Unmatched ) or \); not returned from regcomp. */
+ REG_ERANGEX, /* Range striding over charsets. */
+ REG_ESIZEBR /* n or m too big in \{n,m\} */
+} reg_errcode_t;
static const char *re_error_msgid[] =
{
- gettext_noop ("Success"), /* REG_NOERROR */
- gettext_noop ("No match"), /* REG_NOMATCH */
- gettext_noop ("Invalid regular expression"), /* REG_BADPAT */
- gettext_noop ("Invalid collation character"), /* REG_ECOLLATE */
- gettext_noop ("Invalid character class name"), /* REG_ECTYPE */
- gettext_noop ("Trailing backslash"), /* REG_EESCAPE */
- gettext_noop ("Invalid back reference"), /* REG_ESUBREG */
- gettext_noop ("Unmatched [ or [^"), /* REG_EBRACK */
- gettext_noop ("Unmatched ( or \\("), /* REG_EPAREN */
- gettext_noop ("Unmatched \\{"), /* REG_EBRACE */
- gettext_noop ("Invalid content of \\{\\}"), /* REG_BADBR */
- gettext_noop ("Invalid range end"), /* REG_ERANGE */
- gettext_noop ("Memory exhausted"), /* REG_ESPACE */
- gettext_noop ("Invalid preceding regular expression"), /* REG_BADRPT */
- gettext_noop ("Premature end of regular expression"), /* REG_EEND */
- gettext_noop ("Regular expression too big"), /* REG_ESIZE */
- gettext_noop ("Unmatched ) or \\)"), /* REG_ERPAREN */
- gettext_noop ("Range striding over charsets"), /* REG_ERANGEX */
- gettext_noop ("Invalid content of \\{\\}, repetitions too big") /*
REG_ESIZEBR */
+ [REG_NOERROR] = "Success",
+ [REG_NOMATCH] = "No match",
+ [REG_BADPAT] = "Invalid regular expression",
+ [REG_ECOLLATE] = "Invalid collation character",
+ [REG_ECTYPE] = "Invalid character class name",
+ [REG_EESCAPE] = "Trailing backslash",
+ [REG_ESUBREG] = "Invalid back reference",
+ [REG_EBRACK] = "Unmatched [ or [^",
+ [REG_EPAREN] = "Unmatched ( or \\(",
+ [REG_EBRACE] = "Unmatched \\{",
+ [REG_BADBR] = "Invalid content of \\{\\}",
+ [REG_ERANGE] = "Invalid range end",
+ [REG_ESPACE] = "Memory exhausted",
+ [REG_BADRPT] = "Invalid preceding regular expression",
+ [REG_EEND] = "Premature end of regular expression",
+ [REG_ESIZE] = "Regular expression too big",
+ [REG_ERPAREN] = "Unmatched ) or \\)",
+ [REG_ERANGEX ] = "Range striding over charsets",
+ [REG_ESIZEBR ] = "Invalid content of \\{\\}",
};
-
-/* Whether to allocate memory during matching. */
-
-/* Define MATCH_MAY_ALLOCATE to allow the searching and matching
- functions allocate memory for the failure stack and registers.
- Normally should be defined, because otherwise searching and
- matching routines will have much smaller memory resources at their
- disposal, and therefore might fail to handle complex regexps.
- Therefore undefine MATCH_MAY_ALLOCATE only in the following
- exceptional situations:
-
- . When running on a system where memory is at premium.
- . When alloca cannot be used at all, perhaps due to bugs in
- its implementation, or its being unavailable, or due to a
- very small stack size. This requires to define REGEX_MALLOC
- to use malloc instead, which in turn could lead to memory
- leaks if search is interrupted by a signal. (For these
- reasons, defining REGEX_MALLOC when building Emacs
- automatically undefines MATCH_MAY_ALLOCATE, but outside
- Emacs you may not care about memory leaks.) If you want to
- prevent the memory leaks, undefine MATCH_MAY_ALLOCATE.
- . When code that calls the searching and matching functions
- cannot allow memory allocation, for whatever reasons. */
-
-/* Normally, this is fine. */
-#define MATCH_MAY_ALLOCATE
-
-/* The match routines may not allocate if (1) they would do it with malloc
- and (2) it's not safe for them to use malloc.
- Note that if REL_ALLOC is defined, matching would not use malloc for the
- failure stack, but we would still use it for the register vectors;
- so REL_ALLOC should not affect this. */
-#if defined REGEX_MALLOC && defined emacs
-# undef MATCH_MAY_ALLOCATE
-#endif
-/* While regex matching of a single compiled pattern isn't reentrant
- (because we compile regexes to bytecode programs, and the bytecode
- programs are self-modifying), the regex machinery must nevertheless
- be reentrant with respect to _different_ patterns, and we do that
- by avoiding global variables and using MATCH_MAY_ALLOCATE. */
-#if !defined MATCH_MAY_ALLOCATE && defined emacs
-# error "Emacs requires MATCH_MAY_ALLOCATE"
-#endif
+/* For 'regs_allocated'. */
+enum { REGS_UNALLOCATED, REGS_REALLOCATE, REGS_FIXED };
+/* If 'regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
+ 're_match_2' returns information about at least this many registers
+ the first time a 'regs' structure is passed. */
+enum { RE_NREGS = 30 };
+/* The searching and matching functions allocate memory for the
+ failure stack and registers. Otherwise searching and matching
+ routines would have much smaller memory resources at their
+ disposal, and therefore might fail to handle complex regexps. */
+
/* Failure stack declarations and macros; both re_compile_fastmap and
re_match_2 use a failure stack. These have to be macros because of
- REGEX_ALLOCATE_STACK. */
+ SAFE_ALLOCA. */
/* Approximate number of failure points for which to initially allocate space
when matching. If this number is exceeded, we allocate more
space, so it is not a hard limit. */
-#ifndef INIT_FAILURE_ALLOC
-# define INIT_FAILURE_ALLOC 20
-#endif
+#define INIT_FAILURE_ALLOC 20
/* Roughly the maximum number of failure points on the stack. Would be
- exactly that if always used TYPICAL_FAILURE_SIZE items each time we failed.
+ exactly that if failure always used TYPICAL_FAILURE_SIZE items.
This is a variable only so users of regex can assign to it; we never
change it ourselves. We always multiply it by TYPICAL_FAILURE_SIZE
before using it, so it should probably be a byte-count instead. */
-# if defined MATCH_MAY_ALLOCATE
/* Note that 4400 was enough to cause a crash on Alpha OSF/1,
whose default stack limit is 2mb. In order for a larger
value to work reliably, you have to try to make it accord
with the process stack limit. */
size_t emacs_re_max_failures = 40000;
-# else
-size_t emacs_re_max_failures = 4000;
-# endif
union fail_stack_elt
{
re_char *pointer;
- /* This should be the biggest `int' that's no bigger than a pointer. */
+ /* This should be the biggest 'int' that's no bigger than a pointer. */
long integer;
};
@@ -1291,45 +899,28 @@ typedef struct
#define FAIL_STACK_EMPTY() (fail_stack.frame == 0)
-/* Define macros to initialize and free the failure stack.
- Do `return -2' if the alloc fails. */
+/* Define macros to initialize and free the failure stack. */
-#ifdef MATCH_MAY_ALLOCATE
-# define INIT_FAIL_STACK() \
+#define INIT_FAIL_STACK() \
do { \
fail_stack.stack = \
- REGEX_ALLOCATE_STACK (INIT_FAILURE_ALLOC * TYPICAL_FAILURE_SIZE \
- * sizeof (fail_stack_elt_t)); \
- \
- if (fail_stack.stack == NULL) \
- return -2; \
- \
+ SAFE_ALLOCA (INIT_FAILURE_ALLOC * TYPICAL_FAILURE_SIZE \
+ * sizeof (fail_stack_elt_t)); \
fail_stack.size = INIT_FAILURE_ALLOC; \
fail_stack.avail = 0; \
fail_stack.frame = 0; \
- } while (0)
-#else
-# define INIT_FAIL_STACK() \
- do { \
- fail_stack.avail = 0; \
- fail_stack.frame = 0; \
- } while (0)
-
-# define RETALLOC_IF(addr, n, t) \
- if (addr) RETALLOC((addr), (n), t); else (addr) = TALLOC ((n), t)
-#endif
+ } while (false)
/* Double the size of FAIL_STACK, up to a limit
- which allows approximately `emacs_re_max_failures' items.
+ which allows approximately 'emacs_re_max_failures' items.
Return 1 if succeeds, and 0 if either ran out of memory
allocating space for it or it was already too large.
- REGEX_REALLOCATE_STACK requires `destination' be declared. */
+ REGEX_REALLOCATE requires 'destination' be declared. */
-/* Factor to increase the failure stack size by
- when we increase it.
+/* Factor to increase the failure stack size by.
This used to be 2, but 2 was too wasteful
because the old discarded stacks added up to as much space
were as ultimate, maximum-size stack. */
@@ -1339,34 +930,31 @@ typedef struct
(((fail_stack).size >= emacs_re_max_failures * TYPICAL_FAILURE_SIZE) \
? 0 \
: ((fail_stack).stack \
- = REGEX_REALLOCATE_STACK ((fail_stack).stack, \
+ = REGEX_REALLOCATE ((fail_stack).stack, \
(fail_stack).size * sizeof (fail_stack_elt_t), \
min (emacs_re_max_failures * TYPICAL_FAILURE_SIZE, \
((fail_stack).size * FAIL_STACK_GROWTH_FACTOR)) \
* sizeof (fail_stack_elt_t)), \
- \
- (fail_stack).stack == NULL \
- ? 0 \
- : ((fail_stack).size \
- = (min (emacs_re_max_failures * TYPICAL_FAILURE_SIZE, \
- ((fail_stack).size * FAIL_STACK_GROWTH_FACTOR))), \
- 1)))
+ ((fail_stack).size \
+ = (min (emacs_re_max_failures * TYPICAL_FAILURE_SIZE, \
+ ((fail_stack).size * FAIL_STACK_GROWTH_FACTOR)))), \
+ 1))
/* Push a pointer value onto the failure stack.
- Assumes the variable `fail_stack'. Probably should only
- be called from within `PUSH_FAILURE_POINT'. */
+ Assumes the variable 'fail_stack'. Probably should only
+ be called from within 'PUSH_FAILURE_POINT'. */
#define PUSH_FAILURE_POINTER(item) \
fail_stack.stack[fail_stack.avail++].pointer = (item)
/* This pushes an integer-valued item onto the failure stack.
- Assumes the variable `fail_stack'. Probably should only
- be called from within `PUSH_FAILURE_POINT'. */
+ Assumes the variable 'fail_stack'. Probably should only
+ be called from within 'PUSH_FAILURE_POINT'. */
#define PUSH_FAILURE_INT(item) \
fail_stack.stack[fail_stack.avail++].integer = (item)
/* These POP... operations complement the PUSH... operations.
- All assume that `fail_stack' is nonempty. */
+ All assume that 'fail_stack' is nonempty. */
#define POP_FAILURE_POINTER() fail_stack.stack[--fail_stack.avail].pointer
#define POP_FAILURE_INT() fail_stack.stack[--fail_stack.avail].integer
@@ -1384,8 +972,8 @@ typedef struct
while (REMAINING_AVAIL_SLOTS <= space) { \
if (!GROW_FAIL_STACK (fail_stack)) \
return -2; \
- DEBUG_PRINT ("\n Doubled stack; size now: %zd\n", (fail_stack).size);\
- DEBUG_PRINT (" slots available: %zd\n", REMAINING_AVAIL_SLOTS);\
+ DEBUG_PRINT ("\n Doubled stack; size now: %zu\n", (fail_stack).size);\
+ DEBUG_PRINT (" slots available: %zu\n", REMAINING_AVAIL_SLOTS);\
}
/* Push register NUM onto the stack. */
@@ -1399,7 +987,7 @@ do {
\
PUSH_FAILURE_POINTER (regstart[n]); \
PUSH_FAILURE_POINTER (regend[n]); \
PUSH_FAILURE_INT (n);
\
-} while (0)
+} while (false)
/* Change the counter's value to VAL, but make sure that it will
be reset when backtracking. */
@@ -1414,7 +1002,7 @@ do {
\
PUSH_FAILURE_POINTER (ptr); \
PUSH_FAILURE_INT (-1); \
STORE_NUMBER (ptr, val); \
-} while (0)
+} while (false)
/* Pop a saved register off the stack. */
#define POP_FAILURE_REG_OR_COUNT() \
@@ -1423,7 +1011,7 @@ do {
\
if (pfreg == -1) \
{ \
/* It's a counter. */ \
- /* Here, we discard `const', making re_match non-reentrant. */ \
+ /* Discard 'const', making re_search non-reentrant. */ \
unsigned char *ptr = (unsigned char *) POP_FAILURE_POINTER (); \
pfreg = POP_FAILURE_INT (); \
STORE_NUMBER (ptr, pfreg); \
@@ -1436,19 +1024,19 @@ do {
\
DEBUG_PRINT (" Pop reg %ld (spanning %p -> %p)\n", \
pfreg, regstart[pfreg], regend[pfreg]); \
} \
-} while (0)
+} while (false)
/* Check that we are not stuck in an infinite loop. */
#define CHECK_INFINITE_LOOP(pat_cur, string_place) \
do { \
- ssize_t failure = TOP_FAILURE_HANDLE (); \
+ ptrdiff_t failure = TOP_FAILURE_HANDLE (); \
/* Check for infinite matching loops */ \
while (failure > 0 \
&& (FAILURE_STR (failure) == string_place \
|| FAILURE_STR (failure) == NULL)) \
{ \
- assert (FAILURE_PAT (failure) >= bufp->buffer \
- && FAILURE_PAT (failure) <= bufp->buffer + bufp->used); \
+ eassert (FAILURE_PAT (failure) >= bufp->buffer \
+ && FAILURE_PAT (failure) <= bufp->buffer + bufp->used); \
if (FAILURE_PAT (failure) == pat_cur) \
{ \
cycle = 1; \
@@ -1458,47 +1046,44 @@ do {
\
failure = NEXT_FAILURE_HANDLE(failure); \
} \
DEBUG_PRINT (" Other string: %p\n", FAILURE_STR (failure)); \
-} while (0)
+} while (false)
/* Push the information about the state we will need
if we ever fail back to it.
Requires variables fail_stack, regstart, regend and
- num_regs be declared. GROW_FAIL_STACK requires `destination' be
+ num_regs be declared. GROW_FAIL_STACK requires 'destination' be
declared.
- Does `return FAILURE_CODE' if runs out of memory. */
+ Does 'return FAILURE_CODE' if runs out of memory. */
#define PUSH_FAILURE_POINT(pattern, string_place) \
do { \
char *destination; \
- /* Must be int, so when we don't save any registers, the arithmetic \
- of 0 + -1 isn't done as unsigned. */ \
- \
DEBUG_STATEMENT (nfailure_points_pushed++); \
DEBUG_PRINT ("\nPUSH_FAILURE_POINT:\n"); \
- DEBUG_PRINT (" Before push, next avail: %zd\n", (fail_stack).avail);
\
- DEBUG_PRINT (" size: %zd\n", (fail_stack).size);\
- \
+ DEBUG_PRINT (" Before push, next avail: %zu\n", (fail_stack).avail);
\
+ DEBUG_PRINT (" size: %zu\n", (fail_stack).size);\
+ \
ENSURE_FAIL_STACK (NUM_NONREG_ITEMS);
\
- \
+ \
DEBUG_PRINT ("\n"); \
- \
- DEBUG_PRINT (" Push frame index: %zd\n", fail_stack.frame); \
+ \
+ DEBUG_PRINT (" Push frame index: %zu\n", fail_stack.frame); \
PUSH_FAILURE_INT (fail_stack.frame); \
- \
+ \
DEBUG_PRINT (" Push string %p: \"", string_place); \
DEBUG_PRINT_DOUBLE_STRING (string_place, string1, size1, string2, size2);\
DEBUG_PRINT ("\"\n");
\
PUSH_FAILURE_POINTER (string_place); \
- \
+ \
DEBUG_PRINT (" Push pattern %p: ", pattern);
\
DEBUG_PRINT_COMPILED_PATTERN (bufp, pattern, pend); \
PUSH_FAILURE_POINTER (pattern); \
- \
+ \
/* Close the frame by moving the frame pointer past it. */ \
fail_stack.frame = fail_stack.avail; \
-} while (0)
+} while (false)
/* Estimate the size of data pushed by a typical failure stack entry.
An estimate is all we need, because all we use this for
@@ -1510,24 +1095,24 @@ do {
\
#define REMAINING_AVAIL_SLOTS ((fail_stack).size - (fail_stack).avail)
-/* Pops what PUSH_FAIL_STACK pushes.
+/* Pop what PUSH_FAIL_STACK pushes.
- We restore into the parameters, all of which should be lvalues:
+ Restore into the parameters, all of which should be lvalues:
STR -- the saved data position.
PAT -- the saved pattern position.
REGSTART, REGEND -- arrays of string positions.
- Also assumes the variables `fail_stack' and (if debugging), `bufp',
- `pend', `string1', `size1', `string2', and `size2'. */
+ Also assume the variables FAIL_STACK and (if debugging) BUFP, PEND,
+ STRING1, SIZE1, STRING2, and SIZE2. */
#define POP_FAILURE_POINT(str, pat) \
do { \
- assert (!FAIL_STACK_EMPTY ()); \
+ eassert (!FAIL_STACK_EMPTY ()); \
\
/* Remove failure points and point to how many regs pushed. */ \
DEBUG_PRINT ("POP_FAILURE_POINT:\n");
\
- DEBUG_PRINT (" Before pop, next avail: %zd\n", fail_stack.avail); \
- DEBUG_PRINT (" size: %zd\n", fail_stack.size); \
+ DEBUG_PRINT (" Before pop, next avail: %zu\n", fail_stack.avail); \
+ DEBUG_PRINT (" size: %zu\n", fail_stack.size); \
\
/* Pop the saved registers. */ \
while (fail_stack.frame < fail_stack.avail) \
@@ -1546,13 +1131,13 @@ do {
\
DEBUG_PRINT ("\"\n");
\
\
fail_stack.frame = POP_FAILURE_INT (); \
- DEBUG_PRINT (" Popping frame index: %zd\n", fail_stack.frame); \
+ DEBUG_PRINT (" Popping frame index: %zu\n", fail_stack.frame); \
\
- assert (fail_stack.avail >= 0); \
- assert (fail_stack.frame <= fail_stack.avail); \
+ eassert (fail_stack.avail >= 0); \
+ eassert (fail_stack.frame <= fail_stack.avail); \
\
DEBUG_STATEMENT (nfailure_points_popped++); \
-} while (0) /* POP_FAILURE_POINT */
+} while (false) /* POP_FAILURE_POINT */
@@ -1562,12 +1147,8 @@ do {
\
/* Subroutine declarations and macros for regex_compile. */
static reg_errcode_t regex_compile (re_char *pattern, size_t size,
-#ifdef emacs
bool posix_backtracking,
const char *whitespace_regexp,
-#else
- reg_syntax_t syntax,
-#endif
struct re_pattern_buffer *bufp);
static void store_op1 (re_opcode_t op, unsigned char *loc, int arg);
static void store_op2 (re_opcode_t op, unsigned char *loc, int arg1, int arg2);
@@ -1575,10 +1156,8 @@ static void insert_op1 (re_opcode_t op, unsigned char
*loc,
int arg, unsigned char *end);
static void insert_op2 (re_opcode_t op, unsigned char *loc,
int arg1, int arg2, unsigned char *end);
-static boolean at_begline_loc_p (re_char *pattern, re_char *p,
- reg_syntax_t syntax);
-static boolean at_endline_loc_p (re_char *p, re_char *pend,
- reg_syntax_t syntax);
+static bool at_begline_loc_p (re_char *pattern, re_char *p);
+static bool at_endline_loc_p (re_char *p, re_char *pend);
static re_char *skip_one_char (re_char *p);
static int analyze_first (re_char *p, re_char *pend,
char *fastmap, const int multibyte);
@@ -1591,35 +1170,28 @@ static int analyze_first (re_char *p, re_char *pend,
if (p == pend) return REG_EEND; \
c = RE_STRING_CHAR_AND_LENGTH (p, len, multibyte); \
p += len; \
- } while (0)
+ } while (false)
-/* If `translate' is non-null, return translate[D], else just D. We
- cast the subscript to translate because some data is declared as
- `char *', to avoid warnings when a string constant is passed. But
- when we use a character as a subscript we must make it unsigned. */
-#ifndef TRANSLATE
-# define TRANSLATE(d) \
- (RE_TRANSLATE_P (translate) ? RE_TRANSLATE (translate, (d)) : (d))
-#endif
-
+#define RE_TRANSLATE(TBL, C) char_table_translate (TBL, C)
+#define TRANSLATE(d) (!NILP (translate) ? RE_TRANSLATE (translate, d) : (d))
-/* Macros for outputting the compiled pattern into `buffer'. */
+/* Macros for outputting the compiled pattern into 'buffer'. */
/* If the buffer isn't allocated when it comes in, use this. */
#define INIT_BUF_SIZE 32
-/* Make sure we have at least N more bytes of space in buffer. */
+/* Ensure at least N more bytes of space in buffer. */
#define GET_BUFFER_SPACE(n) \
while ((size_t) (b - bufp->buffer + (n)) > bufp->allocated)
\
EXTEND_BUFFER ()
-/* Make sure we have one more byte of buffer space and then add C to it. */
+/* Ensure one more byte of buffer space and then add C to it. */
#define BUF_PUSH(c) \
do { \
GET_BUFFER_SPACE (1); \
*b++ = (unsigned char) (c);
\
- } while (0)
+ } while (false)
/* Ensure we have two more bytes of buffer space and then append C1 and C2. */
@@ -1628,10 +1200,10 @@ static int analyze_first (re_char *p, re_char *pend,
GET_BUFFER_SPACE (2); \
*b++ = (unsigned char) (c1); \
*b++ = (unsigned char) (c2); \
- } while (0)
+ } while (false)
-/* Store a jump with opcode OP at LOC to location TO. We store a
+/* Store a jump with opcode OP at LOC to location TO. Store a
relative address offset by the three bytes the jump itself occupies. */
#define STORE_JUMP(op, loc, to) \
store_op1 (op, loc, (to) - (loc) - 3)
@@ -1640,11 +1212,11 @@ static int analyze_first (re_char *p, re_char *pend,
#define STORE_JUMP2(op, loc, to, arg) \
store_op2 (op, loc, (to) - (loc) - 3, arg)
-/* Like `STORE_JUMP', but for inserting. Assume `b' is the buffer end. */
+/* Like 'STORE_JUMP', but for inserting. Assume B is the buffer end. */
#define INSERT_JUMP(op, loc, to) \
insert_op1 (op, loc, (to) - (loc) - 3, b)
-/* Like `STORE_JUMP2', but for inserting. Assume `b' is the buffer end. */
+/* Like 'STORE_JUMP2', but for inserting. Assume B is the buffer end. */
#define INSERT_JUMP2(op, loc, to, arg) \
insert_op2 (op, loc, (to) - (loc) - 3, arg, b)
@@ -1652,7 +1224,7 @@ static int analyze_first (re_char *p, re_char *pend,
/* This is not an arbitrary limit: the arguments which represent offsets
into the pattern are two bytes long. So if 2^15 bytes turns out to
be too small, many things would have to change. */
-# define MAX_BUF_SIZE (1L << 15)
+# define MAX_BUF_SIZE (1 << 15)
/* Extend the buffer by twice its current size via realloc and
reset the pointers that pointed into the old block to point to the
@@ -1676,15 +1248,13 @@ static int analyze_first (re_char *p, re_char *pend,
if (laststart_set) laststart_off = laststart - old_buffer; \
if (pending_exact_set) pending_exact_off = pending_exact - old_buffer; \
RETALLOC (bufp->buffer, bufp->allocated, unsigned char); \
- if (bufp->buffer == NULL) \
- return REG_ESPACE; \
unsigned char *new_buffer = bufp->buffer; \
b = new_buffer + b_off; \
begalt = new_buffer + begalt_off; \
if (fixup_alt_jump_set) fixup_alt_jump = new_buffer + fixup_alt_jump_off; \
if (laststart_set) laststart = new_buffer + laststart_off; \
if (pending_exact_set) pending_exact = new_buffer + pending_exact_off; \
- } while (0)
+ } while (false)
/* Since we have one byte reserved for the register number argument to
@@ -1692,7 +1262,7 @@ static int analyze_first (re_char *p, re_char *pend,
things about is what fits in that byte. */
#define MAX_REGNUM 255
-/* But patterns can have more than `MAX_REGNUM' registers. We just
+/* But patterns can have more than 'MAX_REGNUM' registers. Just
ignore the excess. */
typedef int regnum_t;
@@ -1701,7 +1271,6 @@ typedef int regnum_t;
/* Since offsets can go either forwards or backwards, this type needs to
be able to hold values from -(MAX_BUF_SIZE - 1) to MAX_BUF_SIZE - 1. */
-/* int may be not enough when sizeof(int) == 2. */
typedef long pattern_offset_t;
typedef struct
@@ -1728,12 +1297,6 @@ typedef struct
/* The next available element. */
#define COMPILE_STACK_TOP (compile_stack.stack[compile_stack.avail])
-
-/* Explicit quit checking is needed for Emacs, which uses polling to
- process input events. */
-#ifndef emacs
-static void maybe_quit (void) {}
-#endif
/* Structure to manage work area for range table. */
struct range_table_work_area
@@ -1744,8 +1307,6 @@ struct range_table_work_area
int bits; /* flag to record character classes */
};
-#ifdef emacs
-
/* Make sure that WORK_AREA can hold more N multibyte characters.
This is used only in set_image_of_range and set_image_of_range_1.
It expects WORK_AREA to be a pointer.
@@ -1759,7 +1320,7 @@ struct range_table_work_area
if ((work_area).table == 0) \
return (REG_ESPACE); \
}
\
- } while (0)
+ } while (false)
#define SET_RANGE_TABLE_WORK_AREA_BIT(work_area, bit) \
(work_area).bits |= (bit)
@@ -1770,18 +1331,17 @@ struct range_table_work_area
EXTEND_RANGE_TABLE ((work_area), 2); \
(work_area).table[(work_area).used++] = (range_start); \
(work_area).table[(work_area).used++] = (range_end); \
- } while (0)
-
-#endif /* emacs */
+ } while (false)
/* Free allocated memory for WORK_AREA. */
#define FREE_RANGE_TABLE_WORK_AREA(work_area) \
do { \
if ((work_area).table) \
- free ((work_area).table); \
- } while (0)
+ xfree ((work_area).table); \
+ } while (false)
-#define CLEAR_RANGE_TABLE_WORK_USED(work_area) ((work_area).used = 0,
(work_area).bits = 0)
+#define CLEAR_RANGE_TABLE_WORK_USED(work_area) \
+ ((work_area).used = 0, (work_area).bits = 0)
#define RANGE_TABLE_WORK_USED(work_area) ((work_area).used)
#define RANGE_TABLE_WORK_BITS(work_area) ((work_area).bits)
#define RANGE_TABLE_WORK_ELT(work_area, i) ((work_area).table[i])
@@ -1806,8 +1366,6 @@ struct range_table_work_area
#define SET_LIST_BIT(c) (b[((c)) / BYTEWIDTH] |= 1 << ((c) % BYTEWIDTH))
-#ifdef emacs
-
/* Store characters in the range FROM to TO in the bitmap at B (for
ASCII and unibyte characters) and WORK_AREA (for multibyte
characters) while translating them and paying attention to the
@@ -1822,7 +1380,7 @@ struct range_table_work_area
#define SETUP_ASCII_RANGE(work_area, FROM, TO) \
do { \
int C0, C1; \
- \
+ \
for (C0 = (FROM); C0 <= (TO); C0++) \
{ \
C1 = TRANSLATE (C0); \
@@ -1834,7 +1392,7 @@ struct range_table_work_area
} \
SET_LIST_BIT (C1); \
} \
- } while (0)
+ } while (false)
/* Both FROM and TO are unibyte characters (0x80..0xFF). */
@@ -1843,7 +1401,7 @@ struct range_table_work_area
do { \
int C0, C1, C2, I; \
int USED = RANGE_TABLE_WORK_USED (work_area); \
- \
+ \
for (C0 = (FROM); C0 <= (TO); C0++)
\
{
\
C1 = RE_CHAR_TO_MULTIBYTE (C0); \
@@ -1874,7 +1432,7 @@ struct range_table_work_area
SET_RANGE_TABLE_WORK_AREA ((work_area), C2, C2); \
} \
}
\
- } while (0)
+ } while (false)
/* Both FROM and TO are multibyte characters. */
@@ -1882,7 +1440,7 @@ struct range_table_work_area
#define SETUP_MULTIBYTE_RANGE(work_area, FROM, TO) \
do { \
int C0, C1, C2, I, USED = RANGE_TABLE_WORK_USED (work_area); \
- \
+ \
SET_RANGE_TABLE_WORK_AREA ((work_area), (FROM), (TO)); \
for (C0 = (FROM); C0 <= (TO); C0++)
\
{
\
@@ -1896,7 +1454,7 @@ struct range_table_work_area
{ \
int from = RANGE_TABLE_WORK_ELT (work_area, I); \
int to = RANGE_TABLE_WORK_ELT (work_area, I + 1); \
- \
+ \
if (C1 >= from - 1 && C1 <= to + 1) \
{ \
if (C1 == from - 1) \
@@ -1909,9 +1467,7 @@ struct range_table_work_area
if (I < USED) \
SET_RANGE_TABLE_WORK_AREA ((work_area), C1, C1); \
}
\
- } while (0)
-
-#endif /* emacs */
+ } while (false)
/* Get the next unsigned number in the uncompiled pattern. */
#define GET_INTERVAL_COUNT(num) \
@@ -1933,10 +1489,8 @@ struct range_table_work_area
PATFETCH (c); \
} \
}
\
- } while (0)
+ } while (false)
-#if ! WIDE_CHAR_SUPPORT
-
/* Parse a character class, i.e. string such as "[:name:]". *strp
points to the string to be parsed and limit is length, in bytes, of
that string.
@@ -2030,7 +1584,7 @@ re_wctype_parse (const unsigned char **strp, unsigned
limit)
}
/* True if CH is in the char class CC. */
-boolean
+bool
re_iswctype (int ch, re_wctype_t cc)
{
switch (cc)
@@ -2083,7 +1637,6 @@ re_wctype_to_bit (re_wctype_t cc)
abort ();
}
}
-#endif
/* Filling in the work area of a range. */
@@ -2093,357 +1646,75 @@ static void
extend_range_table_work_area (struct range_table_work_area *work_area)
{
work_area->allocated += 16 * sizeof (int);
- work_area->table = realloc (work_area->table, work_area->allocated);
+ work_area->table = xrealloc (work_area->table, work_area->allocated);
}
-
-#if 0
-#ifdef emacs
-
-/* Carefully find the ranges of codes that are equivalent
- under case conversion to the range start..end when passed through
- TRANSLATE. Handle the case where non-letters can come in between
- two upper-case letters (which happens in Latin-1).
- Also handle the case of groups of more than 2 case-equivalent chars.
-
- The basic method is to look at consecutive characters and see
- if they can form a run that can be handled as one.
-
- Returns -1 if successful, REG_ESPACE if ran out of space. */
-
-static int
-set_image_of_range_1 (struct range_table_work_area *work_area,
- re_wchar_t start, re_wchar_t end,
- RE_TRANSLATE_TYPE translate)
-{
- /* `one_case' indicates a character, or a run of characters,
- each of which is an isolate (no case-equivalents).
- This includes all ASCII non-letters.
-
- `two_case' indicates a character, or a run of characters,
- each of which has two case-equivalent forms.
- This includes all ASCII letters.
-
- `strange' indicates a character that has more than one
- case-equivalent. */
-
- enum case_type {one_case, two_case, strange};
-
- /* Describe the run that is in progress,
- which the next character can try to extend.
- If run_type is strange, that means there really is no run.
- If run_type is one_case, then run_start...run_end is the run.
- If run_type is two_case, then the run is run_start...run_end,
- and the case-equivalents end at run_eqv_end. */
-
- enum case_type run_type = strange;
- int run_start, run_end, run_eqv_end;
-
- Lisp_Object eqv_table;
-
- if (!RE_TRANSLATE_P (translate))
- {
- EXTEND_RANGE_TABLE (work_area, 2);
- work_area->table[work_area->used++] = (start);
- work_area->table[work_area->used++] = (end);
- return -1;
- }
-
- eqv_table = XCHAR_TABLE (translate)->extras[2];
-
- for (; start <= end; start++)
- {
- enum case_type this_type;
- int eqv = RE_TRANSLATE (eqv_table, start);
- int minchar, maxchar;
-
- /* Classify this character */
- if (eqv == start)
- this_type = one_case;
- else if (RE_TRANSLATE (eqv_table, eqv) == start)
- this_type = two_case;
- else
- this_type = strange;
-
- if (start < eqv)
- minchar = start, maxchar = eqv;
- else
- minchar = eqv, maxchar = start;
-
- /* Can this character extend the run in progress? */
- if (this_type == strange || this_type != run_type
- || !(minchar == run_end + 1
- && (run_type == two_case
- ? maxchar == run_eqv_end + 1 : 1)))
- {
- /* No, end the run.
- Record each of its equivalent ranges. */
- if (run_type == one_case)
- {
- EXTEND_RANGE_TABLE (work_area, 2);
- work_area->table[work_area->used++] = run_start;
- work_area->table[work_area->used++] = run_end;
- }
- else if (run_type == two_case)
- {
- EXTEND_RANGE_TABLE (work_area, 4);
- work_area->table[work_area->used++] = run_start;
- work_area->table[work_area->used++] = run_end;
- work_area->table[work_area->used++]
- = RE_TRANSLATE (eqv_table, run_start);
- work_area->table[work_area->used++]
- = RE_TRANSLATE (eqv_table, run_end);
- }
- run_type = strange;
- }
-
- if (this_type == strange)
- {
- /* For a strange character, add each of its equivalents, one
- by one. Don't start a range. */
- do
- {
- EXTEND_RANGE_TABLE (work_area, 2);
- work_area->table[work_area->used++] = eqv;
- work_area->table[work_area->used++] = eqv;
- eqv = RE_TRANSLATE (eqv_table, eqv);
- }
- while (eqv != start);
- }
-
- /* Add this char to the run, or start a new run. */
- else if (run_type == strange)
- {
- /* Initialize a new range. */
- run_type = this_type;
- run_start = start;
- run_end = start;
- run_eqv_end = RE_TRANSLATE (eqv_table, run_end);
- }
- else
- {
- /* Extend a running range. */
- run_end = minchar;
- run_eqv_end = RE_TRANSLATE (eqv_table, run_end);
- }
- }
-
- /* If a run is still in progress at the end, finish it now
- by recording its equivalent ranges. */
- if (run_type == one_case)
- {
- EXTEND_RANGE_TABLE (work_area, 2);
- work_area->table[work_area->used++] = run_start;
- work_area->table[work_area->used++] = run_end;
- }
- else if (run_type == two_case)
- {
- EXTEND_RANGE_TABLE (work_area, 4);
- work_area->table[work_area->used++] = run_start;
- work_area->table[work_area->used++] = run_end;
- work_area->table[work_area->used++]
- = RE_TRANSLATE (eqv_table, run_start);
- work_area->table[work_area->used++]
- = RE_TRANSLATE (eqv_table, run_end);
- }
-
- return -1;
-}
-
-#endif /* emacs */
-
-/* Record the image of the range start..end when passed through
- TRANSLATE. This is not necessarily TRANSLATE(start)..TRANSLATE(end)
- and is not even necessarily contiguous.
- Normally we approximate it with the smallest contiguous range that contains
- all the chars we need. However, for Latin-1 we go to extra effort
- to do a better job.
-
- This function is not called for ASCII ranges.
-
- Returns -1 if successful, REG_ESPACE if ran out of space. */
-
-static int
-set_image_of_range (struct range_table_work_area *work_area,
- re_wchar_t start, re_wchar_t end,
- RE_TRANSLATE_TYPE translate)
-{
- re_wchar_t cmin, cmax;
-
-#ifdef emacs
- /* For Latin-1 ranges, use set_image_of_range_1
- to get proper handling of ranges that include letters and nonletters.
- For a range that includes the whole of Latin-1, this is not necessary.
- For other character sets, we don't bother to get this right. */
- if (RE_TRANSLATE_P (translate) && start < 04400
- && !(start < 04200 && end >= 04377))
- {
- int newend;
- int tem;
- newend = end;
- if (newend > 04377)
- newend = 04377;
- tem = set_image_of_range_1 (work_area, start, newend, translate);
- if (tem > 0)
- return tem;
-
- start = 04400;
- if (end < 04400)
- return -1;
- }
-#endif
-
- EXTEND_RANGE_TABLE (work_area, 2);
- work_area->table[work_area->used++] = (start);
- work_area->table[work_area->used++] = (end);
-
- cmin = -1, cmax = -1;
-
- if (RE_TRANSLATE_P (translate))
- {
- int ch;
-
- for (ch = start; ch <= end; ch++)
- {
- re_wchar_t c = TRANSLATE (ch);
- if (! (start <= c && c <= end))
- {
- if (cmin == -1)
- cmin = c, cmax = c;
- else
- {
- cmin = min (cmin, c);
- cmax = max (cmax, c);
- }
- }
- }
-
- if (cmin != -1)
- {
- EXTEND_RANGE_TABLE (work_area, 2);
- work_area->table[work_area->used++] = (cmin);
- work_area->table[work_area->used++] = (cmax);
- }
- }
-
- return -1;
-}
-#endif /* 0 */
-
-#ifndef MATCH_MAY_ALLOCATE
-
-/* If we cannot allocate large objects within re_match_2_internal,
- we make the fail stack and register vectors global.
- The fail stack, we grow to the maximum size when a regexp
- is compiled.
- The register vectors, we adjust in size each time we
- compile a regexp, according to the number of registers it needs. */
-
-static fail_stack_type fail_stack;
-
-/* Size with which the following vectors are currently allocated.
- That is so we can make them bigger as needed,
- but never make them smaller. */
-static int regs_allocated_size;
-
-static re_char ** regstart, ** regend;
-static re_char **best_regstart, **best_regend;
-
-/* Make the register vectors big enough for NUM_REGS registers,
- but don't make them smaller. */
-
-static
-regex_grow_registers (int num_regs)
-{
- if (num_regs > regs_allocated_size)
- {
- RETALLOC_IF (regstart, num_regs, re_char *);
- RETALLOC_IF (regend, num_regs, re_char *);
- RETALLOC_IF (best_regstart, num_regs, re_char *);
- RETALLOC_IF (best_regend, num_regs, re_char *);
-
- regs_allocated_size = num_regs;
- }
-}
-
-#endif /* not MATCH_MAY_ALLOCATE */
-static boolean group_in_compile_stack (compile_stack_type compile_stack,
- regnum_t regnum);
-
-/* `regex_compile' compiles PATTERN (of length SIZE) according to SYNTAX.
- Returns one of error codes defined in `regex.h', or zero for success.
-
- If WHITESPACE_REGEXP is given (only #ifdef emacs), it is used instead of
- a space character in PATTERN.
-
- Assumes the `allocated' (and perhaps `buffer') and `translate'
- fields are set in BUFP on entry.
-
- If it succeeds, results are put in BUFP (if it returns an error, the
- contents of BUFP are undefined):
- `buffer' is the compiled pattern;
- `syntax' is set to SYNTAX;
- `used' is set to the length of the compiled pattern;
- `fastmap_accurate' is zero;
- `re_nsub' is the number of subexpressions in PATTERN;
- `not_bol' and `not_eol' are zero;
+/* regex_compile and helpers. */
- The `fastmap' field is neither examined nor set. */
+static bool group_in_compile_stack (compile_stack_type, regnum_t);
-/* Insert the `jump' from the end of last alternative to "here".
+/* Insert the 'jump' from the end of last alternative to "here".
The space for the jump has already been allocated. */
#define FIXUP_ALT_JUMP() \
do { \
if (fixup_alt_jump) \
STORE_JUMP (jump, fixup_alt_jump, b); \
-} while (0)
+} while (false)
/* Return, freeing storage we allocated. */
#define FREE_STACK_RETURN(value) \
do { \
FREE_RANGE_TABLE_WORK_AREA (range_table_work); \
- free (compile_stack.stack); \
+ xfree (compile_stack.stack); \
return value; \
- } while (0)
+ } while (false)
+
+/* Compile PATTERN (of length SIZE) according to SYNTAX.
+ Return a nonzero error code on failure, or zero for success.
+
+ If WHITESPACE_REGEXP is given, use it instead of a space
+ character in PATTERN.
+
+ Assume the 'allocated' (and perhaps 'buffer') and 'translate'
+ fields are set in BUFP on entry.
+
+ If successful, put results in *BUFP (otherwise the
+ contents of *BUFP are undefined):
+ 'buffer' is the compiled pattern;
+ 'syntax' is set to SYNTAX;
+ 'used' is set to the length of the compiled pattern;
+ 'fastmap_accurate' is zero;
+ 're_nsub' is the number of subexpressions in PATTERN;
+
+ The 'fastmap' field is neither examined nor set. */
static reg_errcode_t
regex_compile (re_char *pattern, size_t size,
-#ifdef emacs
-# define syntax RE_SYNTAX_EMACS
bool posix_backtracking,
const char *whitespace_regexp,
-#else
- reg_syntax_t syntax,
-# define posix_backtracking (!(syntax & RE_NO_POSIX_BACKTRACKING))
-#endif
struct re_pattern_buffer *bufp)
{
- /* We fetch characters from PATTERN here. */
- register re_wchar_t c, c1;
+ /* Fetch characters from PATTERN here. */
+ int c, c1;
/* Points to the end of the buffer, where we should append. */
- register unsigned char *b;
+ unsigned char *b;
/* Keeps track of unclosed groups. */
compile_stack_type compile_stack;
/* Points to the current (ending) position in the pattern. */
-#ifdef AIX
- /* `const' makes AIX compiler fail. */
- unsigned char *p = pattern;
-#else
re_char *p = pattern;
-#endif
re_char *pend = pattern + size;
/* How to translate the characters in the pattern. */
- RE_TRANSLATE_TYPE translate = bufp->translate;
+ Lisp_Object translate = bufp->translate;
- /* Address of the count-byte of the most recently inserted `exactn'
+ /* Address of the count-byte of the most recently inserted 'exactn'
command. This makes it possible to tell if a new exact-match
character can be added to that command or if the character requires
- a new `exactn' command. */
+ a new 'exactn' command. */
unsigned char *pending_exact = 0;
/* Address of start of the most recently finished expression.
@@ -2459,7 +1730,7 @@ regex_compile (re_char *pattern, size_t size,
re_char *beg_interval;
/* Address of the place where a forward jump should go to the end of
- the containing expression. Each alternative of an `or' -- except the
+ the containing expression. Each alternative of an 'or' -- except the
last -- ends with a forward jump of this sort. */
unsigned char *fixup_alt_jump = 0;
@@ -2467,9 +1738,8 @@ regex_compile (re_char *pattern, size_t size,
struct range_table_work_area range_table_work;
/* If the object matched can contain multibyte characters. */
- const boolean multibyte = RE_MULTIBYTE_P (bufp);
+ bool multibyte = RE_MULTIBYTE_P (bufp);
-#ifdef emacs
/* Nonzero if we have pushed down into a subpattern. */
int in_subpattern = 0;
@@ -2478,26 +1748,22 @@ regex_compile (re_char *pattern, size_t size,
re_char *main_p;
re_char *main_pattern;
re_char *main_pend;
-#endif
-#ifdef DEBUG
- debug++;
+#ifdef REGEX_EMACS_DEBUG
+ regex_emacs_debug++;
DEBUG_PRINT ("\nCompiling pattern: ");
- if (debug > 0)
+ if (regex_emacs_debug > 0)
{
- unsigned debug_count;
+ size_t debug_count;
for (debug_count = 0; debug_count < size; debug_count++)
putchar (pattern[debug_count]);
putchar ('\n');
}
-#endif /* DEBUG */
+#endif
/* Initialize the compile stack. */
compile_stack.stack = TALLOC (INIT_COMPILE_STACK_SIZE, compile_stack_elt_t);
- if (compile_stack.stack == NULL)
- return REG_ESPACE;
-
compile_stack.size = INIT_COMPILE_STACK_SIZE;
compile_stack.avail = 0;
@@ -2505,26 +1771,16 @@ regex_compile (re_char *pattern, size_t size,
range_table_work.allocated = 0;
/* Initialize the pattern buffer. */
-#ifndef emacs
- bufp->syntax = syntax;
-#endif
bufp->fastmap_accurate = 0;
- bufp->not_bol = bufp->not_eol = 0;
bufp->used_syntax = 0;
- /* Set `used' to zero, so that if we return an error, the pattern
+ /* Set 'used' to zero, so that if we return an error, the pattern
printer (for debugging) will think there's no pattern. We reset it
at the end. */
bufp->used = 0;
- /* Always count groups, whether or not bufp->no_sub is set. */
bufp->re_nsub = 0;
-#if !defined emacs && !defined SYNTAX_TABLE
- /* Initialize the syntax table. */
- init_syntax_once ();
-#endif
-
if (bufp->allocated == 0)
{
if (bufp->buffer)
@@ -2537,8 +1793,6 @@ regex_compile (re_char *pattern, size_t size,
{ /* Caller did not allocate a buffer. Do it for them. */
bufp->buffer = TALLOC (INIT_BUF_SIZE, unsigned char);
}
- if (!bufp->buffer) FREE_STACK_RETURN (REG_ESPACE);
-
bufp->allocated = INIT_BUF_SIZE;
}
@@ -2549,7 +1803,6 @@ regex_compile (re_char *pattern, size_t size,
{
if (p == pend)
{
-#ifdef emacs
/* If this is the end of an included regexp,
pop back to the main regexp and try again. */
if (in_subpattern)
@@ -2560,7 +1813,6 @@ regex_compile (re_char *pattern, size_t size,
pend = main_pend;
continue;
}
-#endif
/* If this is the end of the main regexp, we are done. */
break;
}
@@ -2569,7 +1821,6 @@ regex_compile (re_char *pattern, size_t size,
switch (c)
{
-#ifdef emacs
case ' ':
{
re_char *p1 = p;
@@ -2602,95 +1853,51 @@ regex_compile (re_char *pattern, size_t size,
pend = p + strlen (whitespace_regexp);
break;
}
-#endif
case '^':
- {
- if ( /* If at start of pattern, it's an operator. */
- p == pattern + 1
- /* If context independent, it's an operator. */
- || syntax & RE_CONTEXT_INDEP_ANCHORS
- /* Otherwise, depends on what's come before. */
- || at_begline_loc_p (pattern, p, syntax))
- BUF_PUSH ((syntax & RE_NO_NEWLINE_ANCHOR) ? begbuf : begline);
- else
- goto normal_char;
- }
+ if (! (p == pattern + 1 || at_begline_loc_p (pattern, p)))
+ goto normal_char;
+ BUF_PUSH (begline);
break;
-
case '$':
- {
- if ( /* If at end of pattern, it's an operator. */
- p == pend
- /* If context independent, it's an operator. */
- || syntax & RE_CONTEXT_INDEP_ANCHORS
- /* Otherwise, depends on what's next. */
- || at_endline_loc_p (p, pend, syntax))
- BUF_PUSH ((syntax & RE_NO_NEWLINE_ANCHOR) ? endbuf : endline);
- else
- goto normal_char;
- }
- break;
+ if (! (p == pend || at_endline_loc_p (p, pend)))
+ goto normal_char;
+ BUF_PUSH (endline);
+ break;
case '+':
case '?':
- if ((syntax & RE_BK_PLUS_QM)
- || (syntax & RE_LIMITED_OPS))
- goto normal_char;
- FALLTHROUGH;
case '*':
- handle_plus:
/* If there is no previous pattern... */
if (!laststart)
- {
- if (syntax & RE_CONTEXT_INVALID_OPS)
- FREE_STACK_RETURN (REG_BADRPT);
- else if (!(syntax & RE_CONTEXT_INDEP_OPS))
- goto normal_char;
- }
+ goto normal_char;
{
/* 1 means zero (many) matches is allowed. */
- boolean zero_times_ok = 0, many_times_ok = 0;
- boolean greedy = 1;
+ bool zero_times_ok = false, many_times_ok = false;
+ bool greedy = true;
/* If there is a sequence of repetition chars, collapse it
down to just one (the right one). We can't combine
- interval operators with these because of, e.g., `a{2}*',
- which should only match an even number of `a's. */
+ interval operators with these because of, e.g., 'a{2}*',
+ which should only match an even number of 'a's. */
for (;;)
{
- if ((syntax & RE_FRUGAL)
- && c == '?' && (zero_times_ok || many_times_ok))
- greedy = 0;
+ if (c == '?' && (zero_times_ok || many_times_ok))
+ greedy = false;
else
{
zero_times_ok |= c != '+';
many_times_ok |= c != '?';
}
- if (p == pend)
- break;
- else if (*p == '*'
- || (!(syntax & RE_BK_PLUS_QM)
- && (*p == '+' || *p == '?')))
- ;
- else if (syntax & RE_BK_PLUS_QM && *p == '\\')
- {
- if (p+1 == pend)
- FREE_STACK_RETURN (REG_EESCAPE);
- if (p[1] == '+' || p[1] == '?')
- PATFETCH (c); /* Gobble up the backslash. */
- else
- break;
- }
- else
+ if (! (p < pend && (*p == '*' || *p == '+' || *p == '?')))
break;
/* If we get here, we found another repeat character. */
- PATFETCH (c);
+ c = *p++;
}
/* Star, etc. applied to an empty pattern is equivalent
@@ -2704,25 +1911,25 @@ regex_compile (re_char *pattern, size_t size,
{
if (many_times_ok)
{
- boolean simple = skip_one_char (laststart) == b;
+ bool simple = skip_one_char (laststart) == b;
size_t startoffset = 0;
re_opcode_t ofj =
/* Check if the loop can match the empty string. */
(simple || !analyze_first (laststart, b, NULL, 0))
? on_failure_jump : on_failure_jump_loop;
- assert (skip_one_char (laststart) <= b);
+ eassert (skip_one_char (laststart) <= b);
if (!zero_times_ok && simple)
{ /* Since simple * loops can be made faster by using
- on_failure_keep_string_jump, we turn simple P+
- into PP* if P is simple. */
- unsigned char *p1, *p2;
- startoffset = b - laststart;
- GET_BUFFER_SPACE (startoffset);
- p1 = b; p2 = laststart;
- while (p2 < p1)
- *b++ = *p2++;
- zero_times_ok = 1;
+ on_failure_keep_string_jump, we turn simple P+
+ into PP* if P is simple. */
+ unsigned char *p1, *p2;
+ startoffset = b - laststart;
+ GET_BUFFER_SPACE (startoffset);
+ p1 = b; p2 = laststart;
+ while (p2 < p1)
+ *b++ = *p2++;
+ zero_times_ok = 1;
}
GET_BUFFER_SPACE (6);
@@ -2743,7 +1950,7 @@ regex_compile (re_char *pattern, size_t size,
else
{
/* A simple ? pattern. */
- assert (zero_times_ok);
+ eassert (zero_times_ok);
GET_BUFFER_SPACE (3);
INSERT_JUMP (on_failure_jump, laststart, b + 3);
b += 3;
@@ -2755,7 +1962,7 @@ regex_compile (re_char *pattern, size_t size,
GET_BUFFER_SPACE (7); /* We might use less. */
if (many_times_ok)
{
- boolean emptyp = analyze_first (laststart, b, NULL, 0);
+ bool emptyp = analyze_first (laststart, b, NULL, 0);
/* The non-greedy multiple match looks like
a repeat..until: we only need a conditional jump
@@ -2807,8 +2014,8 @@ regex_compile (re_char *pattern, size_t size,
laststart = b;
- /* We test `*p == '^' twice, instead of using an if
- statement, so we only need one BUF_PUSH. */
+ /* Test '*p == '^' twice, instead of using an if
+ statement, so we need only one BUF_PUSH. */
BUF_PUSH (*p == '^' ? charset_not : charset);
if (*p == '^')
p++;
@@ -2822,25 +2029,18 @@ regex_compile (re_char *pattern, size_t size,
/* Clear the whole map. */
memset (b, 0, (1 << BYTEWIDTH) / BYTEWIDTH);
- /* charset_not matches newline according to a syntax bit. */
- if ((re_opcode_t) b[-2] == charset_not
- && (syntax & RE_HAT_LISTS_NOT_NEWLINE))
- SET_LIST_BIT ('\n');
-
/* Read in characters and ranges, setting map bits. */
for (;;)
{
- boolean escaped_char = false;
const unsigned char *p2 = p;
- re_wctype_t cc;
- re_wchar_t ch;
+ int ch;
if (p == pend) FREE_STACK_RETURN (REG_EBRACK);
/* See if we're at the beginning of a possible character
class. */
- if (syntax & RE_CHAR_CLASSES &&
- (cc = re_wctype_parse(&p, pend - p)) != -1)
+ re_wctype_t cc = re_wctype_parse (&p, pend - p);
+ if (cc != -1)
{
if (cc == 0)
FREE_STACK_RETURN (REG_ECTYPE);
@@ -2848,15 +2048,6 @@ regex_compile (re_char *pattern, size_t size,
if (p == pend)
FREE_STACK_RETURN (REG_EBRACK);
-#ifndef emacs
- for (ch = 0; ch < (1 << BYTEWIDTH); ++ch)
- if (re_iswctype (btowc (ch), cc))
- {
- c = TRANSLATE (ch);
- if (c < (1 << BYTEWIDTH))
- SET_LIST_BIT (c);
- }
-#else /* emacs */
/* Most character classes in a multibyte match just set
a flag. Exceptions are is_blank, is_digit, is_cntrl, and
is_xdigit, since they can only match ASCII characters.
@@ -2883,7 +2074,7 @@ regex_compile (re_char *pattern, size_t size,
}
SET_RANGE_TABLE_WORK_AREA_BIT
(range_table_work, re_wctype_to_bit (cc));
-#endif /* emacs */
+
/* In most cases the matching rule for char classes only
uses the syntax table for multibyte chars, so that the
content of the syntax-table is not hardcoded in the
@@ -2901,60 +2092,33 @@ regex_compile (re_char *pattern, size_t size,
(let ((case-fold-search t)) (string-match "[A-_]" "A")) */
PATFETCH (c);
- /* \ might escape characters inside [...] and [^...]. */
- if ((syntax & RE_BACKSLASH_ESCAPE_IN_LISTS) && c == '\\')
- {
- if (p == pend) FREE_STACK_RETURN (REG_EESCAPE);
-
- PATFETCH (c);
- escaped_char = true;
- }
- else
- {
- /* Could be the end of the bracket expression. If it's
- not (i.e., when the bracket expression is `[]' so
- far), the ']' character bit gets set way below. */
- if (c == ']' && p2 != p1)
- break;
- }
+ /* Could be the end of the bracket expression. If it's
+ not (i.e., when the bracket expression is '[]' so
+ far), the ']' character bit gets set way below. */
+ if (c == ']' && p2 != p1)
+ break;
if (p < pend && p[0] == '-' && p[1] != ']')
{
- /* Discard the `-'. */
+ /* Discard the '-'. */
PATFETCH (c1);
/* Fetch the character which ends the range. */
PATFETCH (c1);
-#ifdef emacs
+
if (CHAR_BYTE8_P (c1)
&& ! ASCII_CHAR_P (c) && ! CHAR_BYTE8_P (c))
/* Treat the range from a multibyte character to
raw-byte character as empty. */
c = c1 + 1;
-#endif /* emacs */
}
else
/* Range from C to C. */
c1 = c;
- if (c > c1)
+ if (c <= c1)
{
- if (syntax & RE_NO_EMPTY_RANGES)
- FREE_STACK_RETURN (REG_ERANGEX);
- /* Else, repeat the loop. */
- }
- else
- {
-#ifndef emacs
- /* Set the range into bitmap */
- for (; c <= c1; c++)
- {
- ch = TRANSLATE (c);
- if (ch < (1 << BYTEWIDTH))
- SET_LIST_BIT (ch);
- }
-#else /* emacs */
if (c < 128)
{
ch = min (127, c1);
@@ -2963,25 +2127,17 @@ regex_compile (re_char *pattern, size_t size,
if (CHAR_BYTE8_P (c1))
c = BYTE8_TO_CHAR (128);
}
- if (c <= c1)
+ if (CHAR_BYTE8_P (c))
{
- if (CHAR_BYTE8_P (c))
- {
- c = CHAR_TO_BYTE8 (c);
- c1 = CHAR_TO_BYTE8 (c1);
- for (; c <= c1; c++)
- SET_LIST_BIT (c);
- }
- else if (multibyte)
- {
- SETUP_MULTIBYTE_RANGE (range_table_work, c, c1);
- }
- else
- {
- SETUP_UNIBYTE_RANGE (range_table_work, c, c1);
- }
+ c = CHAR_TO_BYTE8 (c);
+ c1 = CHAR_TO_BYTE8 (c1);
+ for (; c <= c1; c++)
+ SET_LIST_BIT (c);
}
-#endif /* emacs */
+ else if (multibyte)
+ SETUP_MULTIBYTE_RANGE (range_table_work, c, c1);
+ else
+ SETUP_UNIBYTE_RANGE (range_table_work, c, c1);
}
}
@@ -3003,56 +2159,20 @@ regex_compile (re_char *pattern, size_t size,
each character. */
GET_BUFFER_SPACE (4 + used * 3);
- /* Indicate the existence of range table. */
- laststart[1] |= 0x80;
-
- /* Store the character class flag bits into the range table.
- If not in emacs, these flag bits are always 0. */
- *b++ = RANGE_TABLE_WORK_BITS (range_table_work) & 0xff;
- *b++ = RANGE_TABLE_WORK_BITS (range_table_work) >> 8;
-
- STORE_NUMBER_AND_INCR (b, used / 2);
- for (i = 0; i < used; i++)
- STORE_CHARACTER_AND_INCR
- (b, RANGE_TABLE_WORK_ELT (range_table_work, i));
- }
- }
- break;
-
-
- case '(':
- if (syntax & RE_NO_BK_PARENS)
- goto handle_open;
- else
- goto normal_char;
-
-
- case ')':
- if (syntax & RE_NO_BK_PARENS)
- goto handle_close;
- else
- goto normal_char;
-
-
- case '\n':
- if (syntax & RE_NEWLINE_ALT)
- goto handle_alt;
- else
- goto normal_char;
-
-
- case '|':
- if (syntax & RE_NO_BK_VBAR)
- goto handle_alt;
- else
- goto normal_char;
-
+ /* Indicate the existence of range table. */
+ laststart[1] |= 0x80;
+
+ /* Store the character class flag bits into the range table. */
+ *b++ = RANGE_TABLE_WORK_BITS (range_table_work) & 0xff;
+ *b++ = RANGE_TABLE_WORK_BITS (range_table_work) >> 8;
- case '{':
- if (syntax & RE_INTERVALS && syntax & RE_NO_BK_BRACES)
- goto handle_interval;
- else
- goto normal_char;
+ STORE_NUMBER_AND_INCR (b, used / 2);
+ for (i = 0; i < used; i++)
+ STORE_CHARACTER_AND_INCR
+ (b, RANGE_TABLE_WORK_ELT (range_table_work, i));
+ }
+ }
+ break;
case '\\':
@@ -3066,17 +2186,13 @@ regex_compile (re_char *pattern, size_t size,
switch (c)
{
case '(':
- if (syntax & RE_NO_BK_PARENS)
- goto normal_backslash;
-
- handle_open:
{
int shy = 0;
regnum_t regnum = 0;
if (p+1 < pend)
{
/* Look for a special (?...) construct */
- if ((syntax & RE_SHY_GROUPS) && *p == '?')
+ if (*p == '?')
{
PATFETCH (c); /* Gobble up the '?'. */
while (!shy)
@@ -3126,8 +2242,6 @@ regex_compile (re_char *pattern, size_t size,
{
RETALLOC (compile_stack.stack, compile_stack.size << 1,
compile_stack_elt_t);
- if (compile_stack.stack == NULL) return REG_ESPACE;
-
compile_stack.size <<= 1;
}
@@ -3159,35 +2273,22 @@ regex_compile (re_char *pattern, size_t size,
}
case ')':
- if (syntax & RE_NO_BK_PARENS) goto normal_backslash;
-
if (COMPILE_STACK_EMPTY)
- {
- if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)
- goto normal_backslash;
- else
- FREE_STACK_RETURN (REG_ERPAREN);
- }
+ FREE_STACK_RETURN (REG_ERPAREN);
- handle_close:
FIXUP_ALT_JUMP ();
/* See similar code for backslashed left paren above. */
if (COMPILE_STACK_EMPTY)
- {
- if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD)
- goto normal_char;
- else
- FREE_STACK_RETURN (REG_ERPAREN);
- }
+ FREE_STACK_RETURN (REG_ERPAREN);
/* Since we just checked for an empty stack above, this
- ``can't happen''. */
- assert (compile_stack.avail != 0);
+ "can't happen". */
+ eassert (compile_stack.avail != 0);
{
- /* We don't just want to restore into `regnum', because
+ /* We don't just want to restore into 'regnum', because
later groups should continue to be numbered higher,
- as in `(ab)c(de)' -- the second group is #2. */
+ as in '(ab)c(de)' -- the second group is #2. */
regnum_t regnum;
compile_stack.avail--;
@@ -3211,13 +2312,7 @@ regex_compile (re_char *pattern, size_t size,
break;
- case '|': /* `\|'. */
- if (syntax & RE_LIMITED_OPS || syntax & RE_NO_BK_VBAR)
- goto normal_backslash;
- handle_alt:
- if (syntax & RE_LIMITED_OPS)
- goto normal_char;
-
+ case '|': /* '\|'. */
/* Insert before the previous alternative a jump which
jumps to this alternative if the former fails. */
GET_BUFFER_SPACE (3);
@@ -3234,12 +2329,12 @@ regex_compile (re_char *pattern, size_t size,
_____ _____
| | | |
| v | v
- a | b | c
+ A | B | C
- If we are at `b', then fixup_alt_jump right now points to a
- three-byte space after `a'. We'll put in the jump, set
- fixup_alt_jump to right after `b', and leave behind three
- bytes which we'll fill in when we get to after `c'. */
+ If we are at B, then fixup_alt_jump right now points to a
+ three-byte space after A. We'll put in the jump, set
+ fixup_alt_jump to right after B, and leave behind three
+ bytes which we'll fill in when we get to after C. */
FIXUP_ALT_JUMP ();
@@ -3256,17 +2351,7 @@ regex_compile (re_char *pattern, size_t size,
case '{':
- /* If \{ is a literal. */
- if (!(syntax & RE_INTERVALS)
- /* If we're at `\{' and it's not the open-interval
- operator. */
- || (syntax & RE_NO_BK_BRACES))
- goto normal_backslash;
-
- handle_interval:
{
- /* If got here, then the syntax allows intervals. */
-
/* At least (most) this many matches must be made. */
int lower_bound = 0, upper_bound = -1;
@@ -3277,37 +2362,23 @@ regex_compile (re_char *pattern, size_t size,
if (c == ',')
GET_INTERVAL_COUNT (upper_bound);
else
- /* Interval such as `{1}' => match exactly once. */
+ /* Interval such as '{1}' => match exactly once. */
upper_bound = lower_bound;
if (lower_bound < 0
- || (0 <= upper_bound && upper_bound < lower_bound))
+ || (0 <= upper_bound && upper_bound < lower_bound)
+ || c != '\\')
FREE_STACK_RETURN (REG_BADBR);
-
- if (!(syntax & RE_NO_BK_BRACES))
- {
- if (c != '\\')
- FREE_STACK_RETURN (REG_BADBR);
- if (p == pend)
- FREE_STACK_RETURN (REG_EESCAPE);
- PATFETCH (c);
- }
-
- if (c != '}')
+ if (p == pend)
+ FREE_STACK_RETURN (REG_EESCAPE);
+ if (*p++ != '}')
FREE_STACK_RETURN (REG_BADBR);
/* We just parsed a valid interval. */
/* If it's invalid to have no preceding re. */
if (!laststart)
- {
- if (syntax & RE_CONTEXT_INVALID_OPS)
- FREE_STACK_RETURN (REG_BADRPT);
- else if (syntax & RE_CONTEXT_INDEP_OPS)
- laststart = b;
- else
- goto unfetch_interval;
- }
+ goto unfetch_interval;
if (upper_bound == 0)
/* If the upper bound is zero, just drop the sub pattern
@@ -3324,8 +2395,8 @@ regex_compile (re_char *pattern, size_t size,
succeed_n <after jump addr> <succeed_n count>
<body of loop>
jump_n <succeed_n addr> <jump count>
- (The upper bound and `jump_n' are omitted if
- `upper_bound' is 1, though.) */
+ (The upper bound and 'jump_n' are omitted if
+ 'upper_bound' is 1, though.) */
else
{ /* If the upper bound is > 1, we need to insert
more at the end of the loop. */
@@ -3345,21 +2416,22 @@ regex_compile (re_char *pattern, size_t size,
}
else
{
- /* Initialize lower bound of the `succeed_n', even
+ /* Initialize lower bound of the 'succeed_n', even
though it will be set during matching by its
- attendant `set_number_at' (inserted next),
- because `re_compile_fastmap' needs to know.
- Jump to the `jump_n' we might insert below. */
+ attendant 'set_number_at' (inserted next),
+ because 're_compile_fastmap' needs to know.
+ Jump to the 'jump_n' we might insert below. */
INSERT_JUMP2 (succeed_n, laststart,
b + 5 + nbytes,
lower_bound);
b += 5;
/* Code to initialize the lower bound. Insert
- before the `succeed_n'. The `5' is the last two
- bytes of this `set_number_at', plus 3 bytes of
- the following `succeed_n'. */
- insert_op2 (set_number_at, laststart, 5, lower_bound,
b);
+ before the 'succeed_n'. The '5' is the last two
+ bytes of this 'set_number_at', plus 3 bytes of
+ the following 'succeed_n'. */
+ insert_op2 (set_number_at, laststart, 5,
+ lower_bound, b);
b += 5;
startoffset += 5;
}
@@ -3373,28 +2445,28 @@ regex_compile (re_char *pattern, size_t size,
}
else if (upper_bound > 1)
{ /* More than one repetition is allowed, so
- append a backward jump to the `succeed_n'
+ append a backward jump to the 'succeed_n'
that starts this interval.
When we've reached this during matching,
we'll have matched the interval once, so
- jump back only `upper_bound - 1' times. */
+ jump back only 'upper_bound - 1' times. */
STORE_JUMP2 (jump_n, b, laststart + startoffset,
upper_bound - 1);
b += 5;
/* The location we want to set is the second
- parameter of the `jump_n'; that is `b-2' as
- an absolute address. `laststart' will be
- the `set_number_at' we're about to insert;
- `laststart+3' the number to set, the source
+ parameter of the 'jump_n'; that is 'b-2' as
+ an absolute address. 'laststart' will be
+ the 'set_number_at' we're about to insert;
+ 'laststart+3' the number to set, the source
for the relative address. But we are
inserting into the middle of the pattern --
so everything is getting moved up by 5.
Conclusion: (b - 2) - (laststart + 3) + 5,
i.e., b - laststart.
- We insert this at the beginning of the loop
+ Insert this at the beginning of the loop
so that if we fail during matching, we'll
reinitialize the bounds. */
insert_op2 (set_number_at, laststart, b - laststart,
@@ -3409,22 +2481,13 @@ regex_compile (re_char *pattern, size_t size,
unfetch_interval:
/* If an invalid interval, match the characters as literals. */
- assert (beg_interval);
+ eassert (beg_interval);
p = beg_interval;
beg_interval = NULL;
-
- /* normal_char and normal_backslash need `c'. */
+ eassert (p > pattern && p[-1] == '\\');
c = '{';
+ goto normal_char;
- if (!(syntax & RE_NO_BK_BRACES))
- {
- assert (p > pattern && p[-1] == '\\');
- goto normal_backslash;
- }
- else
- goto normal_char;
-
-#ifdef emacs
case '=':
laststart = b;
BUF_PUSH (at_dot);
@@ -3453,42 +2516,30 @@ regex_compile (re_char *pattern, size_t size,
PATFETCH (c);
BUF_PUSH_2 (notcategoryspec, c);
break;
-#endif /* emacs */
-
case 'w':
- if (syntax & RE_NO_GNU_OPS)
- goto normal_char;
laststart = b;
BUF_PUSH_2 (syntaxspec, Sword);
break;
case 'W':
- if (syntax & RE_NO_GNU_OPS)
- goto normal_char;
laststart = b;
BUF_PUSH_2 (notsyntaxspec, Sword);
break;
case '<':
- if (syntax & RE_NO_GNU_OPS)
- goto normal_char;
laststart = b;
BUF_PUSH (wordbeg);
break;
case '>':
- if (syntax & RE_NO_GNU_OPS)
- goto normal_char;
laststart = b;
BUF_PUSH (wordend);
break;
case '_':
- if (syntax & RE_NO_GNU_OPS)
- goto normal_char;
laststart = b;
PATFETCH (c);
if (c == '<')
@@ -3500,38 +2551,25 @@ regex_compile (re_char *pattern, size_t size,
break;
case 'b':
- if (syntax & RE_NO_GNU_OPS)
- goto normal_char;
BUF_PUSH (wordbound);
break;
case 'B':
- if (syntax & RE_NO_GNU_OPS)
- goto normal_char;
BUF_PUSH (notwordbound);
break;
case '`':
- if (syntax & RE_NO_GNU_OPS)
- goto normal_char;
BUF_PUSH (begbuf);
break;
case '\'':
- if (syntax & RE_NO_GNU_OPS)
- goto normal_char;
BUF_PUSH (endbuf);
break;
case '1': case '2': case '3': case '4': case '5':
case '6': case '7': case '8': case '9':
{
- regnum_t reg;
-
- if (syntax & RE_NO_BK_REFS)
- goto normal_backslash;
-
- reg = c - '0';
+ regnum_t reg = c - '0';
if (reg > bufp->re_nsub || reg < 1
/* Can't back reference to a subexp before its end. */
@@ -3543,16 +2581,7 @@ regex_compile (re_char *pattern, size_t size,
}
break;
-
- case '+':
- case '?':
- if (syntax & RE_BK_PLUS_QM)
- goto handle_plus;
- else
- goto normal_backslash;
-
default:
- normal_backslash:
/* You might think it would be useful for \ to mean
not to translate; but if we don't translate it
it will never match anything. */
@@ -3562,7 +2591,7 @@ regex_compile (re_char *pattern, size_t size,
default:
- /* Expects the character in `c'. */
+ /* Expects the character in C. */
normal_char:
/* If no exactn currently being built. */
if (!pending_exact
@@ -3570,18 +2599,13 @@ regex_compile (re_char *pattern, size_t size,
/* If last exactn not at current position. */
|| pending_exact + *pending_exact + 1 != b
- /* We have only one byte following the exactn for the count. */
+ /* Only one byte follows the exactn for the count. */
|| *pending_exact >= (1 << BYTEWIDTH) - MAX_MULTIBYTE_LENGTH
/* If followed by a repetition operator. */
- || (p != pend && (*p == '*' || *p == '^'))
- || ((syntax & RE_BK_PLUS_QM)
- ? p + 1 < pend && *p == '\\' && (p[1] == '+' || p[1] == '?')
- : p != pend && (*p == '+' || *p == '?'))
- || ((syntax & RE_INTERVALS)
- && ((syntax & RE_NO_BK_BRACES)
- ? p != pend && *p == '{'
- : p + 1 < pend && p[0] == '\\' && p[1] == '{')))
+ || (p != pend
+ && (*p == '*' || *p == '+' || *p == '?' || *p == '^'))
+ || (p + 1 < pend && p[0] == '\\' && p[1] == '{'))
{
/* Start building a new exactn. */
@@ -3606,7 +2630,7 @@ regex_compile (re_char *pattern, size_t size,
c1 = RE_CHAR_TO_MULTIBYTE (c);
if (! CHAR_BYTE8_P (c1))
{
- re_wchar_t c2 = TRANSLATE (c1);
+ int c2 = TRANSLATE (c1);
if (c1 != c2 && (c1 = RE_CHAR_TO_UNIBYTE (c2)) >= 0)
c = c1;
@@ -3634,47 +2658,24 @@ regex_compile (re_char *pattern, size_t size,
if (!posix_backtracking)
BUF_PUSH (succeed);
- /* We have succeeded; set the length of the buffer. */
+ /* Success; set the length of the buffer. */
bufp->used = b - bufp->buffer;
-#ifdef DEBUG
- if (debug > 0)
+#ifdef REGEX_EMACS_DEBUG
+ if (regex_emacs_debug > 0)
{
re_compile_fastmap (bufp);
DEBUG_PRINT ("\nCompiled pattern: \n");
print_compiled_pattern (bufp);
}
- debug--;
-#endif /* DEBUG */
-
-#ifndef MATCH_MAY_ALLOCATE
- /* Initialize the failure stack to the largest possible stack. This
- isn't necessary unless we're trying to avoid calling alloca in
- the search and match routines. */
- {
- int num_regs = bufp->re_nsub + 1;
-
- if (fail_stack.size < emacs_re_max_failures * TYPICAL_FAILURE_SIZE)
- {
- fail_stack.size = emacs_re_max_failures * TYPICAL_FAILURE_SIZE;
- falk_stack.stack = realloc (fail_stack.stack,
- fail_stack.size * sizeof *falk_stack.stack);
- }
-
- regex_grow_registers (num_regs);
- }
-#endif /* not MATCH_MAY_ALLOCATE */
+ regex_emacs_debug--;
+#endif
FREE_STACK_RETURN (REG_NOERROR);
-#ifdef emacs
-# undef syntax
-#else
-# undef posix_backtracking
-#endif
} /* regex_compile */
-/* Subroutines for `regex_compile'. */
+/* Subroutines for 'regex_compile'. */
/* Store OP at LOC followed by two-byte integer parameter ARG. */
@@ -3686,7 +2687,7 @@ store_op1 (re_opcode_t op, unsigned char *loc, int arg)
}
-/* Like `store_op1', but for two two-byte parameters ARG1 and ARG2. */
+/* Like 'store_op1', but for two two-byte parameters ARG1 and ARG2. */
static void
store_op2 (re_opcode_t op, unsigned char *loc, int arg1, int arg2)
@@ -3713,10 +2714,11 @@ insert_op1 (re_opcode_t op, unsigned char *loc, int
arg, unsigned char *end)
}
-/* Like `insert_op1', but for two two-byte parameters ARG1 and ARG2. */
+/* Like 'insert_op1', but for two two-byte parameters ARG1 and ARG2. */
static void
-insert_op2 (re_opcode_t op, unsigned char *loc, int arg1, int arg2, unsigned
char *end)
+insert_op2 (re_opcode_t op, unsigned char *loc, int arg1, int arg2,
+ unsigned char *end)
{
register unsigned char *pfrom = end;
register unsigned char *pto = end + 5;
@@ -3729,74 +2731,60 @@ insert_op2 (re_opcode_t op, unsigned char *loc, int
arg1, int arg2, unsigned cha
/* P points to just after a ^ in PATTERN. Return true if that ^ comes
- after an alternative or a begin-subexpression. We assume there is at
+ after an alternative or a begin-subexpression. Assume there is at
least one character before the ^. */
-static boolean
-at_begline_loc_p (re_char *pattern, re_char *p, reg_syntax_t syntax)
+static bool
+at_begline_loc_p (re_char *pattern, re_char *p)
{
re_char *prev = p - 2;
- boolean odd_backslashes;
-
- /* After a subexpression? */
- if (*prev == '(')
- odd_backslashes = (syntax & RE_NO_BK_PARENS) == 0;
- /* After an alternative? */
- else if (*prev == '|')
- odd_backslashes = (syntax & RE_NO_BK_VBAR) == 0;
-
- /* After a shy subexpression? */
- else if (*prev == ':' && (syntax & RE_SHY_GROUPS))
+ switch (*prev)
{
+ case '(': /* After a subexpression. */
+ case '|': /* After an alternative. */
+ break;
+
+ case ':': /* After a shy subexpression. */
/* Skip over optional regnum. */
- while (prev - 1 >= pattern && prev[-1] >= '0' && prev[-1] <= '9')
+ while (prev > pattern && '0' <= prev[-1] && prev[-1] <= '9')
--prev;
- if (!(prev - 2 >= pattern
- && prev[-1] == '?' && prev[-2] == '('))
+ if (! (prev > pattern + 1 && prev[-1] == '?' && prev[-2] == '('))
return false;
prev -= 2;
- odd_backslashes = (syntax & RE_NO_BK_PARENS) == 0;
+ break;
+
+ default:
+ return false;
}
- else
- return false;
/* Count the number of preceding backslashes. */
p = prev;
- while (prev - 1 >= pattern && prev[-1] == '\\')
+ while (prev > pattern && prev[-1] == '\\')
--prev;
- return (p - prev) & odd_backslashes;
+ return (p - prev) & 1;
}
-/* The dual of at_begline_loc_p. This one is for $. We assume there is
- at least one character after the $, i.e., `P < PEND'. */
+/* The dual of at_begline_loc_p. This one is for $. Assume there is
+ at least one character after the $, i.e., 'P < PEND'. */
-static boolean
-at_endline_loc_p (re_char *p, re_char *pend, reg_syntax_t syntax)
+static bool
+at_endline_loc_p (re_char *p, re_char *pend)
{
- re_char *next = p;
- boolean next_backslash = *next == '\\';
- re_char *next_next = p + 1 < pend ? p + 1 : 0;
-
- return
- /* Before a subexpression? */
- (syntax & RE_NO_BK_PARENS ? *next == ')'
- : next_backslash && next_next && *next_next == ')')
- /* Before an alternative? */
- || (syntax & RE_NO_BK_VBAR ? *next == '|'
- : next_backslash && next_next && *next_next == '|');
+ /* Before a subexpression or an alternative? */
+ return *p == '\\' && p + 1 < pend && (p[1] == ')' || p[1] == '|');
}
/* Returns true if REGNUM is in one of COMPILE_STACK's elements and
false if it's not. */
-static boolean
+static bool
group_in_compile_stack (compile_stack_type compile_stack, regnum_t regnum)
{
- ssize_t this_element;
+ ptrdiff_t this_element;
for (this_element = compile_stack.avail - 1;
this_element >= 0;
@@ -3822,35 +2810,35 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
const int multibyte)
{
int j, k;
- boolean not;
+ bool not;
/* If all elements for base leading-codes in fastmap is set, this
flag is set true. */
- boolean match_any_multibyte_characters = false;
+ bool match_any_multibyte_characters = false;
- assert (p);
+ eassert (p);
/* The loop below works as follows:
- It has a working-list kept in the PATTERN_STACK and which basically
starts by only containing a pointer to the first operation.
- If the opcode we're looking at is a match against some set of
chars, then we add those chars to the fastmap and go on to the
- next work element from the worklist (done via `break').
+ next work element from the worklist (done via 'break').
- If the opcode is a control operator on the other hand, we either
- ignore it (if it's meaningless at this point, such as `start_memory')
+ ignore it (if it's meaningless at this point, such as 'start_memory')
or execute it (if it's a jump). If the jump has several destinations
- (i.e. `on_failure_jump'), then we push the other destination onto the
+ (i.e. 'on_failure_jump'), then we push the other destination onto the
worklist.
We guarantee termination by ignoring backward jumps (more or less),
- so that `p' is monotonically increasing. More to the point, we
- never set `p' (or push) anything `<= p1'. */
+ so that P is monotonically increasing. More to the point, we
+ never set P (or push) anything '<= p1'. */
while (p < pend)
{
- /* `p1' is used as a marker of how far back a `on_failure_jump'
- can go without being ignored. It is normally equal to `p'
- (which prevents any backward `on_failure_jump') except right
- after a plain `jump', to allow patterns such as:
+ /* P1 is used as a marker of how far back a 'on_failure_jump'
+ can go without being ignored. It is normally equal to P
+ (which prevents any backward 'on_failure_jump') except right
+ after a plain 'jump', to allow patterns such as:
0: jump 10
3..9: <body>
10: on_failure_jump 3
@@ -3872,7 +2860,7 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
/* Following are the cases which match a character. These end
- with `break'. */
+ with 'break'. */
case exactn:
if (fastmap)
@@ -3919,7 +2907,6 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
if (!!(p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH))) ^ not)
fastmap[j] = 1;
-#ifdef emacs
if (/* Any leading code can possibly start a character
which doesn't match the specified set of characters. */
not
@@ -3947,7 +2934,7 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
int c, count;
unsigned char lc1, lc2;
- /* Make P points the range table. `+ 2' is to skip flag
+ /* Make P points the range table. '+ 2' is to skip flag
bits for a character class. */
p += CHARSET_BITMAP_SIZE (&p[-2]) + 2;
@@ -3965,20 +2952,11 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
fastmap[j] = 1;
}
}
-#endif
break;
case syntaxspec:
case notsyntaxspec:
if (!fastmap) break;
-#ifndef emacs
- not = (re_opcode_t)p[-1] == notsyntaxspec;
- k = *p++;
- for (j = 0; j < (1 << BYTEWIDTH); j++)
- if ((SYNTAX (j) == (enum syntaxcode) k) ^ not)
- fastmap[j] = 1;
- break;
-#else /* emacs */
/* This match depends on text properties. These end with
aborting optimizations. */
return -1;
@@ -4004,10 +2982,9 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
break;
/* All cases after this match the empty string. These end with
- `continue'. */
+ 'continue'. */
case at_dot:
-#endif /* !emacs */
case no_op:
case begline:
case endline:
@@ -4026,7 +3003,7 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
EXTRACT_NUMBER_AND_INCR (j, p);
if (j < 0)
/* Backward jumps can only go back to code that we've already
- visited. `re_compile' should make sure this is true. */
+ visited. 're_compile' should make sure this is true. */
break;
p += j;
switch (*p)
@@ -4041,7 +3018,7 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
default:
continue;
};
- /* Keep `p1' to allow the `on_failure_jump' we are jumping to
+ /* Keep P1 to allow the 'on_failure_jump' we are jumping to
to jump back to "just after here". */
FALLTHROUGH;
case on_failure_jump:
@@ -4065,7 +3042,7 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
case jump_n:
/* This code simply does not properly handle forward jump_n. */
- DEBUG_STATEMENT (EXTRACT_NUMBER (j, p); assert (j < 0));
+ DEBUG_STATEMENT (EXTRACT_NUMBER (j, p); eassert (j < 0));
p += 4;
/* jump_n can either jump or fall through. The (backward) jump
case has already been handled, so we only need to look at the
@@ -4074,7 +3051,7 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
case succeed_n:
/* If N == 0, it should be an on_failure_jump_loop instead. */
- DEBUG_STATEMENT (EXTRACT_NUMBER (j, p + 2); assert (j > 0));
+ DEBUG_STATEMENT (EXTRACT_NUMBER (j, p + 2); eassert (j > 0));
p += 4;
/* We only care about one iteration of the loop, so we don't
need to consider the case where this behaves like an
@@ -4108,8 +3085,8 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
} /* analyze_first */
-/* re_compile_fastmap computes a ``fastmap'' for the compiled pattern in
- BUFP. A fastmap records which of the (1 << BYTEWIDTH) possible
+/* Compute a fastmap for the compiled pattern in BUFP.
+ A fastmap records which of the (1 << BYTEWIDTH) possible
characters can start a string that matches the pattern. This fastmap
is used by re_search to skip quickly over impossible starting points.
@@ -4120,18 +3097,16 @@ analyze_first (re_char *p, re_char *pend, char *fastmap,
The caller must supply the address of a (1 << BYTEWIDTH)-byte data
area as BUFP->fastmap.
- We set the `fastmap', `fastmap_accurate', and `can_be_null' fields in
- the pattern buffer.
+ Set the 'fastmap', 'fastmap_accurate', and 'can_be_null' fields in
+ the pattern buffer. */
- Returns 0 if we succeed, -2 if an internal error. */
-
-int
+static void
re_compile_fastmap (struct re_pattern_buffer *bufp)
{
char *fastmap = bufp->fastmap;
int analysis;
- assert (fastmap && bufp->buffer);
+ eassert (fastmap && bufp->buffer);
memset (fastmap, 0, 1 << BYTEWIDTH); /* Assume nothing's valid. */
bufp->fastmap_accurate = 1; /* It will be when we're done. */
@@ -4139,14 +3114,13 @@ re_compile_fastmap (struct re_pattern_buffer *bufp)
analysis = analyze_first (bufp->buffer, bufp->buffer + bufp->used,
fastmap, RE_MULTIBYTE_P (bufp));
bufp->can_be_null = (analysis != 0);
- return 0;
} /* re_compile_fastmap */
/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
ENDS. Subsequent matches using PATTERN_BUFFER and REGS will use
this memory for recording register information. STARTS and ENDS
must be allocated using the malloc library routine, and must each
- be at least NUM_REGS * sizeof (regoff_t) bytes long.
+ be at least NUM_REGS * sizeof (ptrdiff_t) bytes long.
If NUM_REGS == 0, then subsequent matches should allocate their own
register data.
@@ -4156,7 +3130,8 @@ re_compile_fastmap (struct re_pattern_buffer *bufp)
freeing the old data. */
void
-re_set_registers (struct re_pattern_buffer *bufp, struct re_registers *regs,
unsigned int num_regs, regoff_t *starts, regoff_t *ends)
+re_set_registers (struct re_pattern_buffer *bufp, struct re_registers *regs,
+ unsigned int num_regs, ptrdiff_t *starts, ptrdiff_t *ends)
{
if (num_regs)
{
@@ -4172,21 +3147,19 @@ re_set_registers (struct re_pattern_buffer *bufp,
struct re_registers *regs, uns
regs->start = regs->end = 0;
}
}
-WEAK_ALIAS (__re_set_registers, re_set_registers)
/* Searching routines. */
/* Like re_search_2, below, but only one string is specified, and
doesn't let you say where to stop matching. */
-regoff_t
+ptrdiff_t
re_search (struct re_pattern_buffer *bufp, const char *string, size_t size,
- ssize_t startpos, ssize_t range, struct re_registers *regs)
+ ptrdiff_t startpos, ptrdiff_t range, struct re_registers *regs)
{
return re_search_2 (bufp, NULL, 0, string, size, startpos, range,
regs, size);
}
-WEAK_ALIAS (__re_search, re_search)
/* Head address of virtual concatenation of string. */
#define HEAD_ADDR_VSTRING(P) \
@@ -4213,25 +3186,26 @@ WEAK_ALIAS (__re_search, re_search)
Do not consider matching one past the index STOP in the virtual
concatenation of STRING1 and STRING2.
- We return either the position in the strings at which the match was
+ Return either the position in the strings at which the match was
found, -1 if no match, or -2 if error (such as failure
stack overflow). */
-regoff_t
+ptrdiff_t
re_search_2 (struct re_pattern_buffer *bufp, const char *str1, size_t size1,
- const char *str2, size_t size2, ssize_t startpos, ssize_t range,
- struct re_registers *regs, ssize_t stop)
+ const char *str2, size_t size2,
+ ptrdiff_t startpos, ptrdiff_t range,
+ struct re_registers *regs, ptrdiff_t stop)
{
- regoff_t val;
+ ptrdiff_t val;
re_char *string1 = (re_char *) str1;
re_char *string2 = (re_char *) str2;
- register char *fastmap = bufp->fastmap;
- register RE_TRANSLATE_TYPE translate = bufp->translate;
+ char *fastmap = bufp->fastmap;
+ Lisp_Object translate = bufp->translate;
size_t total_size = size1 + size2;
- ssize_t endpos = startpos + range;
- boolean anchored_start;
+ ptrdiff_t endpos = startpos + range;
+ bool anchored_start;
/* Nonzero if we are searching multibyte string. */
- const boolean multibyte = RE_TARGET_MULTIBYTE_P (bufp);
+ bool multibyte = RE_TARGET_MULTIBYTE_P (bufp);
/* Check for out-of-range STARTPOS. */
if (startpos < 0 || startpos > total_size)
@@ -4255,7 +3229,6 @@ re_search_2 (struct re_pattern_buffer *bufp, const char
*str1, size_t size1,
range = 0;
}
-#ifdef emacs
/* In a forward search for something that starts with \=.
don't keep searching past point. */
if (bufp->used > 0 && (re_opcode_t) bufp->buffer[0] == at_dot && range > 0)
@@ -4264,7 +3237,6 @@ re_search_2 (struct re_pattern_buffer *bufp, const char
*str1, size_t size1,
if (range < 0)
return -1;
}
-#endif /* emacs */
/* Update the fastmap now if not correct already. */
if (fastmap && !bufp->fastmap_accurate)
@@ -4273,21 +3245,19 @@ re_search_2 (struct re_pattern_buffer *bufp, const char
*str1, size_t size1,
/* See whether the pattern is anchored. */
anchored_start = (bufp->buffer[0] == begline);
-#ifdef emacs
gl_state.object = re_match_object; /* Used by SYNTAX_TABLE_BYTE_TO_CHAR. */
{
- ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (POS_AS_IN_BUFFER (startpos));
+ ptrdiff_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (POS_AS_IN_BUFFER
(startpos));
SETUP_SYNTAX_TABLE_FOR_OBJECT (re_match_object, charpos, 1);
}
-#endif
/* Loop through the string, looking for a place to start matching. */
for (;;)
{
/* If the pattern is anchored,
skip quickly past places we cannot match.
- We don't bother to treat startpos == 0 specially
+ Don't bother to treat startpos == 0 specially
because that case doesn't repeat. */
if (anchored_start && startpos > 0)
{
@@ -4303,21 +3273,21 @@ re_search_2 (struct re_pattern_buffer *bufp, const char
*str1, size_t size1,
the first null string. */
if (fastmap && startpos < total_size && !bufp->can_be_null)
{
- register re_char *d;
- register re_wchar_t buf_ch;
+ re_char *d;
+ int buf_ch;
d = POS_ADDR_VSTRING (startpos);
if (range > 0) /* Searching forwards. */
{
- ssize_t irange = range, lim = 0;
+ ptrdiff_t irange = range, lim = 0;
if (startpos < size1 && startpos + range >= size1)
lim = range - (size1 - startpos);
- /* Written out as an if-else to avoid testing `translate'
+ /* Written out as an if-else to avoid testing 'translate'
inside the loop. */
- if (RE_TRANSLATE_P (translate))
+ if (!NILP (translate))
{
if (multibyte)
while (range > lim)
@@ -4335,11 +3305,9 @@ re_search_2 (struct re_pattern_buffer *bufp, const char
*str1, size_t size1,
else
while (range > lim)
{
- register re_wchar_t ch, translated;
-
buf_ch = *d;
- ch = RE_CHAR_TO_MULTIBYTE (buf_ch);
- translated = RE_TRANSLATE (translate, ch);
+ int ch = RE_CHAR_TO_MULTIBYTE (buf_ch);
+ int translated = RE_TRANSLATE (translate, ch);
if (translated != ch
&& (ch = RE_CHAR_TO_UNIBYTE (translated)) >= 0)
buf_ch = ch;
@@ -4382,11 +3350,9 @@ re_search_2 (struct re_pattern_buffer *bufp, const char
*str1, size_t size1,
}
else
{
- register re_wchar_t ch, translated;
-
buf_ch = *d;
- ch = RE_CHAR_TO_MULTIBYTE (buf_ch);
- translated = TRANSLATE (ch);
+ int ch = RE_CHAR_TO_MULTIBYTE (buf_ch);
+ int translated = TRANSLATE (ch);
if (translated != ch
&& (ch = RE_CHAR_TO_UNIBYTE (translated)) >= 0)
buf_ch = ch;
@@ -4456,17 +3422,16 @@ re_search_2 (struct re_pattern_buffer *bufp, const char
*str1, size_t size1,
}
return -1;
} /* re_search_2 */
-WEAK_ALIAS (__re_search_2, re_search_2)
/* Declarations and macros for re_match_2. */
static int bcmp_translate (re_char *s1, re_char *s2,
- register ssize_t len,
- RE_TRANSLATE_TYPE translate,
+ ptrdiff_t len,
+ Lisp_Object translate,
const int multibyte);
-/* This converts PTR, a pointer into one of the search strings `string1'
- and `string2' into an offset from the beginning of that string. */
+/* This converts PTR, a pointer into one of the search strings 'string1'
+ and 'string2' into an offset from the beginning of that string. */
#define POINTER_TO_OFFSET(ptr) \
(FIRST_STRING_P (ptr) \
? (ptr) - string1 \
@@ -4490,7 +3455,7 @@ static int bcmp_translate (re_char *s1, re_char *s2,
/* Call before fetching a char with *d if you already checked other limits.
This is meant for use in lookahead operations like wordend, etc..
where we might need to look at parts of the string that might be
- outside of the LIMITs (i.e past `stop'). */
+ outside of the LIMITs (i.e past 'stop'). */
#define PREFETCH_NOLIMIT() \
if (d == end1) \
{ \
@@ -4499,7 +3464,7 @@ static int bcmp_translate (re_char *s1, re_char *s2,
} \
/* Test if at very beginning or at very end of the virtual concatenation
- of `string1' and `string2'. If only one string, it's `string2'. */
+ of STRING1 and STRING2. If only one string, it's STRING2. */
#define AT_STRINGS_BEG(d) ((d) == (size1 ? string1 : string2) || !size2)
#define AT_STRINGS_END(d) ((d) == end2)
@@ -4530,29 +3495,6 @@ static int bcmp_translate (re_char *s1, re_char *s2,
|| WORDCHAR_P (d - 1) != WORDCHAR_P (d))
#endif
-/* Free everything we malloc. */
-#ifdef MATCH_MAY_ALLOCATE
-# define FREE_VAR(var) \
- do { \
- if (var) \
- {
\
- REGEX_FREE (var); \
- var = NULL; \
- }
\
- } while (0)
-# define FREE_VARIABLES() \
- do { \
- REGEX_FREE_STACK (fail_stack.stack); \
- FREE_VAR (regstart); \
- FREE_VAR (regend); \
- FREE_VAR (best_regstart); \
- FREE_VAR (best_regend); \
- REGEX_SAFE_FREE ();
\
- } while (0)
-#else
-# define FREE_VARIABLES() ((void)0) /* Do nothing! But inhibit gcc warning.
*/
-#endif /* not MATCH_MAY_ALLOCATE */
-
/* Optimization routines. */
@@ -4585,10 +3527,8 @@ skip_one_char (re_char *p)
case syntaxspec:
case notsyntaxspec:
-#ifdef emacs
case categoryspec:
case notcategoryspec:
-#endif /* emacs */
p++;
break;
@@ -4622,7 +3562,7 @@ skip_noops (re_char *p, re_char *pend)
return p;
}
}
- assert (p == pend);
+ eassert (p == pend);
return p;
}
@@ -4649,17 +3589,16 @@ execute_charset (re_char **pp, unsigned c, unsigned
corig, bool unibyte)
if (unibyte && c < (1 << BYTEWIDTH))
{ /* Lookup bitmap. */
- /* Cast to `unsigned' instead of `unsigned char' in
+ /* Cast to 'unsigned' instead of 'unsigned char' in
case the bit list is a full 32 bytes long. */
if (c < (unsigned) (CHARSET_BITMAP_SIZE (p) * BYTEWIDTH)
&& p[2 + c / BYTEWIDTH] & (1 << (c % BYTEWIDTH)))
return !not;
}
-#ifdef emacs
else if (rtp)
{
int class_bits = CHARSET_RANGE_TABLE_BITS (p);
- re_wchar_t range_start, range_end;
+ int range_start, range_end;
/* Sort tests by the most commonly used classes with some adjustment to which
tests are easiest to perform. Take a look at comment in re_wctype_parse
@@ -4690,7 +3629,7 @@ execute_charset (re_char **pp, unsigned c, unsigned
corig, bool unibyte)
return !not;
}
}
-#endif /* emacs */
+
return not;
}
@@ -4700,11 +3639,11 @@ mutually_exclusive_p (struct re_pattern_buffer *bufp,
re_char *p1,
re_char *p2)
{
re_opcode_t op2;
- const boolean multibyte = RE_MULTIBYTE_P (bufp);
+ bool multibyte = RE_MULTIBYTE_P (bufp);
unsigned char *pend = bufp->buffer + bufp->used;
- assert (p1 >= bufp->buffer && p1 < pend
- && p2 >= bufp->buffer && p2 <= pend);
+ eassert (p1 >= bufp->buffer && p1 < pend
+ && p2 >= bufp->buffer && p2 <= pend);
/* Skip over open/close-group commands.
If what follows this loop is a ...+ construct,
@@ -4715,8 +3654,8 @@ mutually_exclusive_p (struct re_pattern_buffer *bufp,
re_char *p1,
is only used in the case where p1 is a simple match operator. */
/* p1 = skip_noops (p1, pend); */
- assert (p1 >= bufp->buffer && p1 < pend
- && p2 >= bufp->buffer && p2 <= pend);
+ eassert (p1 >= bufp->buffer && p1 < pend
+ && p2 >= bufp->buffer && p2 <= pend);
op2 = p2 == pend ? succeed : *p2;
@@ -4735,7 +3674,7 @@ mutually_exclusive_p (struct re_pattern_buffer *bufp,
re_char *p1,
case endline:
case exactn:
{
- register re_wchar_t c
+ int c
= (re_opcode_t) *p2 == endline ? '\n'
: RE_STRING_CHAR (p2 + 2, multibyte);
@@ -4751,7 +3690,7 @@ mutually_exclusive_p (struct re_pattern_buffer *bufp,
re_char *p1,
else if ((re_opcode_t) *p1 == charset
|| (re_opcode_t) *p1 == charset_not)
{
- if (!execute_charset (&p1, c, c, !multibyte || IS_REAL_ASCII (c)))
+ if (!execute_charset (&p1, c, c, !multibyte || ASCII_CHAR_P (c)))
{
DEBUG_PRINT (" No match => fast loop.\n");
return 1;
@@ -4778,10 +3717,10 @@ mutually_exclusive_p (struct re_pattern_buffer *bufp,
re_char *p1,
else if (!multibyte || !CHARSET_RANGE_TABLE_EXISTS_P (p2))
{
/* Now, we are sure that P2 has no range table.
- So, for the size of bitmap in P2, `p2[1]' is
+ So, for the size of bitmap in P2, 'p2[1]' is
enough. But P1 may have range table, so the
size of bitmap table of P1 is extracted by
- using macro `CHARSET_BITMAP_SIZE'.
+ using macro 'CHARSET_BITMAP_SIZE'.
In a multibyte case, we know that all the character
listed in P2 is ASCII. In a unibyte case, P1 has only a
@@ -4865,12 +3804,10 @@ mutually_exclusive_p (struct re_pattern_buffer *bufp,
re_char *p1,
|| (re_opcode_t) *p1 == syntaxspec)
&& p1[1] == Sword);
-#ifdef emacs
case categoryspec:
return ((re_opcode_t) *p1 == notcategoryspec && p1[1] == p2[1]);
case notcategoryspec:
return ((re_opcode_t) *p1 == categoryspec && p1[1] == p2[1]);
-#endif /* emacs */
default:
;
@@ -4883,61 +3820,43 @@ mutually_exclusive_p (struct re_pattern_buffer *bufp,
re_char *p1,
/* Matching routines. */
-#ifndef emacs /* Emacs never uses this. */
-/* re_match is like re_match_2 except it takes only a single string. */
-
-regoff_t
-re_match (struct re_pattern_buffer *bufp, const char *string,
- size_t size, ssize_t pos, struct re_registers *regs)
-{
- regoff_t result = re_match_2_internal (bufp, NULL, 0, (re_char *) string,
- size, pos, regs, size);
- return result;
-}
-WEAK_ALIAS (__re_match, re_match)
-#endif /* not emacs */
-
/* re_match_2 matches the compiled pattern in BUFP against the
the (virtual) concatenation of STRING1 and STRING2 (of length SIZE1
and SIZE2, respectively). We start matching at POS, and stop
matching at STOP.
- If REGS is non-null and the `no_sub' field of BUFP is nonzero, we
- store offsets for the substring each group matched in REGS. See the
- documentation for exactly how many groups we fill.
+ If REGS is non-null, store offsets for the substring each group
+ matched in REGS.
We return -1 if no match, -2 if an internal error (such as the
failure stack overflowing). Otherwise, we return the length of the
matched substring. */
-regoff_t
+ptrdiff_t
re_match_2 (struct re_pattern_buffer *bufp, const char *string1,
- size_t size1, const char *string2, size_t size2, ssize_t pos,
- struct re_registers *regs, ssize_t stop)
+ size_t size1, const char *string2, size_t size2, ptrdiff_t pos,
+ struct re_registers *regs, ptrdiff_t stop)
{
- regoff_t result;
+ ptrdiff_t result;
-#ifdef emacs
- ssize_t charpos;
+ ptrdiff_t charpos;
gl_state.object = re_match_object; /* Used by SYNTAX_TABLE_BYTE_TO_CHAR. */
charpos = SYNTAX_TABLE_BYTE_TO_CHAR (POS_AS_IN_BUFFER (pos));
SETUP_SYNTAX_TABLE_FOR_OBJECT (re_match_object, charpos, 1);
-#endif
result = re_match_2_internal (bufp, (re_char *) string1, size1,
(re_char *) string2, size2,
pos, regs, stop);
return result;
}
-WEAK_ALIAS (__re_match_2, re_match_2)
/* This is a separate function so that we can force an alloca cleanup
afterwards. */
-static regoff_t
+static ptrdiff_t
re_match_2_internal (struct re_pattern_buffer *bufp, re_char *string1,
size_t size1, re_char *string2, size_t size2,
- ssize_t pos, struct re_registers *regs, ssize_t stop)
+ ptrdiff_t pos, struct re_registers *regs, ptrdiff_t stop)
{
/* General temporaries. */
int mcnt;
@@ -4964,13 +3883,13 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
re_char *pend = p + bufp->used;
/* We use this to map every character in the string. */
- RE_TRANSLATE_TYPE translate = bufp->translate;
+ Lisp_Object translate = bufp->translate;
- /* Nonzero if BUFP is setup from a multibyte regex. */
- const boolean multibyte = RE_MULTIBYTE_P (bufp);
+ /* True if BUFP is setup from a multibyte regex. */
+ bool multibyte = RE_MULTIBYTE_P (bufp);
- /* Nonzero if STRING1/STRING2 are multibyte. */
- const boolean target_multibyte = RE_TARGET_MULTIBYTE_P (bufp);
+ /* True if STRING1/STRING2 are multibyte. */
+ bool target_multibyte = RE_TARGET_MULTIBYTE_P (bufp);
/* Failure point stack. Each place that can handle a failure further
down the line pushes a failure point on this stack. It consists of
@@ -4979,19 +3898,11 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
registers, and, finally, two char *'s. The first char * is where
to resume scanning the pattern; the second one is where to resume
scanning the strings. */
-#ifdef MATCH_MAY_ALLOCATE /* otherwise, this is global. */
fail_stack_type fail_stack;
-#endif
#ifdef DEBUG_COMPILES_ARGUMENTS
unsigned nfailure_points_pushed = 0, nfailure_points_popped = 0;
#endif
-#if defined REL_ALLOC && defined REGEX_MALLOC
- /* This holds the pointer to the failure stack, when
- it is allocated relocatably. */
- fail_stack_elt_t *failure_stack_ptr;
-#endif
-
/* We fill all the registers internally, independent of what we
return, for use in backreferences. The number here includes
an element for register zero. */
@@ -5004,24 +3915,20 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
matching and the regnum-th regend points to right after where we
stopped matching the regnum-th subexpression. (The zeroth register
keeps track of what the whole pattern matches.) */
-#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */
- re_char **regstart, **regend;
-#endif
+ re_char **regstart UNINIT, **regend UNINIT;
/* The following record the register info as found in the above
variables when we find a match better than any we've seen before.
This happens as we backtrack through the failure points, which in
turn happens only if we have not yet matched the entire string. */
unsigned best_regs_set = false;
-#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */
- re_char **best_regstart, **best_regend;
-#endif
+ re_char **best_regstart UNINIT, **best_regend UNINIT;
- /* Logically, this is `best_regend[0]'. But we don't want to have to
+ /* Logically, this is 'best_regend[0]'. But we don't want to have to
allocate space for that if we're not allocating space for anything
else (see below). Also, we never need info about register 0 for
any of the other register vectors, and it seems rather a kludge to
- treat `best_regend' differently than the rest. So we keep track of
+ treat 'best_regend' differently than the rest. So we keep track of
the end of the best match so far in a separate variable. We
initialize this to NULL so that when we backtrack the first time
and need to test it, it's not garbage. */
@@ -5038,7 +3945,6 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
INIT_FAIL_STACK ();
-#ifdef MATCH_MAY_ALLOCATE
/* Do not bother to initialize all the register variables if there are
no groups in the pattern, as it takes a fair amount of time. If
there are groups, we include space for register 0 (the whole
@@ -5046,29 +3952,16 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
array indexing. We should fix this. */
if (bufp->re_nsub)
{
- regstart = REGEX_TALLOC (num_regs, re_char *);
- regend = REGEX_TALLOC (num_regs, re_char *);
- best_regstart = REGEX_TALLOC (num_regs, re_char *);
- best_regend = REGEX_TALLOC (num_regs, re_char *);
-
- if (!(regstart && regend && best_regstart && best_regend))
- {
- FREE_VARIABLES ();
- return -2;
- }
- }
- else
- {
- /* We must initialize all our variables to NULL, so that
- `FREE_VARIABLES' doesn't try to free them. */
- regstart = regend = best_regstart = best_regend = NULL;
+ regstart = SAFE_ALLOCA (num_regs * 4 * sizeof *regstart);
+ regend = regstart + num_regs;
+ best_regstart = regend + num_regs;
+ best_regend = best_regstart + num_regs;
}
-#endif /* MATCH_MAY_ALLOCATE */
/* The starting position is bogus. */
if (pos < 0 || pos > size1 + size2)
{
- FREE_VARIABLES ();
+ SAFE_FREE ();
return -1;
}
@@ -5078,8 +3971,8 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
for (reg = 1; reg < num_regs; reg++)
regstart[reg] = regend[reg] = NULL;
- /* We move `string1' into `string2' if the latter's empty -- but not if
- `string1' is null. */
+ /* We move 'string1' into 'string2' if the latter's empty -- but not if
+ 'string1' is null. */
if (size2 == 0 && string1 != NULL)
{
string2 = string1;
@@ -5090,12 +3983,12 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
end1 = string1 + size1;
end2 = string2 + size2;
- /* `p' scans through the pattern as `d' scans through the data.
- `dend' is the end of the input string that `d' points within. `d'
- is advanced into the following input string whenever necessary, but
+ /* P scans through the pattern as D scans through the data.
+ DEND is the end of the input string that D points within.
+ Advance D into the following input string whenever necessary, but
this happens before fetching; therefore, at the beginning of the
- loop, `d' can be pointing at the end of a string, but it cannot
- equal `string2'. */
+ loop, D can be pointing at the end of a string, but it cannot
+ equal STRING2. */
if (pos >= size1)
{
/* Only match within string2. */
@@ -5112,7 +4005,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
/* BEWARE!
When we reach end_match_1, PREFETCH normally switches to string2.
But in the present case, this means that just doing a PREFETCH
- makes us jump from `stop' to `gap' within the string.
+ makes us jump from 'stop' to 'gap' within the string.
What we really want here is for the search to stop as
soon as we hit end_match_1. That's why we set end_match_2
to end_match_1 (since PREFETCH fails as soon as we hit
@@ -5120,8 +4013,8 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
end_match_2 = end_match_1;
}
else
- { /* It's important to use this code when stop == size so that
- moving `d' from end1 to string2 will not prevent the d == dend
+ { /* It's important to use this code when STOP == SIZE so that
+ moving D from end1 to string2 will not prevent the D == DEND
check from catching the end of string. */
end_match_1 = end1;
end_match_2 = string2 + stop - size1;
@@ -5197,10 +4090,10 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
else if (best_regs_set && !best_match_p)
{
restore_best_regs:
- /* Restore best match. It may happen that `dend ==
+ /* Restore best match. It may happen that 'dend ==
end_match_1' while the restored d is in string2.
- For example, the pattern `x.*y.*z' against the
- strings `x-' and `y-z-', if the two strings are
+ For example, the pattern 'x.*y.*z' against the
+ strings 'x-' and 'y-z-', if the two strings are
not consecutive in memory. */
DEBUG_PRINT ("Restoring best registers.\n");
@@ -5220,21 +4113,16 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
DEBUG_PRINT ("Accepting match.\n");
/* If caller wants register contents data back, do it. */
- if (regs && !bufp->no_sub)
+ if (regs)
{
/* Have the register data arrays been allocated? */
if (bufp->regs_allocated == REGS_UNALLOCATED)
{ /* No. So allocate them with malloc. We need one
- extra element beyond `num_regs' for the `-1' marker
+ extra element beyond 'num_regs' for the '-1' marker
GNU code uses. */
regs->num_regs = max (RE_NREGS, num_regs + 1);
- regs->start = TALLOC (regs->num_regs, regoff_t);
- regs->end = TALLOC (regs->num_regs, regoff_t);
- if (regs->start == NULL || regs->end == NULL)
- {
- FREE_VARIABLES ();
- return -2;
- }
+ regs->start = TALLOC (regs->num_regs, ptrdiff_t);
+ regs->end = TALLOC (regs->num_regs, ptrdiff_t);
bufp->regs_allocated = REGS_REALLOCATE;
}
else if (bufp->regs_allocated == REGS_REALLOCATE)
@@ -5244,23 +4132,14 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
if (regs->num_regs < num_regs + 1)
{
regs->num_regs = num_regs + 1;
- RETALLOC (regs->start, regs->num_regs, regoff_t);
- RETALLOC (regs->end, regs->num_regs, regoff_t);
- if (regs->start == NULL || regs->end == NULL)
- {
- FREE_VARIABLES ();
- return -2;
- }
+ RETALLOC (regs->start, regs->num_regs, ptrdiff_t);
+ RETALLOC (regs->end, regs->num_regs, ptrdiff_t);
}
}
else
- {
- /* These braces fend off a "empty body in an else-statement"
- warning under GCC when assert expands to nothing. */
- assert (bufp->regs_allocated == REGS_FIXED);
- }
+ eassert (bufp->regs_allocated == REGS_FIXED);
- /* Convert the pointer data in `regstart' and `regend' to
+ /* Convert the pointer data in 'regstart' and 'regend' to
indices. Register zero has to be set differently,
since we haven't kept track of any info for it. */
if (regs->num_regs > 0)
@@ -5269,7 +4148,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
regs->end[0] = POINTER_TO_OFFSET (d);
}
- /* Go through the first `min (num_regs, regs->num_regs)'
+ /* Go through the first 'min (num_regs, regs->num_regs)'
registers, since that is all we initialized. */
for (reg = 1; reg < min (num_regs, regs->num_regs); reg++)
{
@@ -5289,7 +4168,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
-1 at the end. */
for (reg = num_regs; reg < regs->num_regs; reg++)
regs->start[reg] = regs->end[reg] = -1;
- } /* regs && !bufp->no_sub */
+ }
DEBUG_PRINT ("%u failure points pushed, %u popped (%u remain).\n",
nfailure_points_pushed, nfailure_points_popped,
@@ -5300,7 +4179,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
DEBUG_PRINT ("Returning %td from re_match_2.\n", dcnt);
- FREE_VARIABLES ();
+ SAFE_FREE ();
return dcnt;
}
@@ -5327,34 +4206,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
/* Remember the start point to rollback upon failure. */
dfail = d;
-#ifndef emacs
- /* This is written out as an if-else so we don't waste time
- testing `translate' inside the loop. */
- if (RE_TRANSLATE_P (translate))
- do
- {
- PREFETCH ();
- if (RE_TRANSLATE (translate, *d) != *p++)
- {
- d = dfail;
- goto fail;
- }
- d++;
- }
- while (--mcnt);
- else
- do
- {
- PREFETCH ();
- if (*d++ != *p++)
- {
- d = dfail;
- goto fail;
- }
- }
- while (--mcnt);
-#else /* emacs */
- /* The cost of testing `translate' is comparatively small. */
+ /* The cost of testing 'translate' is comparatively small. */
if (target_multibyte)
do
{
@@ -5418,16 +4270,15 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
d++;
}
while (--mcnt);
-#endif
+
break;
- /* Match any character except possibly a newline or a null. */
+ /* Match any character except newline. */
case anychar:
{
int buf_charlen;
- re_wchar_t buf_ch;
- reg_syntax_t syntax;
+ int buf_ch;
DEBUG_PRINT ("EXECUTING anychar.\n");
@@ -5435,15 +4286,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
buf_ch = RE_STRING_CHAR_AND_LENGTH (d, buf_charlen,
target_multibyte);
buf_ch = TRANSLATE (buf_ch);
-
-#ifdef emacs
- syntax = RE_SYNTAX_EMACS;
-#else
- syntax = bufp->syntax;
-#endif
-
- if ((!(syntax & RE_DOT_NEWLINE) && buf_ch == '\n')
- || ((syntax & RE_DOT_NOT_NULL) && buf_ch == '\000'))
+ if (buf_ch == '\n')
goto fail;
DEBUG_PRINT (" Matched \"%d\".\n", *d);
@@ -5459,7 +4302,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
int len;
/* Whether matching against a unibyte character. */
- boolean unibyte_char = false;
+ bool unibyte_char = false;
DEBUG_PRINT ("EXECUTING charset%s.\n",
(re_opcode_t) *(p - 1) == charset_not ? "_not" : "");
@@ -5529,10 +4372,10 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
case stop_memory:
DEBUG_PRINT ("EXECUTING stop_memory %d:\n", *p);
- assert (!REG_UNSET (regstart[*p]));
+ eassert (!REG_UNSET (regstart[*p]));
/* Strictly speaking, there should be code such as:
- assert (REG_UNSET (regend[*p]));
+ eassert (REG_UNSET (regend[*p]));
PUSH_FAILURE_REGSTOP ((unsigned int)*p);
But the only info to be pushed is regend[*p] and it is known to
@@ -5552,11 +4395,11 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
break;
- /* \<digit> has been turned into a `duplicate' command which is
+ /* \<digit> has been turned into a 'duplicate' command which is
followed by the numeric value of <digit> as the register number. */
case duplicate:
{
- register re_char *d2, *dend2;
+ re_char *d2, *dend2;
int regno = *p++; /* Get which register to match against. */
DEBUG_PRINT ("EXECUTING duplicate %d.\n", regno);
@@ -5609,7 +4452,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
/* Compare that many; failure if mismatch, else move
past them. */
- if (RE_TRANSLATE_P (translate)
+ if (!NILP (translate)
? bcmp_translate (d, d2, dcnt, translate, target_multibyte)
: memcmp (d, d2, dcnt))
{
@@ -5622,15 +4465,13 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
break;
- /* begline matches the empty string at the beginning of the string
- (unless `not_bol' is set in `bufp'), and after newlines. */
+ /* begline matches the empty string at the beginning of the string,
+ and after newlines. */
case begline:
DEBUG_PRINT ("EXECUTING begline.\n");
if (AT_STRINGS_BEG (d))
- {
- if (!bufp->not_bol) break;
- }
+ break;
else
{
unsigned c;
@@ -5638,7 +4479,6 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
if (c == '\n')
break;
}
- /* In all other cases, we fail. */
goto fail;
@@ -5647,15 +4487,10 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
DEBUG_PRINT ("EXECUTING endline.\n");
if (AT_STRINGS_END (d))
- {
- if (!bufp->not_eol) break;
- }
- else
- {
- PREFETCH_NOLIMIT ();
- if (*d == '\n')
- break;
- }
+ break;
+ PREFETCH_NOLIMIT ();
+ if (*d == '\n')
+ break;
goto fail;
@@ -5675,21 +4510,21 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
goto fail;
- /* on_failure_keep_string_jump is used to optimize `.*\n'. It
+ /* on_failure_keep_string_jump is used to optimize '.*\n'. It
pushes NULL as the value for the string on the stack. Then
- `POP_FAILURE_POINT' will keep the current value for the
+ 'POP_FAILURE_POINT' will keep the current value for the
string, instead of restoring it. To see why, consider
- matching `foo\nbar' against `.*\n'. The .* matches the foo;
+ matching 'foo\nbar' against '.*\n'. The .* matches the foo;
then the . fails against the \n. But the next thing we want
to do is match the \n against the \n; if we restored the
string value, we would be back at the foo.
Because this is used only in specific cases, we don't need to
- check all the things that `on_failure_jump' does, to make
+ check all the things that 'on_failure_jump' does, to make
sure the right things get saved on the stack. Hence we don't
share its code. The only reason to push anything on the
stack at all is that otherwise we would have to change
- `anychar's code to do something besides goto fail in this
+ 'anychar's code to do something besides goto fail in this
case; that seems worse than this. */
case on_failure_keep_string_jump:
EXTRACT_NUMBER_AND_INCR (mcnt, p);
@@ -5718,7 +4553,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
DEBUG_PRINT ("EXECUTING on_failure_jump_nastyloop %d (to %p):\n",
mcnt, p + mcnt);
- assert ((re_opcode_t)p[-4] == no_op);
+ eassert ((re_opcode_t)p[-4] == no_op);
{
int cycle = 0;
CHECK_INFINITE_LOOP (p - 4, d);
@@ -5743,7 +4578,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
CHECK_INFINITE_LOOP (p - 3, d);
if (cycle)
/* If there's a cycle, get out of the loop, as if the matching
- had failed. We used to just `goto fail' here, but that was
+ had failed. We used to just 'goto fail' here, but that was
aborting the search a bit too early: we want to keep the
empty-loop-match and keep matching after the loop.
We want (x?)*y\1z to match both xxyz and xxyxz. */
@@ -5778,7 +4613,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
Compare the beginning of the repeat with what in the
pattern follows its end. If we can establish that there
is nothing that they would both match, i.e., that we
- would have to backtrack because of (as in, e.g., `a*a')
+ would have to backtrack because of (as in, e.g., 'a*a')
then we can use a non-backtracking loop based on
on_failure_keep_string_jump instead of on_failure_jump. */
case on_failure_jump_smart:
@@ -5787,7 +4622,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
mcnt, p + mcnt);
{
re_char *p1 = p; /* Next operation. */
- /* Here, we discard `const', making re_match non-reentrant. */
+ /* Discard 'const', making re_search non-reentrant. */
unsigned char *p2 = (unsigned char *) p + mcnt; /* Jump dest. */
unsigned char *p3 = (unsigned char *) p - 3; /* opcode location. */
@@ -5798,23 +4633,23 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
/* Ensure this is indeed the trivial kind of loop
we are expecting. */
- assert (skip_one_char (p1) == p2 - 3);
- assert ((re_opcode_t) p2[-3] == jump && p2 + mcnt == p);
- DEBUG_STATEMENT (debug += 2);
+ eassert (skip_one_char (p1) == p2 - 3);
+ eassert ((re_opcode_t) p2[-3] == jump && p2 + mcnt == p);
+ DEBUG_STATEMENT (regex_emacs_debug += 2);
if (mutually_exclusive_p (bufp, p1, p2))
{
- /* Use a fast `on_failure_keep_string_jump' loop. */
+ /* Use a fast 'on_failure_keep_string_jump' loop. */
DEBUG_PRINT (" smart exclusive => fast loop.\n");
*p3 = (unsigned char) on_failure_keep_string_jump;
STORE_NUMBER (p2 - 2, mcnt + 3);
}
else
{
- /* Default to a safe `on_failure_jump' loop. */
+ /* Default to a safe 'on_failure_jump' loop. */
DEBUG_PRINT (" smart default => slow loop.\n");
*p3 = (unsigned char) on_failure_jump;
}
- DEBUG_STATEMENT (debug -= 2);
+ DEBUG_STATEMENT (regex_emacs_debug -= 2);
}
break;
@@ -5830,7 +4665,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
/* Have to succeed matching what follows at least n times.
- After that, handle like `on_failure_jump'. */
+ After that, handle like 'on_failure_jump'. */
case succeed_n:
/* Signedness doesn't matter since we only compare MCNT to 0. */
EXTRACT_NUMBER (mcnt, p + 2);
@@ -5839,7 +4674,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
/* Originally, mcnt is how many times we HAVE to succeed. */
if (mcnt != 0)
{
- /* Here, we discard `const', making re_match non-reentrant. */
+ /* Discard 'const', making re_search non-reentrant. */
unsigned char *p2 = (unsigned char *) p + 2; /* counter loc. */
mcnt--;
p += 4;
@@ -5858,7 +4693,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
/* Originally, this is how many times we CAN jump. */
if (mcnt != 0)
{
- /* Here, we discard `const', making re_match non-reentrant. */
+ /* Discard 'const', making re_search non-reentrant. */
unsigned char *p2 = (unsigned char *) p + 2; /* counter loc. */
mcnt--;
PUSH_NUMBER (p2, mcnt);
@@ -5875,7 +4710,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
DEBUG_PRINT ("EXECUTING set_number_at.\n");
EXTRACT_NUMBER_AND_INCR (mcnt, p);
- /* Here, we discard `const', making re_match non-reentrant. */
+ /* Discard 'const', making re_search non-reentrant. */
p2 = (unsigned char *) p + mcnt;
/* Signedness doesn't matter since we only copy MCNT's bits. */
EXTRACT_NUMBER_AND_INCR (mcnt, p);
@@ -5887,7 +4722,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
case wordbound:
case notwordbound:
{
- boolean not = (re_opcode_t) *(p - 1) == notwordbound;
+ bool not = (re_opcode_t) *(p - 1) == notwordbound;
DEBUG_PRINT ("EXECUTING %swordbound.\n", not ? "not" : "");
/* We SUCCEED (or FAIL) in one of the following cases: */
@@ -5899,19 +4734,15 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
{
/* C1 is the character before D, S1 is the syntax of C1, C2
is the character at D, and S2 is the syntax of C2. */
- re_wchar_t c1, c2;
+ int c1, c2;
int s1, s2;
int dummy;
-#ifdef emacs
- ssize_t offset = PTR_TO_OFFSET (d - 1);
- ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ptrdiff_t offset = PTR_TO_OFFSET (d - 1);
+ ptrdiff_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (charpos);
-#endif
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
s1 = SYNTAX (c1);
-#ifdef emacs
UPDATE_SYNTAX_TABLE_FORWARD (charpos + 1);
-#endif
PREFETCH_NOLIMIT ();
GET_CHAR_AFTER (c2, d, dummy);
s2 = SYNTAX (c2);
@@ -5941,14 +4772,12 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
{
/* C1 is the character before D, S1 is the syntax of C1, C2
is the character at D, and S2 is the syntax of C2. */
- re_wchar_t c1, c2;
+ int c1, c2;
int s1, s2;
int dummy;
-#ifdef emacs
- ssize_t offset = PTR_TO_OFFSET (d);
- ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ptrdiff_t offset = PTR_TO_OFFSET (d);
+ ptrdiff_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (charpos);
-#endif
PREFETCH ();
GET_CHAR_AFTER (c2, d, dummy);
s2 = SYNTAX (c2);
@@ -5961,9 +4790,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
if (!AT_STRINGS_BEG (d))
{
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
-#ifdef emacs
UPDATE_SYNTAX_TABLE_BACKWARD (charpos - 1);
-#endif
s1 = SYNTAX (c1);
/* ... and S1 is Sword, and WORD_BOUNDARY_P (C1, C2)
@@ -5986,14 +4813,12 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
{
/* C1 is the character before D, S1 is the syntax of C1, C2
is the character at D, and S2 is the syntax of C2. */
- re_wchar_t c1, c2;
+ int c1, c2;
int s1, s2;
int dummy;
-#ifdef emacs
- ssize_t offset = PTR_TO_OFFSET (d) - 1;
- ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ptrdiff_t offset = PTR_TO_OFFSET (d) - 1;
+ ptrdiff_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (charpos);
-#endif
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
s1 = SYNTAX (c1);
@@ -6006,9 +4831,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
{
PREFETCH_NOLIMIT ();
GET_CHAR_AFTER (c2, d, dummy);
-#ifdef emacs
UPDATE_SYNTAX_TABLE_FORWARD (charpos);
-#endif
s2 = SYNTAX (c2);
/* ... and S2 is Sword, and WORD_BOUNDARY_P (C1, C2)
@@ -6031,13 +4854,11 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
{
/* C1 is the character before D, S1 is the syntax of C1, C2
is the character at D, and S2 is the syntax of C2. */
- re_wchar_t c1, c2;
+ int c1, c2;
int s1, s2;
-#ifdef emacs
- ssize_t offset = PTR_TO_OFFSET (d);
- ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ptrdiff_t offset = PTR_TO_OFFSET (d);
+ ptrdiff_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (charpos);
-#endif
PREFETCH ();
c2 = RE_STRING_CHAR (d, target_multibyte);
s2 = SYNTAX (c2);
@@ -6050,9 +4871,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
if (!AT_STRINGS_BEG (d))
{
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
-#ifdef emacs
UPDATE_SYNTAX_TABLE_BACKWARD (charpos - 1);
-#endif
s1 = SYNTAX (c1);
/* ... and S1 is Sword or Ssymbol. */
@@ -6074,13 +4893,11 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
{
/* C1 is the character before D, S1 is the syntax of C1, C2
is the character at D, and S2 is the syntax of C2. */
- re_wchar_t c1, c2;
+ int c1, c2;
int s1, s2;
-#ifdef emacs
- ssize_t offset = PTR_TO_OFFSET (d) - 1;
- ssize_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ptrdiff_t offset = PTR_TO_OFFSET (d) - 1;
+ ptrdiff_t charpos = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (charpos);
-#endif
GET_CHAR_BEFORE_2 (c1, d, string1, end1, string2, end2);
s1 = SYNTAX (c1);
@@ -6093,9 +4910,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
{
PREFETCH_NOLIMIT ();
c2 = RE_STRING_CHAR (d, target_multibyte);
-#ifdef emacs
UPDATE_SYNTAX_TABLE_FORWARD (charpos + 1);
-#endif
s2 = SYNTAX (c2);
/* ... and S2 is Sword or Ssymbol. */
@@ -6108,21 +4923,19 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
case syntaxspec:
case notsyntaxspec:
{
- boolean not = (re_opcode_t) *(p - 1) == notsyntaxspec;
+ bool not = (re_opcode_t) *(p - 1) == notsyntaxspec;
mcnt = *p++;
DEBUG_PRINT ("EXECUTING %ssyntaxspec %d.\n", not ? "not" : "",
mcnt);
PREFETCH ();
-#ifdef emacs
{
- ssize_t offset = PTR_TO_OFFSET (d);
- ssize_t pos1 = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
+ ptrdiff_t offset = PTR_TO_OFFSET (d);
+ ptrdiff_t pos1 = SYNTAX_TABLE_BYTE_TO_CHAR (offset);
UPDATE_SYNTAX_TABLE (pos1);
}
-#endif
{
int len;
- re_wchar_t c;
+ int c;
GET_CHAR_AFTER (c, d, len);
if ((SYNTAX (c) != (enum syntaxcode) mcnt) ^ not)
@@ -6132,7 +4945,6 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
}
break;
-#ifdef emacs
case at_dot:
DEBUG_PRINT ("EXECUTING at_dot.\n");
if (PTR_BYTE_POS (d) != PT_BYTE)
@@ -6142,7 +4954,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
case categoryspec:
case notcategoryspec:
{
- boolean not = (re_opcode_t) *(p - 1) == notcategoryspec;
+ bool not = (re_opcode_t) *(p - 1) == notcategoryspec;
mcnt = *p++;
DEBUG_PRINT ("EXECUTING %scategoryspec %d.\n",
not ? "not" : "", mcnt);
@@ -6150,7 +4962,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
{
int len;
- re_wchar_t c;
+ int c;
GET_CHAR_AFTER (c, d, len);
if ((!CHAR_HAS_CATEGORY (c, mcnt)) ^ not)
goto fail;
@@ -6159,8 +4971,6 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
}
break;
-#endif /* emacs */
-
default:
abort ();
}
@@ -6179,11 +4989,11 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
switch (*pat++)
{
case on_failure_keep_string_jump:
- assert (str == NULL);
+ eassert (str == NULL);
goto continue_failure_jump;
case on_failure_jump_nastyloop:
- assert ((re_opcode_t)pat[-2] == no_op);
+ eassert ((re_opcode_t)pat[-2] == no_op);
PUSH_FAILURE_POINT (pat - 2, str);
FALLTHROUGH;
case on_failure_jump_loop:
@@ -6203,7 +5013,7 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
abort ();
}
- assert (p >= bufp->buffer && p <= pend);
+ eassert (p >= bufp->buffer && p <= pend);
if (d >= string1 && d <= end1)
dend = end_match_1;
@@ -6215,9 +5025,9 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
if (best_regs_set)
goto restore_best_regs;
- FREE_VARIABLES ();
+ SAFE_FREE ();
- return -1; /* Failure to match. */
+ return -1; /* Failure to match. */
}
/* Subroutine definitions for re_match_2. */
@@ -6226,19 +5036,19 @@ re_match_2_internal (struct re_pattern_buffer *bufp,
re_char *string1,
bytes; nonzero otherwise. */
static int
-bcmp_translate (re_char *s1, re_char *s2, ssize_t len,
- RE_TRANSLATE_TYPE translate, const int target_multibyte)
+bcmp_translate (re_char *s1, re_char *s2, ptrdiff_t len,
+ Lisp_Object translate, int target_multibyte)
{
re_char *p1 = s1, *p2 = s2;
re_char *p1_end = s1 + len;
re_char *p2_end = s2 + len;
/* FIXME: Checking both p1 and p2 presumes that the two strings might have
- different lengths, but relying on a single `len' would break this. -sm */
+ different lengths, but relying on a single LEN would break this. -sm */
while (p1 < p1_end && p2 < p2_end)
{
int p1_charlen, p2_charlen;
- re_wchar_t p1_ch, p2_ch;
+ int p1_ch, p2_ch;
GET_CHAR_AFTER (p1_ch, p1, p1_charlen);
GET_CHAR_AFTER (p2_ch, p2, p2_charlen);
@@ -6262,16 +5072,14 @@ bcmp_translate (re_char *s1, re_char *s2, ssize_t len,
compiles PATTERN (of length SIZE) and puts the result in BUFP.
Returns 0 if the pattern was valid, otherwise an error string.
- Assumes the `allocated' (and perhaps `buffer') and `translate' fields
+ Assumes the 'allocated' (and perhaps 'buffer') and 'translate' fields
are set in BUFP on entry.
We call regex_compile to do the actual compilation. */
const char *
re_compile_pattern (const char *pattern, size_t length,
-#ifdef emacs
bool posix_backtracking, const char *whitespace_regexp,
-#endif
struct re_pattern_buffer *bufp)
{
reg_errcode_t ret;
@@ -6280,335 +5088,12 @@ re_compile_pattern (const char *pattern, size_t length,
(and at least one extra will be -1). */
bufp->regs_allocated = REGS_UNALLOCATED;
- /* And GNU code determines whether or not to get register information
- by passing null for the REGS argument to re_match, etc., not by
- setting no_sub. */
- bufp->no_sub = 0;
-
ret = regex_compile ((re_char *) pattern, length,
-#ifdef emacs
posix_backtracking,
whitespace_regexp,
-#else
- re_syntax_options,
-#endif
bufp);
if (!ret)
return NULL;
- return gettext (re_error_msgid[(int) ret]);
-}
-WEAK_ALIAS (__re_compile_pattern, re_compile_pattern)
-
-/* Entry points compatible with 4.2 BSD regex library. We don't define
- them unless specifically requested. */
-
-#if defined _REGEX_RE_COMP || defined _LIBC
-
-/* BSD has one and only one pattern buffer. */
-static struct re_pattern_buffer re_comp_buf;
-
-char *
-# ifdef _LIBC
-/* Make these definitions weak in libc, so POSIX programs can redefine
- these names if they don't use our functions, and still use
- regcomp/regexec below without link errors. */
-weak_function
-# endif
-re_comp (const char *s)
-{
- reg_errcode_t ret;
-
- if (!s)
- {
- if (!re_comp_buf.buffer)
- /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
- return (char *) gettext ("No previous regular expression");
- return 0;
- }
-
- if (!re_comp_buf.buffer)
- {
- re_comp_buf.buffer = malloc (200);
- if (re_comp_buf.buffer == NULL)
- /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
- return (char *) gettext (re_error_msgid[(int) REG_ESPACE]);
- re_comp_buf.allocated = 200;
-
- re_comp_buf.fastmap = malloc (1 << BYTEWIDTH);
- if (re_comp_buf.fastmap == NULL)
- /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
- return (char *) gettext (re_error_msgid[(int) REG_ESPACE]);
- }
-
- /* Since `re_exec' always passes NULL for the `regs' argument, we
- don't need to initialize the pattern buffer fields which affect it. */
-
- ret = regex_compile (s, strlen (s), re_syntax_options, &re_comp_buf);
-
- if (!ret)
- return NULL;
-
- /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */
- return (char *) gettext (re_error_msgid[(int) ret]);
-}
-
-
-int
-# ifdef _LIBC
-weak_function
-# endif
-re_exec (const char *s)
-{
- const size_t len = strlen (s);
- return re_search (&re_comp_buf, s, len, 0, len, 0) >= 0;
-}
-#endif /* _REGEX_RE_COMP */
-
-/* POSIX.2 functions. Don't define these for Emacs. */
-
-#ifndef emacs
-
-/* regcomp takes a regular expression as a string and compiles it.
-
- PREG is a regex_t *. We do not expect any fields to be initialized,
- since POSIX says we shouldn't. Thus, we set
-
- `buffer' to the compiled pattern;
- `used' to the length of the compiled pattern;
- `syntax' to RE_SYNTAX_POSIX_EXTENDED if the
- REG_EXTENDED bit in CFLAGS is set; otherwise, to
- RE_SYNTAX_POSIX_BASIC;
- `fastmap' to an allocated space for the fastmap;
- `fastmap_accurate' to zero;
- `re_nsub' to the number of subexpressions in PATTERN.
-
- PATTERN is the address of the pattern string.
-
- CFLAGS is a series of bits which affect compilation.
-
- If REG_EXTENDED is set, we use POSIX extended syntax; otherwise, we
- use POSIX basic syntax.
-
- If REG_NEWLINE is set, then . and [^...] don't match newline.
- Also, regexec will try a match beginning after every newline.
-
- If REG_ICASE is set, then we considers upper- and lowercase
- versions of letters to be equivalent when matching.
-
- If REG_NOSUB is set, then when PREG is passed to regexec, that
- routine will report only success or failure, and nothing about the
- registers.
-
- It returns 0 if it succeeds, nonzero if it doesn't. (See regex.h for
- the return codes and their meanings.) */
-
-reg_errcode_t
-regcomp (regex_t *_Restrict_ preg, const char *_Restrict_ pattern,
- int cflags)
-{
- reg_errcode_t ret;
- reg_syntax_t syntax
- = (cflags & REG_EXTENDED) ?
- RE_SYNTAX_POSIX_EXTENDED : RE_SYNTAX_POSIX_BASIC;
-
- /* regex_compile will allocate the space for the compiled pattern. */
- preg->buffer = 0;
- preg->allocated = 0;
- preg->used = 0;
-
- /* Try to allocate space for the fastmap. */
- preg->fastmap = malloc (1 << BYTEWIDTH);
-
- if (cflags & REG_ICASE)
- {
- unsigned i;
-
- preg->translate = malloc (CHAR_SET_SIZE * sizeof *preg->translate);
- if (preg->translate == NULL)
- return (int) REG_ESPACE;
-
- /* Map uppercase characters to corresponding lowercase ones. */
- for (i = 0; i < CHAR_SET_SIZE; i++)
- preg->translate[i] = ISUPPER (i) ? TOLOWER (i) : i;
- }
- else
- preg->translate = NULL;
-
- /* If REG_NEWLINE is set, newlines are treated differently. */
- if (cflags & REG_NEWLINE)
- { /* REG_NEWLINE implies neither . nor [^...] match newline. */
- syntax &= ~RE_DOT_NEWLINE;
- syntax |= RE_HAT_LISTS_NOT_NEWLINE;
- }
- else
- syntax |= RE_NO_NEWLINE_ANCHOR;
-
- preg->no_sub = !!(cflags & REG_NOSUB);
-
- /* POSIX says a null character in the pattern terminates it, so we
- can use strlen here in compiling the pattern. */
- ret = regex_compile ((re_char *) pattern, strlen (pattern), syntax, preg);
-
- /* POSIX doesn't distinguish between an unmatched open-group and an
- unmatched close-group: both are REG_EPAREN. */
- if (ret == REG_ERPAREN)
- ret = REG_EPAREN;
-
- if (ret == REG_NOERROR && preg->fastmap)
- { /* Compute the fastmap now, since regexec cannot modify the pattern
- buffer. */
- re_compile_fastmap (preg);
- if (preg->can_be_null)
- { /* The fastmap can't be used anyway. */
- free (preg->fastmap);
- preg->fastmap = NULL;
- }
- }
- return ret;
-}
-WEAK_ALIAS (__regcomp, regcomp)
-
-
-/* regexec searches for a given pattern, specified by PREG, in the
- string STRING.
-
- If NMATCH is zero or REG_NOSUB was set in the cflags argument to
- `regcomp', we ignore PMATCH. Otherwise, we assume PMATCH has at
- least NMATCH elements, and we set them to the offsets of the
- corresponding matched substrings.
-
- EFLAGS specifies `execution flags' which affect matching: if
- REG_NOTBOL is set, then ^ does not match at the beginning of the
- string; if REG_NOTEOL is set, then $ does not match at the end.
-
- We return 0 if we find a match and REG_NOMATCH if not. */
-
-reg_errcode_t
-regexec (const regex_t *_Restrict_ preg, const char *_Restrict_ string,
- size_t nmatch, regmatch_t pmatch[_Restrict_arr_], int eflags)
-{
- regoff_t ret;
- struct re_registers regs;
- regex_t private_preg;
- size_t len = strlen (string);
- boolean want_reg_info = !preg->no_sub && nmatch > 0 && pmatch;
-
- private_preg = *preg;
-
- private_preg.not_bol = !!(eflags & REG_NOTBOL);
- private_preg.not_eol = !!(eflags & REG_NOTEOL);
-
- /* The user has told us exactly how many registers to return
- information about, via `nmatch'. We have to pass that on to the
- matching routines. */
- private_preg.regs_allocated = REGS_FIXED;
-
- if (want_reg_info)
- {
- regs.num_regs = nmatch;
- regs.start = TALLOC (nmatch * 2, regoff_t);
- if (regs.start == NULL)
- return REG_NOMATCH;
- regs.end = regs.start + nmatch;
- }
-
- /* Instead of using not_eol to implement REG_NOTEOL, we could simply
- pass (&private_preg, string, len + 1, 0, len, ...) pretending the string
- was a little bit longer but still only matching the real part.
- This works because the `endline' will check for a '\n' and will find a
- '\0', correctly deciding that this is not the end of a line.
- But it doesn't work out so nicely for REG_NOTBOL, since we don't have
- a convenient '\0' there. For all we know, the string could be preceded
- by '\n' which would throw things off. */
-
- /* Perform the searching operation. */
- ret = re_search (&private_preg, string, len,
- /* start: */ 0, /* range: */ len,
- want_reg_info ? ®s : 0);
-
- /* Copy the register information to the POSIX structure. */
- if (want_reg_info)
- {
- if (ret >= 0)
- {
- unsigned r;
-
- for (r = 0; r < nmatch; r++)
- {
- pmatch[r].rm_so = regs.start[r];
- pmatch[r].rm_eo = regs.end[r];
- }
- }
-
- /* If we needed the temporary register info, free the space now. */
- free (regs.start);
- }
-
- /* We want zero return to mean success, unlike `re_search'. */
- return ret >= 0 ? REG_NOERROR : REG_NOMATCH;
-}
-WEAK_ALIAS (__regexec, regexec)
-
-
-/* Returns a message corresponding to an error code, ERR_CODE, returned
- from either regcomp or regexec. We don't use PREG here.
-
- ERR_CODE was previously called ERRCODE, but that name causes an
- error with msvc8 compiler. */
-
-size_t
-regerror (int err_code, const regex_t *preg, char *errbuf, size_t errbuf_size)
-{
- const char *msg;
- size_t msg_size;
-
- if (err_code < 0
- || err_code >= (sizeof (re_error_msgid) / sizeof (re_error_msgid[0])))
- /* Only error codes returned by the rest of the code should be passed
- to this routine. If we are given anything else, or if other regex
- code generates an invalid error code, then the program has a bug.
- Dump core so we can fix it. */
- abort ();
-
- msg = gettext (re_error_msgid[err_code]);
-
- msg_size = strlen (msg) + 1; /* Includes the null. */
-
- if (errbuf_size != 0)
- {
- if (msg_size > errbuf_size)
- {
- memcpy (errbuf, msg, errbuf_size - 1);
- errbuf[errbuf_size - 1] = 0;
- }
- else
- strcpy (errbuf, msg);
- }
-
- return msg_size;
-}
-WEAK_ALIAS (__regerror, regerror)
-
-
-/* Free dynamically allocated space used by PREG. */
-
-void
-regfree (regex_t *preg)
-{
- free (preg->buffer);
- preg->buffer = NULL;
-
- preg->allocated = 0;
- preg->used = 0;
-
- free (preg->fastmap);
- preg->fastmap = NULL;
- preg->fastmap_accurate = 0;
-
- free (preg->translate);
- preg->translate = NULL;
+ return re_error_msgid[ret];
}
-WEAK_ALIAS (__regfree, regfree)
-
-#endif /* not emacs */
diff --git a/src/regex-emacs.h b/src/regex-emacs.h
new file mode 100644
index 0000000..a849cbe
--- /dev/null
+++ b/src/regex-emacs.h
@@ -0,0 +1,197 @@
+/* Emacs regular expression API
+
+ Copyright (C) 1985, 1989-1993, 1995, 2000-2018 Free Software
+ Foundation, Inc.
+
+ This program is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 3, or (at your option)
+ any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <https://www.gnu.org/licenses/>. */
+
+#ifndef EMACS_REGEX_H
+#define EMACS_REGEX_H 1
+
+#include <stddef.h>
+
+/* This is the structure we store register match data in.
+ Declare this before including lisp.h, since lisp.h (via thread.h)
+ uses struct re_registers. */
+struct re_registers
+{
+ unsigned num_regs;
+ ptrdiff_t *start;
+ ptrdiff_t *end;
+};
+
+#include "lisp.h"
+
+/* The string or buffer being matched.
+ It is used for looking up syntax properties.
+
+ If the value is a Lisp string object, match text in that string; if
+ it's nil, match text in the current buffer; if it's t, match text
+ in a C string.
+
+ This value is effectively another parameter to re_search_2 and
+ re_match_2. No calls into Lisp or thread switches are allowed
+ before setting re_match_object and calling into the regex search
+ and match functions. These functions capture the current value of
+ re_match_object into gl_state on entry.
+
+ TODO: turn into an actual function parameter. */
+extern Lisp_Object re_match_object;
+
+/* Roughly the maximum number of failure points on the stack. */
+extern size_t emacs_re_max_failures;
+
+/* Amount of memory that we can safely stack allocate. */
+extern ptrdiff_t emacs_re_safe_alloca;
+
+/* This data structure represents a compiled pattern. Before calling
+ the pattern compiler, the fields 'buffer', 'allocated', 'fastmap',
+ and 'translate' can be set. After the pattern has been
+ compiled, the 're_nsub' field is available. All other fields are
+ private to the regex routines. */
+
+struct re_pattern_buffer
+{
+ /* Space that holds the compiled pattern. It is declared as
+ 'unsigned char *' because its elements are
+ sometimes used as array indexes. */
+ unsigned char *buffer;
+
+ /* Number of bytes to which 'buffer' points. */
+ size_t allocated;
+
+ /* Number of bytes actually used in 'buffer'. */
+ size_t used;
+
+ /* Charset of unibyte characters at compiling time. */
+ int charset_unibyte;
+
+ /* Pointer to a fastmap, if any, otherwise zero. re_search uses
+ the fastmap, if there is one, to skip over impossible
+ starting points for matches. */
+ char *fastmap;
+
+ /* Either a translate table to apply to all characters before
+ comparing them, or zero for no translation. The translation
+ applies to a pattern when it is compiled and to a string
+ when it is matched. */
+ Lisp_Object translate;
+
+ /* Number of subexpressions found by the compiler. */
+ size_t re_nsub;
+
+ /* True if and only if this pattern can match the empty string.
+ Well, in truth it's used only in 're_search_2', to see
+ whether or not we should use the fastmap, so we don't set
+ this absolutely perfectly; see 're_compile_fastmap'. */
+ unsigned can_be_null : 1;
+
+ /* If REGS_UNALLOCATED, allocate space in the 'regs' structure
+ for 'max (RE_NREGS, re_nsub + 1)' groups.
+ If REGS_REALLOCATE, reallocate space if necessary.
+ If REGS_FIXED, use what's there. */
+ unsigned regs_allocated : 2;
+
+ /* Set to false when 'regex_compile' compiles a pattern; set to true
+ by 're_compile_fastmap' if it updates the fastmap. */
+ unsigned fastmap_accurate : 1;
+
+ /* If true, the compilation of the pattern had to look up the syntax table,
+ so the compiled pattern is valid for the current syntax table only. */
+ unsigned used_syntax : 1;
+
+ /* If true, multi-byte form in the regexp pattern should be
+ recognized as a multibyte character. */
+ unsigned multibyte : 1;
+
+ /* If true, multi-byte form in the target of match should be
+ recognized as a multibyte character. */
+ unsigned target_multibyte : 1;
+};
+
+/* Declarations for routines. */
+
+/* Compile the regular expression PATTERN, with length LENGTH
+ and syntax given by the global 're_syntax_options', into the buffer
+ BUFFER. Return NULL if successful, and an error string if not. */
+extern const char *re_compile_pattern (const char *pattern, size_t length,
+ bool posix_backtracking,
+ const char *whitespace_regexp,
+ struct re_pattern_buffer *buffer);
+
+
+/* Search in the string STRING (with length LENGTH) for the pattern
+ compiled into BUFFER. Start searching at position START, for RANGE
+ characters. Return the starting position of the match, -1 for no
+ match, or -2 for an internal error. Also return register
+ information in REGS (if REGS is non-null). */
+extern ptrdiff_t re_search (struct re_pattern_buffer *buffer,
+ const char *string, size_t length,
+ ptrdiff_t start, ptrdiff_t range,
+ struct re_registers *regs);
+
+
+/* Like 're_search', but search in the concatenation of STRING1 and
+ STRING2. Also, stop searching at index START + STOP. */
+extern ptrdiff_t re_search_2 (struct re_pattern_buffer *buffer,
+ const char *string1, size_t length1,
+ const char *string2, size_t length2,
+ ptrdiff_t start, ptrdiff_t range,
+ struct re_registers *regs,
+ ptrdiff_t stop);
+
+
+/* Like 're_search_2', but return how many characters in STRING the regexp
+ in BUFFER matched, starting at position START. */
+extern ptrdiff_t re_match_2 (struct re_pattern_buffer *buffer,
+ const char *string1, size_t length1,
+ const char *string2, size_t length2,
+ ptrdiff_t start, struct re_registers *regs,
+ ptrdiff_t stop);
+
+
+/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
+ ENDS. Subsequent matches using BUFFER and REGS will use this memory
+ for recording register information. STARTS and ENDS must be
+ allocated with malloc, and must each be at least 'NUM_REGS * sizeof
+ (ptrdiff_t)' bytes long.
+
+ If NUM_REGS == 0, then subsequent matches should allocate their own
+ register data.
+
+ Unless this function is called, the first search or match using
+ PATTERN_BUFFER will allocate its own register data, without
+ freeing the old data. */
+extern void re_set_registers (struct re_pattern_buffer *buffer,
+ struct re_registers *regs,
+ unsigned num_regs,
+ ptrdiff_t *starts, ptrdiff_t *ends);
+
+/* Character classes. */
+typedef enum { RECC_ERROR = 0,
+ RECC_ALNUM, RECC_ALPHA, RECC_WORD,
+ RECC_GRAPH, RECC_PRINT,
+ RECC_LOWER, RECC_UPPER,
+ RECC_PUNCT, RECC_CNTRL,
+ RECC_DIGIT, RECC_XDIGIT,
+ RECC_BLANK, RECC_SPACE,
+ RECC_MULTIBYTE, RECC_NONASCII,
+ RECC_ASCII, RECC_UNIBYTE
+} re_wctype_t;
+
+extern bool re_iswctype (int ch, re_wctype_t cc);
+extern re_wctype_t re_wctype_parse (const unsigned char **strp,
+ unsigned limit);
+
+#endif /* EMACS_REGEX_H */
diff --git a/src/regex.h b/src/regex.h
deleted file mode 100644
index 3a2d74d..0000000
--- a/src/regex.h
+++ /dev/null
@@ -1,654 +0,0 @@
-/* Definitions for data structures and routines for the regular
- expression library, version 0.12.
-
- Copyright (C) 1985, 1989-1993, 1995, 2000-2018 Free Software
- Foundation, Inc.
-
- This program is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; either version 3, or (at your option)
- any later version.
-
- This program is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- GNU General Public License for more details.
-
- You should have received a copy of the GNU General Public License
- along with this program. If not, see <https://www.gnu.org/licenses/>. */
-
-#ifndef _REGEX_H
-#define _REGEX_H 1
-
-#if defined emacs && (defined _REGEX_RE_COMP || defined _LIBC)
-/* We're not defining re_set_syntax and using a different prototype of
- re_compile_pattern when building Emacs so fail compilation early with
- a (somewhat helpful) error message when conflict is detected. */
-# error "_REGEX_RE_COMP nor _LIBC can be defined if emacs is defined."
-#endif
-
-#include <sys/types.h>
-
-/* Allow the use in C++ code. */
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-#if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS
-/* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
- should be there. */
-# include <stddef.h>
-#endif
-
-/* The following bits are used to determine the regexp syntax we
- recognize. The set/not-set meanings where historically chosen so
- that Emacs syntax had the value 0.
- The bits are given in alphabetical order, and
- the definitions shifted by one from the previous bit; thus, when we
- add or remove a bit, only one other definition need change. */
-typedef unsigned long reg_syntax_t;
-
-/* If this bit is not set, then \ inside a bracket expression is literal.
- If set, then such a \ quotes the following character. */
-#define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
-
-/* If this bit is not set, then + and ? are operators, and \+ and \? are
- literals.
- If set, then \+ and \? are operators and + and ? are literals. */
-#define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
-
-/* If this bit is set, then character classes are supported. They are:
- [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
- [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
- If not set, then character classes are not supported. */
-#define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
-
-/* If this bit is set, then ^ and $ are always anchors (outside bracket
- expressions, of course).
- If this bit is not set, then it depends:
- ^ is an anchor if it is at the beginning of a regular
- expression or after an open-group or an alternation operator;
- $ is an anchor if it is at the end of a regular expression, or
- before a close-group or an alternation operator.
-
- This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
- POSIX draft 11.2 says that * etc. in leading positions is undefined.
- We already implemented a previous draft which made those constructs
- invalid, though, so we haven't changed the code back. */
-#define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
-
-/* If this bit is set, then special characters are always special
- regardless of where they are in the pattern.
- If this bit is not set, then special characters are special only in
- some contexts; otherwise they are ordinary. Specifically,
- * + ? and intervals are only special when not after the beginning,
- open-group, or alternation operator. */
-#define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
-
-/* If this bit is set, then *, +, ?, and { cannot be first in an re or
- immediately after an alternation or begin-group operator. */
-#define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
-
-/* If this bit is set, then . matches newline.
- If not set, then it doesn't. */
-#define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
-
-/* If this bit is set, then . doesn't match NUL.
- If not set, then it does. */
-#define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
-
-/* If this bit is set, nonmatching lists [^...] do not match newline.
- If not set, they do. */
-#define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
-
-/* If this bit is set, either \{...\} or {...} defines an
- interval, depending on RE_NO_BK_BRACES.
- If not set, \{, \}, {, and } are literals. */
-#define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
-
-/* If this bit is set, +, ? and | aren't recognized as operators.
- If not set, they are. */
-#define RE_LIMITED_OPS (RE_INTERVALS << 1)
-
-/* If this bit is set, newline is an alternation operator.
- If not set, newline is literal. */
-#define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
-
-/* If this bit is set, then `{...}' defines an interval, and \{ and \}
- are literals.
- If not set, then `\{...\}' defines an interval. */
-#define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
-
-/* If this bit is set, (...) defines a group, and \( and \) are literals.
- If not set, \(...\) defines a group, and ( and ) are literals. */
-#define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
-
-/* If this bit is set, then \<digit> matches <digit>.
- If not set, then \<digit> is a back-reference. */
-#define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
-
-/* If this bit is set, then | is an alternation operator, and \| is literal.
- If not set, then \| is an alternation operator, and | is literal. */
-#define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
-
-/* If this bit is set, then an ending range point collating higher
- than the starting range point, as in [z-a], is invalid.
- If not set, then when ending range point collates higher than the
- starting range point, the range is ignored. */
-#define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
-
-/* If this bit is set, then an unmatched ) is ordinary.
- If not set, then an unmatched ) is invalid. */
-#define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
-
-/* If this bit is set, succeed as soon as we match the whole pattern,
- without further backtracking. */
-#define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
-
-/* If this bit is set, do not process the GNU regex operators.
- If not set, then the GNU regex operators are recognized. */
-#define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
-
-/* If this bit is set, then *?, +? and ?? match non greedily. */
-#define RE_FRUGAL (RE_NO_GNU_OPS << 1)
-
-/* If this bit is set, then (?:...) is treated as a shy group. */
-#define RE_SHY_GROUPS (RE_FRUGAL << 1)
-
-/* If this bit is set, ^ and $ only match at beg/end of buffer. */
-#define RE_NO_NEWLINE_ANCHOR (RE_SHY_GROUPS << 1)
-
-/* If this bit is set, turn on internal regex debugging.
- If not set, and debugging was on, turn it off.
- This only works if regex.c is compiled -DDEBUG.
- We define this bit always, so that all that's needed to turn on
- debugging is to recompile regex.c; the calling code can always have
- this bit set, and it won't affect anything in the normal case. */
-#define RE_DEBUG (RE_NO_NEWLINE_ANCHOR << 1)
-
-/* This global variable defines the particular regexp syntax to use (for
- some interfaces). When a regexp is compiled, the syntax used is
- stored in the pattern buffer, so changing this does not affect
- already-compiled regexps. */
-/* extern reg_syntax_t re_syntax_options; */
-
-#ifdef emacs
-# include "lisp.h"
-/* In Emacs, this is the string or buffer in which we are matching.
- It is used for looking up syntax properties.
-
- If the value is a Lisp string object, we are matching text in that
- string; if it's nil, we are matching text in the current buffer; if
- it's t, we are matching text in a C string.
-
- This value is effectively another parameter to re_search_2 and
- re_match_2. No calls into Lisp or thread switches are allowed
- before setting re_match_object and calling into the regex search
- and match functions. These functions capture the current value of
- re_match_object into gl_state on entry.
-
- TODO: once we get rid of the !emacs case in this code, turn into an
- actual function parameter. */
-extern Lisp_Object re_match_object;
-#endif
-
-/* Roughly the maximum number of failure points on the stack. */
-extern size_t emacs_re_max_failures;
-
-#ifdef emacs
-/* Amount of memory that we can safely stack allocate. */
-extern ptrdiff_t emacs_re_safe_alloca;
-#endif
-
-
-/* Define combinations of the above bits for the standard possibilities.
- (The [[[ comments delimit what gets put into the Texinfo file, so
- don't delete them!) */
-/* [[[begin syntaxes]]] */
-#define RE_SYNTAX_EMACS
\
- (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS | RE_FRUGAL)
-
-#define RE_SYNTAX_AWK \
- (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
- | RE_NO_BK_PARENS | RE_NO_BK_REFS \
- | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
- | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
- | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
-
-#define RE_SYNTAX_GNU_AWK \
- ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG)
\
- & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
-
-#define RE_SYNTAX_POSIX_AWK \
- (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
- | RE_INTERVALS | RE_NO_GNU_OPS)
-
-#define RE_SYNTAX_GREP \
- (RE_BK_PLUS_QM | RE_CHAR_CLASSES
\
- | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
- | RE_NEWLINE_ALT)
-
-#define RE_SYNTAX_EGREP
\
- (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
- | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
- | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
- | RE_NO_BK_VBAR)
-
-#define RE_SYNTAX_POSIX_EGREP \
- (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
-
-/* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
-#define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
-
-#define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
-
-/* Syntax bits common to both basic and extended POSIX regex syntax. */
-#define _RE_SYNTAX_POSIX_COMMON
\
- (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
- | RE_INTERVALS | RE_NO_EMPTY_RANGES)
-
-#define RE_SYNTAX_POSIX_BASIC \
- (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
-
-/* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
- RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
- isn't minimal, since other operators, such as \`, aren't disabled. */
-#define RE_SYNTAX_POSIX_MINIMAL_BASIC \
- (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
-
-#define RE_SYNTAX_POSIX_EXTENDED \
- (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
- | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
- | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
- | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
-
-/* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
- removed and RE_NO_BK_REFS is added. */
-#define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
- (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
- | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
- | RE_NO_BK_PARENS | RE_NO_BK_REFS \
- | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
-/* [[[end syntaxes]]] */
-
-/* Maximum number of duplicates an interval can allow. Some systems
- (erroneously) define this in other header files, but we want our
- value, so remove any previous define. */
-#ifdef RE_DUP_MAX
-# undef RE_DUP_MAX
-#endif
-/* Repeat counts are stored in opcodes as 2 byte integers. This was
- previously limited to 7fff because the parsing code uses signed
- ints. But Emacs only runs on 32 bit platforms anyway. */
-#define RE_DUP_MAX (0xffff)
-
-
-/* POSIX `cflags' bits (i.e., information for `regcomp'). */
-
-/* If this bit is set, then use extended regular expression syntax.
- If not set, then use basic regular expression syntax. */
-#define REG_EXTENDED 1
-
-/* If this bit is set, then ignore case when matching.
- If not set, then case is significant. */
-#define REG_ICASE (REG_EXTENDED << 1)
-
-/* If this bit is set, then anchors do not match at newline
- characters in the string.
- If not set, then anchors do match at newlines. */
-#define REG_NEWLINE (REG_ICASE << 1)
-
-/* If this bit is set, then report only success or fail in regexec.
- If not set, then returns differ between not matching and errors. */
-#define REG_NOSUB (REG_NEWLINE << 1)
-
-
-/* POSIX `eflags' bits (i.e., information for regexec). */
-
-/* If this bit is set, then the beginning-of-line operator doesn't match
- the beginning of the string (presumably because it's not the
- beginning of a line).
- If not set, then the beginning-of-line operator does match the
- beginning of the string. */
-#define REG_NOTBOL 1
-
-/* Like REG_NOTBOL, except for the end-of-line. */
-#define REG_NOTEOL (1 << 1)
-
-
-/* If any error codes are removed, changed, or added, update the
- `re_error_msg' table in regex.c. */
-typedef enum
-{
-#ifdef _XOPEN_SOURCE
- REG_ENOSYS = -1, /* This will never happen for this implementation. */
-#endif
-
- REG_NOERROR = 0, /* Success. */
- REG_NOMATCH, /* Didn't find a match (for regexec). */
-
- /* POSIX regcomp return error codes. (In the order listed in the
- standard.) */
- REG_BADPAT, /* Invalid pattern. */
- REG_ECOLLATE, /* Not implemented. */
- REG_ECTYPE, /* Invalid character class name. */
- REG_EESCAPE, /* Trailing backslash. */
- REG_ESUBREG, /* Invalid back reference. */
- REG_EBRACK, /* Unmatched left bracket. */
- REG_EPAREN, /* Parenthesis imbalance. */
- REG_EBRACE, /* Unmatched \{. */
- REG_BADBR, /* Invalid contents of \{\}. */
- REG_ERANGE, /* Invalid range end. */
- REG_ESPACE, /* Ran out of memory. */
- REG_BADRPT, /* No preceding re for repetition op. */
-
- /* Error codes we've added. */
- REG_EEND, /* Premature end. */
- REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
- REG_ERPAREN, /* Unmatched ) or \); not returned from regcomp. */
- REG_ERANGEX, /* Range striding over charsets. */
- REG_ESIZEBR /* n or m too big in \{n,m\} */
-} reg_errcode_t;
-
-/* This data structure represents a compiled pattern. Before calling
- the pattern compiler, the fields `buffer', `allocated', `fastmap',
- `translate', and `no_sub' can be set. After the pattern has been
- compiled, the `re_nsub' field is available. All other fields are
- private to the regex routines. */
-
-#ifndef RE_TRANSLATE_TYPE
-# define RE_TRANSLATE_TYPE char *
-#endif
-
-struct re_pattern_buffer
-{
-/* [[[begin pattern_buffer]]] */
- /* Space that holds the compiled pattern. It is declared as
- `unsigned char *' because its elements are
- sometimes used as array indexes. */
- unsigned char *buffer;
-
- /* Number of bytes to which `buffer' points. */
- size_t allocated;
-
- /* Number of bytes actually used in `buffer'. */
- size_t used;
-
-#ifdef emacs
- /* Charset of unibyte characters at compiling time. */
- int charset_unibyte;
-#else
- /* Syntax setting with which the pattern was compiled. */
- reg_syntax_t syntax;
-#endif
- /* Pointer to a fastmap, if any, otherwise zero. re_search uses
- the fastmap, if there is one, to skip over impossible
- starting points for matches. */
- char *fastmap;
-
- /* Either a translate table to apply to all characters before
- comparing them, or zero for no translation. The translation
- is applied to a pattern when it is compiled and to a string
- when it is matched. */
- RE_TRANSLATE_TYPE translate;
-
- /* Number of subexpressions found by the compiler. */
- size_t re_nsub;
-
- /* Zero if this pattern cannot match the empty string, one else.
- Well, in truth it's used only in `re_search_2', to see
- whether or not we should use the fastmap, so we don't set
- this absolutely perfectly; see `re_compile_fastmap'. */
- unsigned can_be_null : 1;
-
- /* If REGS_UNALLOCATED, allocate space in the `regs' structure
- for `max (RE_NREGS, re_nsub + 1)' groups.
- If REGS_REALLOCATE, reallocate space if necessary.
- If REGS_FIXED, use what's there. */
-#define REGS_UNALLOCATED 0
-#define REGS_REALLOCATE 1
-#define REGS_FIXED 2
- unsigned regs_allocated : 2;
-
- /* Set to zero when `regex_compile' compiles a pattern; set to one
- by `re_compile_fastmap' if it updates the fastmap. */
- unsigned fastmap_accurate : 1;
-
- /* If set, `re_match_2' does not return information about
- subexpressions. */
- unsigned no_sub : 1;
-
- /* If set, a beginning-of-line anchor doesn't match at the
- beginning of the string. */
- unsigned not_bol : 1;
-
- /* Similarly for an end-of-line anchor. */
- unsigned not_eol : 1;
-
- /* If true, the compilation of the pattern had to look up the syntax table,
- so the compiled pattern is only valid for the current syntax table. */
- unsigned used_syntax : 1;
-
-#ifdef emacs
- /* If true, multi-byte form in the regexp pattern should be
- recognized as a multibyte character. */
- unsigned multibyte : 1;
-
- /* If true, multi-byte form in the target of match should be
- recognized as a multibyte character. */
- unsigned target_multibyte : 1;
-#endif
-
-/* [[[end pattern_buffer]]] */
-};
-
-typedef struct re_pattern_buffer regex_t;
-
-/* POSIX 1003.1-2008 requires that regoff_t be at least as wide as
- ptrdiff_t and ssize_t. We don't know of any hosts where ptrdiff_t
- is wider than ssize_t, so ssize_t is safe. ptrdiff_t is not
- necessarily visible here, so use ssize_t. */
-typedef ssize_t regoff_t;
-
-
-/* This is the structure we store register match data in. See
- regex.texinfo for a full description of what registers match. */
-struct re_registers
-{
- unsigned num_regs;
- regoff_t *start;
- regoff_t *end;
-};
-
-
-/* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
- `re_match_2' returns information about at least this many registers
- the first time a `regs' structure is passed. */
-#ifndef RE_NREGS
-# define RE_NREGS 30
-#endif
-
-
-/* POSIX specification for registers. Aside from the different names than
- `re_registers', POSIX uses an array of structures, instead of a
- structure of arrays. */
-typedef struct
-{
- regoff_t rm_so; /* Byte offset from string's start to substring's start. */
- regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
-} regmatch_t;
-
-/* Declarations for routines. */
-
-#ifndef emacs
-
-/* Sets the current default syntax to SYNTAX, and return the old syntax.
- You can also simply assign to the `re_syntax_options' variable. */
-extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
-
-#endif
-
-/* Compile the regular expression PATTERN, with length LENGTH
- and syntax given by the global `re_syntax_options', into the buffer
- BUFFER. Return NULL if successful, and an error string if not. */
-extern const char *re_compile_pattern (const char *__pattern, size_t __length,
-#ifdef emacs
- bool posix_backtracking,
- const char *whitespace_regexp,
-#endif
- struct re_pattern_buffer *__buffer);
-
-
-/* Compile a fastmap for the compiled pattern in BUFFER; used to
- accelerate searches. Return 0 if successful and -2 if was an
- internal error. */
-extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
-
-
-/* Search in the string STRING (with length LENGTH) for the pattern
- compiled into BUFFER. Start searching at position START, for RANGE
- characters. Return the starting position of the match, -1 for no
- match, or -2 for an internal error. Also return register
- information in REGS (if REGS and BUFFER->no_sub are nonzero). */
-extern regoff_t re_search (struct re_pattern_buffer *__buffer,
- const char *__string, size_t __length,
- ssize_t __start, ssize_t __range,
- struct re_registers *__regs);
-
-
-/* Like `re_search', but search in the concatenation of STRING1 and
- STRING2. Also, stop searching at index START + STOP. */
-extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
- const char *__string1, size_t __length1,
- const char *__string2, size_t __length2,
- ssize_t __start, ssize_t __range,
- struct re_registers *__regs,
- ssize_t __stop);
-
-
-/* Like `re_search', but return how many characters in STRING the regexp
- in BUFFER matched, starting at position START. */
-extern regoff_t re_match (struct re_pattern_buffer *__buffer,
- const char *__string, size_t __length,
- ssize_t __start, struct re_registers *__regs);
-
-
-/* Relates to `re_match' as `re_search_2' relates to `re_search'. */
-extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
- const char *__string1, size_t __length1,
- const char *__string2, size_t __length2,
- ssize_t __start, struct re_registers *__regs,
- ssize_t __stop);
-
-
-/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
- ENDS. Subsequent matches using BUFFER and REGS will use this memory
- for recording register information. STARTS and ENDS must be
- allocated with malloc, and must each be at least `NUM_REGS * sizeof
- (regoff_t)' bytes long.
-
- If NUM_REGS == 0, then subsequent matches should allocate their own
- register data.
-
- Unless this function is called, the first search or match using
- PATTERN_BUFFER will allocate its own register data, without
- freeing the old data. */
-extern void re_set_registers (struct re_pattern_buffer *__buffer,
- struct re_registers *__regs,
- unsigned __num_regs,
- regoff_t *__starts, regoff_t *__ends);
-
-#if defined _REGEX_RE_COMP || defined _LIBC
-# ifndef _CRAY
-/* 4.2 bsd compatibility. */
-extern char *re_comp (const char *);
-extern int re_exec (const char *);
-# endif
-#endif
-
-/* GCC 2.95 and later have "__restrict"; C99 compilers have
- "restrict", and "configure" may have defined "restrict".
- Other compilers use __restrict, __restrict__, and _Restrict, and
- 'configure' might #define 'restrict' to those words, so pick a
- different name. */
-#ifndef _Restrict_
-# if 199901L <= __STDC_VERSION__
-# define _Restrict_ restrict
-# elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)
-# define _Restrict_ __restrict
-# else
-# define _Restrict_
-# endif
-#endif
-/* gcc 3.1 and up support the [restrict] syntax. Don't trust
- sys/cdefs.h's definition of __restrict_arr, though, as it
- mishandles gcc -ansi -pedantic. */
-#ifndef _Restrict_arr_
-# if ((199901L <= __STDC_VERSION__ \
- || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__)) \
- && !defined __STRICT_ANSI__))
\
- && !defined __GNUG__)
-# define _Restrict_arr_ _Restrict_
-# else
-# define _Restrict_arr_
-# endif
-#endif
-
-/* POSIX compatibility. */
-extern reg_errcode_t regcomp (regex_t *_Restrict_ __preg,
- const char *_Restrict_ __pattern,
- int __cflags);
-
-extern reg_errcode_t regexec (const regex_t *_Restrict_ __preg,
- const char *_Restrict_ __string, size_t __nmatch,
- regmatch_t __pmatch[_Restrict_arr_],
- int __eflags);
-
-extern size_t regerror (int __errcode, const regex_t * __preg,
- char *__errbuf, size_t __errbuf_size);
-
-extern void regfree (regex_t *__preg);
-
-
-#ifdef __cplusplus
-}
-#endif /* C++ */
-
-/* For platform which support the ISO C amendment 1 functionality we
- support user defined character classes. */
-#if WIDE_CHAR_SUPPORT
-/* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. */
-# include <wchar.h>
-# include <wctype.h>
-
-typedef wctype_t re_wctype_t;
-typedef wchar_t re_wchar_t;
-# define re_wctype wctype
-# define re_iswctype iswctype
-# define re_wctype_to_bit(cc) 0
-#else
-# ifndef emacs
-# define btowc(c) c
-# endif
-
-/* Character classes. */
-typedef enum { RECC_ERROR = 0,
- RECC_ALNUM, RECC_ALPHA, RECC_WORD,
- RECC_GRAPH, RECC_PRINT,
- RECC_LOWER, RECC_UPPER,
- RECC_PUNCT, RECC_CNTRL,
- RECC_DIGIT, RECC_XDIGIT,
- RECC_BLANK, RECC_SPACE,
- RECC_MULTIBYTE, RECC_NONASCII,
- RECC_ASCII, RECC_UNIBYTE
-} re_wctype_t;
-
-extern char re_iswctype (int ch, re_wctype_t cc);
-extern re_wctype_t re_wctype_parse (const unsigned char **strp, unsigned
limit);
-
-typedef int re_wchar_t;
-
-#endif /* not WIDE_CHAR_SUPPORT */
-
-#endif /* regex.h */
-
diff --git a/src/search.c b/src/search.c
index 5385c89..f5c7719 100644
--- a/src/search.c
+++ b/src/search.c
@@ -30,7 +30,7 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#include "blockinput.h"
#include "intervals.h"
-#include "regex.h"
+#include "regex-emacs.h"
#define REGEXP_CACHE_SIZE 20
@@ -59,8 +59,8 @@ static struct regexp_cache searchbufs[REGEXP_CACHE_SIZE];
static struct regexp_cache *searchbuf_head;
-/* Every call to re_match, etc., must pass &search_regs as the regs
- argument unless you can show it is unnecessary (i.e., if re_match
+/* Every call to re_search, etc., must pass &search_regs as the regs
+ argument unless you can show it is unnecessary (i.e., if re_search
is certainly going to be called again before region-around-match
can be called).
@@ -132,7 +132,7 @@ compile_pattern_1 (struct regexp_cache *cp, Lisp_Object
pattern,
eassert (!cp->busy);
cp->regexp = Qnil;
- cp->buf.translate = (! NILP (translate) ? translate : make_fixnum (0));
+ cp->buf.translate = translate;
cp->posix = posix;
cp->buf.multibyte = STRING_MULTIBYTE (pattern);
cp->buf.charset_unibyte = charset_unibyte;
@@ -238,7 +238,7 @@ compile_pattern (Lisp_Object pattern, struct re_registers
*regp,
&& !cp->busy
&& STRING_MULTIBYTE (cp->regexp) == STRING_MULTIBYTE (pattern)
&& !NILP (Fstring_equal (cp->regexp, pattern))
- && EQ (cp->buf.translate, (! NILP (translate) ? translate :
make_fixnum (0)))
+ && EQ (cp->buf.translate, translate)
&& cp->posix == posix
&& (EQ (cp->syntax_table, Qt)
|| EQ (cp->syntax_table, BVAR (current_buffer, syntax_table)))
@@ -290,7 +290,8 @@ looking_at_1 (Lisp_Object string, bool posix)
if (running_asynch_code)
save_search_regs ();
- /* This is so set_image_of_range_1 in regex.c can find the EQV table. */
+ /* This is so set_image_of_range_1 in regex-emacs.c can find the EQV
+ table. */
set_char_table_extras (BVAR (current_buffer, case_canon_table), 2,
BVAR (current_buffer, case_eqv_table));
@@ -410,7 +411,8 @@ string_match_1 (Lisp_Object regexp, Lisp_Object string,
Lisp_Object start,
pos_byte = string_char_to_byte (string, pos);
}
- /* This is so set_image_of_range_1 in regex.c can find the EQV table. */
+ /* This is so set_image_of_range_1 in regex-emacs.c can find the EQV
+ table. */
set_char_table_extras (BVAR (current_buffer, case_canon_table), 2,
BVAR (current_buffer, case_eqv_table));
@@ -1062,7 +1064,8 @@ search_command (Lisp_Object string, Lisp_Object bound,
Lisp_Object noerror,
lim_byte = CHAR_TO_BYTE (lim);
}
- /* This is so set_image_of_range_1 in regex.c can find the EQV table. */
+ /* This is so set_image_of_range_1 in regex-emacs.c can find the EQV
+ table. */
set_char_table_extras (BVAR (current_buffer, case_canon_table), 2,
BVAR (current_buffer, case_eqv_table));
@@ -2186,8 +2189,8 @@ set_search_regs (ptrdiff_t beg_byte, ptrdiff_t nbytes)
the match position. */
if (search_regs.num_regs == 0)
{
- search_regs.start = xmalloc (2 * sizeof (regoff_t));
- search_regs.end = xmalloc (2 * sizeof (regoff_t));
+ search_regs.start = xmalloc (2 * sizeof *search_regs.start);
+ search_regs.end = xmalloc (2 * sizeof *search_regs.end);
search_regs.num_regs = 2;
}
@@ -2998,9 +3001,9 @@ If optional arg RESEAT is non-nil, make markers on LIST
point nowhere. */)
memory_full (SIZE_MAX);
search_regs.start =
xpalloc (search_regs.start, &num_regs, length - num_regs,
- min (PTRDIFF_MAX, UINT_MAX), sizeof (regoff_t));
+ min (PTRDIFF_MAX, UINT_MAX), sizeof *search_regs.start);
search_regs.end =
- xrealloc (search_regs.end, num_regs * sizeof (regoff_t));
+ xrealloc (search_regs.end, num_regs * sizeof *search_regs.end);
for (i = search_regs.num_regs; i < num_regs; i++)
search_regs.start[i] = -1;
@@ -3055,12 +3058,9 @@ If optional arg RESEAT is non-nil, make markers on LIST
point nowhere. */)
XSETFASTINT (marker, 0);
CHECK_FIXNUM_COERCE_MARKER (marker);
- if ((XFIXNUM (from) < 0
- ? TYPE_MINIMUM (regoff_t) <= XFIXNUM (from)
- : XFIXNUM (from) <= TYPE_MAXIMUM (regoff_t))
- && (XFIXNUM (marker) < 0
- ? TYPE_MINIMUM (regoff_t) <= XFIXNUM (marker)
- : XFIXNUM (marker) <= TYPE_MAXIMUM (regoff_t)))
+ if (PTRDIFF_MIN <= XFIXNUM (from) && XFIXNUM (from) <= PTRDIFF_MAX
+ && PTRDIFF_MIN <= XFIXNUM (marker)
+ && XFIXNUM (marker) <= PTRDIFF_MAX)
{
search_regs.start[i] = XFIXNUM (from);
search_regs.end[i] = XFIXNUM (marker);
diff --git a/src/syntax.c b/src/syntax.c
index ee83ed0..a9bc36a 100644
--- a/src/syntax.c
+++ b/src/syntax.c
@@ -23,7 +23,7 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#include "lisp.h"
#include "character.h"
#include "buffer.h"
-#include "regex.h"
+#include "regex-emacs.h"
#include "syntax.h"
#include "intervals.h"
#include "category.h"
@@ -267,9 +267,10 @@ SETUP_SYNTAX_TABLE (ptrdiff_t from, ptrdiff_t count)
If it is t (which is only used in fast_c_string_match_ignore_case),
ignore properties altogether.
- This is meant for regex.c to use. For buffers, regex.c passes arguments
- to the UPDATE_SYNTAX_TABLE functions which are relative to BEGV.
- So if it is a buffer, we set the offset field to BEGV. */
+ This is meant for regex-emacs.c to use. For buffers, regex-emacs.c
+ passes arguments to the UPDATE_SYNTAX_TABLE functions which are
+ relative to BEGV. So if it is a buffer, we set the offset field to
+ BEGV. */
void
SETUP_SYNTAX_TABLE_FOR_OBJECT (Lisp_Object object,
@@ -3729,7 +3730,7 @@ syms_of_syntax (void)
staticpro (&gl_state.current_syntax_table);
staticpro (&gl_state.old_prop);
- /* Defined in regex.c. */
+ /* Defined in regex-emacs.c. */
staticpro (&re_match_object);
DEFSYM (Qscan_error, "scan-error");
diff --git a/src/terminal.c b/src/terminal.c
index 76ac4ca..18982fe 100644
--- a/src/terminal.c
+++ b/src/terminal.c
@@ -483,7 +483,7 @@ static Lisp_Object
store_terminal_param (struct terminal *t, Lisp_Object parameter, Lisp_Object
value)
{
Lisp_Object old_alist_elt = Fassq (parameter, t->param_alist);
- if (EQ (old_alist_elt, Qnil))
+ if (NILP (old_alist_elt))
{
tset_param_alist (t, Fcons (Fcons (parameter, value), t->param_alist));
return Qnil;
diff --git a/src/textprop.c b/src/textprop.c
index 4bd4892..8e8baf4 100644
--- a/src/textprop.c
+++ b/src/textprop.c
@@ -1350,6 +1350,7 @@ set_text_properties (Lisp_Object start, Lisp_Object end,
Lisp_Object properties,
{
register INTERVAL i;
Lisp_Object ostart, oend;
+ bool first_time = true;
ostart = start;
oend = end;
@@ -1372,6 +1373,7 @@ set_text_properties (Lisp_Object start, Lisp_Object end,
Lisp_Object properties,
return Qt;
}
+ retry:
i = validate_interval_range (object, &start, &end, soft);
if (!i)
@@ -1391,8 +1393,22 @@ set_text_properties (Lisp_Object start, Lisp_Object end,
Lisp_Object properties,
return Qnil;
}
- if (BUFFERP (object) && !NILP (coherent_change_p))
- modify_text_properties (object, start, end);
+ if (BUFFERP (object) && !NILP (coherent_change_p) && first_time)
+ {
+ ptrdiff_t prev_length = LENGTH (i);
+ ptrdiff_t prev_pos = i->position;
+
+ modify_text_properties (object, start, end);
+ /* If someone called us recursively as a side effect of
+ modify_text_properties, and changed the intervals behind our
+ back, we cannot continue with I, because its data changed.
+ So we restart the interval analysis anew. */
+ if (LENGTH (i) != prev_length || i->position != prev_pos)
+ {
+ first_time = false;
+ goto retry;
+ }
+ }
set_text_properties_1 (start, end, properties, object, i);
@@ -2269,7 +2285,7 @@ verify_interval_modification (struct buffer *buf,
if (!inhibit_modification_hooks)
{
hooks = Fnreverse (hooks);
- while (! EQ (hooks, Qnil))
+ while (! NILP (hooks))
{
call_mod_hooks (Fcar (hooks), make_fixnum (start),
make_fixnum (end));
diff --git a/src/thread.c b/src/thread.c
index f11e3e5..1c73d93 100644
--- a/src/thread.c
+++ b/src/thread.c
@@ -681,7 +681,7 @@ invoke_thread_function (void)
{
ptrdiff_t count = SPECPDL_INDEX ();
- Ffuncall (1, ¤t_thread->function);
+ current_thread->result = Ffuncall (1, ¤t_thread->function);
return unbind_to (count, Qnil);
}
@@ -789,6 +789,7 @@ If NAME is given, it must be a string; it names the new
thread. */)
new_thread->m_last_thing_searched = Qnil; /* copy from parent? */
new_thread->m_saved_last_thing_searched = Qnil;
new_thread->m_current_buffer = current_thread->m_current_buffer;
+ new_thread->result = Qnil;
new_thread->error_symbol = Qnil;
new_thread->error_data = Qnil;
new_thread->event_object = Qnil;
@@ -933,12 +934,13 @@ thread_join_callback (void *arg)
DEFUN ("thread-join", Fthread_join, Sthread_join, 1, 1, 0,
doc: /* Wait for THREAD to exit.
-This blocks the current thread until THREAD exits or until
-the current thread is signaled.
-It is an error for a thread to try to join itself. */)
+This blocks the current thread until THREAD exits or until the current
+thread is signaled. It returns the result of the THREAD function. It
+is an error for a thread to try to join itself. */)
(Lisp_Object thread)
{
struct thread_state *tstate;
+ Lisp_Object error_symbol, error_data;
CHECK_THREAD (thread);
tstate = XTHREAD (thread);
@@ -946,10 +948,16 @@ It is an error for a thread to try to join itself. */)
if (tstate == current_thread)
error ("Cannot join current thread");
+ error_symbol = tstate->error_symbol;
+ error_data = tstate->error_data;
+
if (thread_alive_p (tstate))
flush_stack_call_func (thread_join_callback, tstate);
- return Qnil;
+ if (!NILP (error_symbol))
+ Fsignal (error_symbol, error_data);
+
+ return tstate->result;
}
DEFUN ("all-threads", Fall_threads, Sall_threads, 0, 0, 0,
@@ -973,11 +981,17 @@ DEFUN ("all-threads", Fall_threads, Sall_threads, 0, 0, 0,
return result;
}
-DEFUN ("thread-last-error", Fthread_last_error, Sthread_last_error, 0, 0, 0,
- doc: /* Return the last error form recorded by a dying thread. */)
- (void)
+DEFUN ("thread-last-error", Fthread_last_error, Sthread_last_error, 0, 1, 0,
+ doc: /* Return the last error form recorded by a dying thread.
+If CLEANUP is non-nil, remove this error form from history. */)
+ (Lisp_Object cleanup)
{
- return last_thread_error;
+ Lisp_Object result = last_thread_error;
+
+ if (!NILP (cleanup))
+ last_thread_error = Qnil;
+
+ return result;
}
@@ -1011,6 +1025,7 @@ init_main_thread (void)
main_thread.m_saved_last_thing_searched = Qnil;
main_thread.name = Qnil;
main_thread.function = Qnil;
+ main_thread.result = Qnil;
main_thread.error_symbol = Qnil;
main_thread.error_data = Qnil;
main_thread.event_object = Qnil;
@@ -1076,9 +1091,19 @@ syms_of_threads (void)
staticpro (&last_thread_error);
last_thread_error = Qnil;
+
+ Fprovide (intern_c_string ("threads"), Qnil);
}
DEFSYM (Qthreadp, "threadp");
DEFSYM (Qmutexp, "mutexp");
DEFSYM (Qcondition_variable_p, "condition-variable-p");
+
+ DEFVAR_LISP ("main-thread", Vmain_thread,
+ doc: /* The main thread of Emacs. */);
+#ifdef THREADS_ENABLED
+ XSETTHREAD (Vmain_thread, &main_thread);
+#else
+ Vmain_thread = Qnil;
+#endif
}
diff --git a/src/thread.h b/src/thread.h
index c10e5ec..8ecb008 100644
--- a/src/thread.h
+++ b/src/thread.h
@@ -19,7 +19,7 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#ifndef THREAD_H
#define THREAD_H
-#include "regex.h"
+#include "regex-emacs.h"
#ifdef WINDOWSNT
#include <sys/socket.h>
@@ -52,6 +52,9 @@ struct thread_state
/* The thread's function. */
Lisp_Object function;
+ /* The thread's result, if function has finished. */
+ Lisp_Object result;
+
/* If non-nil, this thread has been signaled. */
Lisp_Object error_symbol;
Lisp_Object error_data;
@@ -109,8 +112,8 @@ struct thread_state
struct buffer *m_current_buffer;
#define current_buffer (current_thread->m_current_buffer)
- /* Every call to re_match, etc., must pass &search_regs as the regs
- argument unless you can show it is unnecessary (i.e., if re_match
+ /* Every call to re_match_2, etc., must pass &search_regs as the regs
+ argument unless you can show it is unnecessary (i.e., if re_match_2
is certainly going to be called again before region-around-match
can be called).
diff --git a/src/w32cygwinx.c b/src/w32cygwinx.c
index 7bbb8be..bc40123 100644
--- a/src/w32cygwinx.c
+++ b/src/w32cygwinx.c
@@ -24,6 +24,16 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
#include "lisp.h"
#include "w32common.h"
+static Lisp_Object ATTRIBUTE_FORMAT_PRINTF (1, 2)
+format_string (char const *format, ...)
+{
+ va_list args;
+ va_start (args, format);
+ Lisp_Object str = vformat_string (format, args);
+ va_end (args);
+ return str;
+}
+
DEFUN ("w32-battery-status", Fw32_battery_status, Sw32_battery_status, 0, 0, 0,
doc: /* Get power status information from Windows system.
@@ -92,32 +102,17 @@ The following %-sequences are provided:
if (system_status.BatteryLifePercent > 100)
load_percentage = build_string ("N/A");
else
- {
- char buffer[16];
- snprintf (buffer, 16, "%d", system_status.BatteryLifePercent);
- load_percentage = build_string (buffer);
- }
+ load_percentage = format_string ("%d",
system_status.BatteryLifePercent);
if (seconds_left < 0)
seconds = minutes = hours = remain = build_string ("N/A");
else
{
- long m;
- double h;
- char buffer[16];
- snprintf (buffer, 16, "%ld", seconds_left);
- seconds = build_string (buffer);
-
- m = seconds_left / 60;
- snprintf (buffer, 16, "%ld", m);
- minutes = build_string (buffer);
-
- h = seconds_left / 3600.0;
- snprintf (buffer, 16, "%3.1f", h);
- hours = build_string (buffer);
-
- snprintf (buffer, 16, "%ld:%02ld", m / 60, m % 60);
- remain = build_string (buffer);
+ long m = seconds_left / 60;
+ seconds = format_string ("%ld", seconds_left);
+ minutes = format_string ("%ld", m);
+ hours = format_string ("%3.1f", seconds_left / 3600.0);
+ remain = format_string ("%ld:%02ld", m / 60, m % 60);
}
status = listn (CONSTYPE_HEAP, 8,
diff --git a/src/w32term.c b/src/w32term.c
index 9d224fe..8d189ae 100644
--- a/src/w32term.c
+++ b/src/w32term.c
@@ -1476,7 +1476,7 @@ x_draw_glyphless_glyph_string_foreground (struct
glyph_string *s)
{
sprintf ((char *) buf, "%0*X",
glyph->u.glyphless.ch < 0x10000 ? 4 : 6,
- (unsigned int) glyph->u.glyphless.ch);
+ (unsigned int) glyph->u.glyphless.ch & 0xffffff);
str = buf;
}
diff --git a/src/xdisp.c b/src/xdisp.c
index e30f800..76fde99 100644
--- a/src/xdisp.c
+++ b/src/xdisp.c
@@ -265,7 +265,7 @@ along with GNU Emacs. If not, see
<https://www.gnu.org/licenses/>. */
character to be delivered is a composed character, the iteration
calls composition_reseat_it and next_element_from_composition. If
they succeed to compose the character with one or more of the
- following characters, the whole sequence of characters that where
+ following characters, the whole sequence of characters that were
composed is recorded in the `struct composition_it' object that is
part of the buffer iterator. The composed sequence could produce
one or more font glyphs (called "grapheme clusters") on the screen.
@@ -467,12 +467,12 @@ static bool message_enable_multibyte;
looking for those `redisplay' bits (actually, there might be some such bits
set, but then only on objects which aren't displayed anyway).
- OTOH if it's non-zero we wil have to loop through all windows and then check
- the `redisplay' bit of the corresponding window, frame, and buffer, in order
- to decide whether that window needs attention or not. Note that we can't
- just look at the frame's redisplay bit to decide that the whole frame can be
- skipped, since even if the frame's redisplay bit is unset, some of its
- windows's redisplay bits may be set.
+ OTOH if it's non-zero we will have to loop through all windows and then
+ check the `redisplay' bit of the corresponding window, frame, and buffer, in
+ order to decide whether that window needs attention or not. Note that we
+ can't just look at the frame's redisplay bit to decide that the whole frame
+ can be skipped, since even if the frame's redisplay bit is unset, some of
+ its windows's redisplay bits may be set.
Mostly for historical reasons, windows_or_buffers_changed can also take
other non-zero values. In that case, the precise value doesn't matter (it
@@ -483,7 +483,7 @@ static bool message_enable_multibyte;
int windows_or_buffers_changed;
/* Nonzero if we should redraw the mode lines on the next redisplay.
- Similarly to `windows_or_buffers_changed', If it has value REDISPLAY_SOME,
+ Similarly to `windows_or_buffers_changed', if it has value REDISPLAY_SOME,
then only redisplay the mode lines in those buffers/windows/frames where the
`redisplay' bit has been set.
For any other value, redisplay all mode lines (the number used is then only
@@ -2281,9 +2281,9 @@ get_phys_cursor_geometry (struct window *w, struct
glyph_row *row,
int x, y, wd, h, h0, y0, ascent;
/* Compute the width of the rectangle to draw. If on a stretch
- glyph, and `x-stretch-block-cursor' is nil, don't draw a
- rectangle as wide as the glyph, but use a canonical character
- width instead. */
+ glyph, and `x-stretch-cursor' is nil, don't draw a rectangle
+ as wide as the glyph, but use a canonical character width
+ instead. */
wd = glyph->pixel_width;
x = w->phys_cursor.x;
@@ -8384,7 +8384,7 @@ next_element_from_buffer (struct it *it)
eassert (IT_CHARPOS (*it) >= BEGV);
eassert (NILP (it->string) && !it->s);
eassert (!it->bidi_p
- || (EQ (it->bidi_it.string.lstring, Qnil)
+ || (NILP (it->bidi_it.string.lstring)
&& it->bidi_it.string.s == NULL));
/* With bidi reordering, the character to display might not be the
@@ -11148,7 +11148,7 @@ display_echo_area (struct window *w)
/* Helper for display_echo_area. Display the current buffer which
contains the current echo area message in window W, a mini-window,
- a pointer to which is passed in A1. A2..A4 are currently not used.
+ a pointer to which is passed in A1. A2 is currently not used.
Change the height of W so that all of the message is displayed.
Value is true if height of W was changed. */
@@ -11209,8 +11209,8 @@ resize_echo_area_exactly (void)
/* Callback function for with_echo_area_buffer, when used from
resize_echo_area_exactly. A1 contains a pointer to the window to
resize, EXACTLY non-nil means resize the mini-window exactly to the
- size of the text displayed. A3 and A4 are not used. Value is what
- resize_mini_window returns. */
+ size of the text displayed. Value is what resize_mini_window
+ returns. */
static bool
resize_mini_window_1 (ptrdiff_t a1, Lisp_Object exactly)
@@ -16916,6 +16916,7 @@ redisplay_window (Lisp_Object window, bool
just_this_one_p)
/* We used to issue a CHECK_MARGINS argument to try_window here,
but this causes scrolling to fail when point begins inside
the scroll margin (bug#148) -- cyd */
+ clear_glyph_matrix (w->desired_matrix);
if (!try_window (window, startp, 0))
{
w->force_start = true;
@@ -23516,6 +23517,17 @@ move_elt_to_front (Lisp_Object elt, Lisp_Object list)
return list;
}
+/* Subroutine to call Fset_text_properties through
+ internal_condition_case_n. ARGS are the arguments of
+ Fset_text_properties, in order. */
+
+static Lisp_Object
+safe_set_text_properties (ptrdiff_t nargs, Lisp_Object *args)
+{
+ eassert (nargs == 4);
+ return Fset_text_properties (args[0], args[1], args[2], args[3]);
+}
+
/* Contribute ELT to the mode line for window IT->w. How it
translates into text depends on its data type.
@@ -23610,8 +23622,17 @@ display_mode_element (struct it *it, int depth, int
field_width, int precision,
= Fdelq (aelt, mode_line_proptrans_alist);
elt = Fcopy_sequence (elt);
- Fset_text_properties (make_fixnum (0), Flength (elt),
- props, elt);
+ /* PROPS might cause set-text-properties to signal
+ an error, so we call it via internal_condition_case_n,
+ to avoid an infloop in redisplay due to the error. */
+ internal_condition_case_n (safe_set_text_properties,
+ 4,
+ ((Lisp_Object [])
+ {make_fixnum (0),
+ Flength (elt),
+ props,
+ elt}),
+ Qt, safe_eval_handler);
/* Add this item to mode_line_proptrans_alist. */
mode_line_proptrans_alist
= Fcons (Fcons (elt, props),
diff --git a/src/xfaces.c b/src/xfaces.c
index 29168e2..23822b1 100644
--- a/src/xfaces.c
+++ b/src/xfaces.c
@@ -2971,7 +2971,7 @@ FRAME 0 means change the face on all frames, and change
the default
if (!UNSPECIFIEDP (value) && !IGNORE_DEFFACE_P (value))
if ((SYMBOLP (value)
&& !EQ (value, Qt)
- && !EQ (value, Qnil))
+ && !NILP (value))
/* Overline color. */
|| (STRINGP (value)
&& SCHARS (value) == 0))
@@ -2985,7 +2985,7 @@ FRAME 0 means change the face on all frames, and change
the default
if (!UNSPECIFIEDP (value) && !IGNORE_DEFFACE_P (value))
if ((SYMBOLP (value)
&& !EQ (value, Qt)
- && !EQ (value, Qnil))
+ && !NILP (value))
/* Strike-through color. */
|| (STRINGP (value)
&& SCHARS (value) == 0))
diff --git a/src/xfns.c b/src/xfns.c
index d9df03e..c455cfe 100644
--- a/src/xfns.c
+++ b/src/xfns.c
@@ -1456,7 +1456,7 @@ x_set_icon_type (struct frame *f, Lisp_Object arg,
Lisp_Object oldval)
if (STRINGP (oldval) && EQ (Fstring_equal (oldval, arg), Qt))
return;
}
- else if (!STRINGP (oldval) && EQ (oldval, Qnil) == EQ (arg, Qnil))
+ else if (!STRINGP (oldval) && NILP (oldval) == NILP (arg))
return;
block_input ();
@@ -5722,7 +5722,7 @@ If TERMINAL is omitted or nil, that stands for the
selected frame's display. */
{
struct x_display_info *dpyinfo = check_x_display_info (terminal);
- XSynchronize (dpyinfo->display, !EQ (on, Qnil));
+ XSynchronize (dpyinfo->display, !NILP (on));
return Qnil;
}
diff --git a/src/xmenu.c b/src/xmenu.c
index e6740af..10e882a 100644
--- a/src/xmenu.c
+++ b/src/xmenu.c
@@ -1487,7 +1487,7 @@ x_menu_show (struct frame *f, int x, int y, int menuflags,
i = 0;
while (i < menu_items_used)
{
- if (EQ (AREF (menu_items, i), Qnil))
+ if (NILP (AREF (menu_items, i)))
{
submenu_stack[submenu_depth++] = save_wv;
save_wv = prev_wv;
@@ -1656,7 +1656,7 @@ x_menu_show (struct frame *f, int x, int y, int menuflags,
i = 0;
while (i < menu_items_used)
{
- if (EQ (AREF (menu_items, i), Qnil))
+ if (NILP (AREF (menu_items, i)))
{
subprefix_stack[submenu_depth++] = prefix;
prefix = entry;
diff --git a/src/xselect.c b/src/xselect.c
index 984d95d..dd3da8e 100644
--- a/src/xselect.c
+++ b/src/xselect.c
@@ -2094,7 +2094,7 @@ On Nextstep, TERMINAL is unused. */)
struct frame *f = frame_for_x_selection (terminal);
CHECK_SYMBOL (selection);
- if (EQ (selection, Qnil)) selection = QPRIMARY;
+ if (NILP (selection)) selection = QPRIMARY;
if (EQ (selection, Qt)) selection = QSECONDARY;
if (f && !NILP (LOCAL_SELECTION (selection, FRAME_DISPLAY_INFO (f))))
@@ -2124,7 +2124,7 @@ On Nextstep, TERMINAL is unused. */)
struct x_display_info *dpyinfo;
CHECK_SYMBOL (selection);
- if (EQ (selection, Qnil)) selection = QPRIMARY;
+ if (NILP (selection)) selection = QPRIMARY;
if (EQ (selection, Qt)) selection = QSECONDARY;
if (!f)
diff --git a/src/xwidget.c b/src/xwidget.c
index d6b39fe..4739993 100644
--- a/src/xwidget.c
+++ b/src/xwidget.c
@@ -1099,7 +1099,7 @@ xwidget_view_lookup (struct xwidget *xw, struct window *w)
ret = Fxwidget_view_lookup (xwidget, window);
- return EQ (ret, Qnil) ? NULL : XXWIDGET_VIEW (ret);
+ return NILP (ret) ? NULL : XXWIDGET_VIEW (ret);
}
struct xwidget *
diff --git a/test/lisp/auth-source-tests.el b/test/lisp/auth-source-tests.el
index be516f2..ca8a3eb 100644
--- a/test/lisp/auth-source-tests.el
+++ b/test/lisp/auth-source-tests.el
@@ -344,5 +344,25 @@
"session"
(format "address@hidden" (plist-get auth-info :user) (plist-get auth-info
:host)))))
+(ert-deftest auth-source-delete ()
+ (let* ((netrc-file (make-temp-file "auth-source-test" nil nil "\
+machine a1 port a2 user a3 password a4
+machine b1 port b2 user b3 password b4
+machine c1 port c2 user c3 password c4\n"))
+ (auth-sources (list netrc-file))
+ (auth-source-do-cache nil)
+ (expected '((:host "a1" :port "a2" :user "a3" :secret "a4")))
+ (parameters '(:max 1 :host t)))
+ (unwind-protect
+ (let ((found (apply #'auth-source-delete parameters)))
+ (dolist (f found)
+ (let ((s (plist-get f :secret)))
+ (setf f (plist-put f :secret
+ (if (functionp s) (funcall s) s)))))
+ ;; Note: The netrc backend doesn't delete anything, so
+ ;; this is actually the same as `auth-source-search'.
+ (should (equal found expected)))
+ (delete-file netrc-file))))
+
(provide 'auth-source-tests)
;;; auth-source-tests.el ends here
diff --git a/test/lisp/calendar/todo-mode-tests.el
b/test/lisp/calendar/todo-mode-tests.el
index 159294f..325faef 100644
--- a/test/lisp/calendar/todo-mode-tests.el
+++ b/test/lisp/calendar/todo-mode-tests.el
@@ -25,6 +25,7 @@
;;; Code:
(require 'ert)
+(require 'ert-x)
(require 'todo-mode)
(defvar todo-test-data-dir
@@ -561,11 +562,12 @@ source file is different."
;; Headers in the todo file are still hidden.
(should (equal (overlay-get (todo-get-overlay 'header) 'display) "")))))
-(defun todo-test--insert-item (item &optional priority)
+(defun todo-test--insert-item (item &optional priority
+ _arg diary-type date-type time where)
"Insert string ITEM into current category with priority PRIORITY.
-Use defaults for all other item insertion parameters. This
-provides a noninteractive API for todo-insert-item for use in
-automatic testing."
+The remaining arguments (except _ARG, which is ignored) specify
+item insertion parameters. This provides a noninteractive API
+for todo-insert-item for use in automatic testing."
(cl-letf (((symbol-function 'read-from-minibuffer)
(lambda (_prompt) item))
((symbol-function 'read-number) ; For todo-set-item-priority
@@ -581,6 +583,186 @@ automatic testing."
(todo-test--insert-item item 1)
(should (equal (overlay-get (todo-get-overlay 'header) 'display) "")))))
+(defun todo-test--done-items-separator (&optional eol)
+ "Set up test of command interaction with done items separator.
+With non-nil argument EOL, return the position at the end of the
+separator, otherwise, return the position at the beginning."
+ (todo-test--show 1)
+ (goto-char (point-max))
+ ;; See comment about recentering in todo-test-raise-lower-priority.
+ (set-window-buffer nil (current-buffer))
+ (todo-toggle-view-done-items)
+ ;; FIXME: Point should now be on the first done item, and in batch
+ ;; testing it is, so we have to move back one line to the done items
+ ;; separator; but for some reason, in the graphical test
+ ;; environment, it stays on the last empty line of the todo items
+ ;; section, so there we have to advance one character to the done
+ ;; items separator.
+ (if (display-graphic-p)
+ (forward-char)
+ (forward-line -1))
+ (if eol (forward-char)))
+
+(ert-deftest todo-test-done-items-separator01-bol ()
+ "Test item copying and here insertion at BOL of separator.
+Both should be user errors."
+ (with-todo-test
+ (todo-test--done-items-separator)
+ (let* ((copy-err "Item copying is not valid here")
+ (here-err "Item insertion is not valid here")
+ (insert-item-test (lambda (where)
+ (should-error (todo-insert-item--basic
+ nil nil nil nil where)))))
+ (should (string= copy-err (cadr (funcall insert-item-test 'copy))))
+ (should (string= here-err (cadr (funcall insert-item-test 'here)))))))
+
+(ert-deftest todo-test-done-items-separator01-eol ()
+ "Test item copying and here insertion at EOL of separator.
+Both should be user errors."
+ (with-todo-test
+ (todo-test--done-items-separator 'eol)
+ (let* ((copy-err "Item copying is not valid here")
+ (here-err "Item insertion is not valid here")
+ (insert-item-test (lambda (where)
+ (should-error (todo-insert-item--basic
+ nil nil nil nil where)))))
+ (should (string= copy-err (cadr (funcall insert-item-test 'copy))))
+ (should (string= here-err (cadr (funcall insert-item-test 'here)))))))
+
+(ert-deftest todo-test-done-items-separator02-bol ()
+ "Test item editing commands at BOL of done items separator.
+They should all be noops."
+ (with-todo-test
+ (todo-test--done-items-separator)
+ (should-not (todo-item-done))
+ (should-not (todo-raise-item-priority))
+ (should-not (todo-lower-item-priority))
+ (should-not (called-interactively-p #'todo-set-item-priority))
+ (should-not (called-interactively-p #'todo-move-item))
+ (should-not (called-interactively-p #'todo-delete-item))
+ (should-not (called-interactively-p #'todo-edit-item))))
+
+(ert-deftest todo-test-done-items-separator02-eol ()
+ "Test item editing command at EOL of done items separator.
+They should all be noops."
+ (with-todo-test
+ (todo-test--done-items-separator 'eol)
+ (should-not (todo-item-done))
+ (should-not (todo-raise-item-priority))
+ (should-not (todo-lower-item-priority))
+ (should-not (called-interactively-p #'todo-set-item-priority))
+ (should-not (called-interactively-p #'todo-move-item))
+ (should-not (called-interactively-p #'todo-delete-item))
+ (should-not (called-interactively-p #'todo-edit-item))))
+
+(ert-deftest todo-test-done-items-separator03-bol ()
+ "Test item marking at BOL of done items separator.
+This should be a noop, adding no marks to the category."
+ (with-todo-test
+ (todo-test--done-items-separator)
+ (call-interactively #'todo-toggle-mark-item)
+ (should-not (assoc (todo-current-category) todo-categories-with-marks))))
+
+(ert-deftest todo-test-done-items-separator03-eol ()
+ "Test item marking at EOL of done items separator.
+This should be a noop, adding no marks to the category."
+ (with-todo-test
+ (todo-test--done-items-separator 'eol)
+ (call-interactively #'todo-toggle-mark-item)
+ (should-not (assoc (todo-current-category) todo-categories-with-marks))))
+
+(ert-deftest todo-test-done-items-separator04-bol ()
+ "Test moving to previous item from BOL of done items separator.
+This should move point to the last not done todo item."
+ (with-todo-test
+ (todo-test--done-items-separator)
+ (let ((last-item (save-excursion
+ ;; Move to empty line after last todo item.
+ (forward-line -1)
+ (todo-previous-item)
+ (todo-item-string))))
+ (should (string= last-item (save-excursion
+ (todo-previous-item)
+ (todo-item-string)))))))
+
+(ert-deftest todo-test-done-items-separator04-eol ()
+ "Test moving to previous item from EOL of done items separator.
+This should move point to the last not done todo item."
+ (with-todo-test
+ (todo-test--done-items-separator 'eol)
+ (let ((last-item (save-excursion
+ ;; Move to empty line after last todo item.
+ (forward-line -1)
+ (todo-previous-item)
+ (todo-item-string))))
+ (should (string= last-item (save-excursion
+ (todo-previous-item)
+ (todo-item-string)))))))
+
+(ert-deftest todo-test-done-items-separator05-bol ()
+ "Test moving to next item from BOL of done items separator.
+This should move point to the first done todo item."
+ (with-todo-test
+ (todo-test--done-items-separator)
+ (let ((first-done (save-excursion
+ ;; Move to empty line after last todo item.
+ (forward-line -1)
+ (todo-next-item)
+ (todo-item-string))))
+ (should (string= first-done (save-excursion
+ (todo-next-item)
+ (todo-item-string)))))))
+
+(ert-deftest todo-test-done-items-separator05-eol ()
+ "Test moving to next item from EOL of done items separator.
+This should move point to the first done todo item."
+ (with-todo-test
+ (todo-test--done-items-separator 'eol)
+ (let ((first-done (save-excursion
+ ;; Move to empty line after last todo item.
+ (forward-line -1)
+ (todo-next-item)
+ (todo-item-string))))
+ (should (string= first-done (save-excursion
+ (todo-next-item)
+ (todo-item-string)))))))
+
+;; Item highlighting uses hl-line-mode, which enables highlighting in
+;; post-command-hook. For some reason, in the test environment, the
+;; hook function is not automatically run, so after enabling item
+;; highlighting, use ert-simulate-command around the next command,
+;; which explicitly runs the hook function.
+(ert-deftest todo-test-done-items-separator06-bol ()
+ "Test enabling item highlighting at BOL of done items separator.
+Subsequently moving to an item should show it highlighted."
+ (with-todo-test
+ (todo-test--done-items-separator)
+ (call-interactively #'todo-toggle-item-highlighting)
+ (ert-simulate-command '(todo-previous-item))
+ (should (eq 'hl-line (get-char-property (point) 'face)))))
+
+(ert-deftest todo-test-done-items-separator06-eol ()
+ "Test enabling item highlighting at EOL of done items separator.
+Subsequently moving to an item should show it highlighted."
+ (with-todo-test
+ (todo-test--done-items-separator 'eol)
+ (todo-toggle-item-highlighting)
+ (forward-line -1)
+ (ert-simulate-command '(todo-previous-item))
+ (should (eq 'hl-line (get-char-property (point) 'face)))))
+
+(ert-deftest todo-test-done-items-separator07 ()
+ "Test item highlighting when crossing done items separator.
+The highlighting should remain enabled."
+ (with-todo-test
+ (todo-test--done-items-separator)
+ (todo-previous-item)
+ (todo-toggle-item-highlighting)
+ (todo-next-item) ; Now on empty line above separator.
+ (forward-line) ; Now on separator.
+ (ert-simulate-command '(forward-line)) ; Now on first done item.
+ (should (eq 'hl-line (get-char-property (point) 'face)))))
+
(provide 'todo-mode-tests)
;;; todo-mode-tests.el ends here
diff --git a/test/lisp/custom-tests.el b/test/lisp/custom-tests.el
new file mode 100644
index 0000000..96887f8
--- /dev/null
+++ b/test/lisp/custom-tests.el
@@ -0,0 +1,87 @@
+;;; custom-tests.el --- tests for custom.el -*- lexical-binding: t -*-
+
+;; Copyright (C) 2018 Free Software Foundation, Inc.
+
+;; This file is part of GNU Emacs.
+
+;; GNU Emacs is free software; you can redistribute it and/or modify
+;; it under the terms of the GNU General Public License as published by
+;; the Free Software Foundation, either version 3 of the License, or
+;; (at your option) any later version.
+
+;; GNU Emacs is distributed in the hope that it will be useful,
+;; but WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+;; GNU General Public License for more details.
+
+;; You should have received a copy of the GNU General Public License
+;; along with this program. If not, see <https://www.gnu.org/licenses/>.
+
+;;; Code:
+
+(require 'ert)
+
+(ert-deftest custom-theme--load-path ()
+ "Test `custom-theme--load-path' behavior."
+ (let ((tmpdir (file-name-as-directory (make-temp-file "custom-tests-" t))))
+ (unwind-protect
+ ;; Create all temporary files under the same deletable parent.
+ (let ((temporary-file-directory tmpdir))
+ ;; Path is empty.
+ (let ((custom-theme-load-path ()))
+ (should (null (custom-theme--load-path))))
+
+ ;; Path comprises non-existent file.
+ (let* ((name (make-temp-name tmpdir))
+ (custom-theme-load-path (list name)))
+ (should (not (file-exists-p name)))
+ (should (null (custom-theme--load-path))))
+
+ ;; Path comprises existing file.
+ (let* ((file (make-temp-file "file"))
+ (custom-theme-load-path (list file)))
+ (should (file-exists-p file))
+ (should (not (file-directory-p file)))
+ (should (null (custom-theme--load-path))))
+
+ ;; Path comprises existing directory.
+ (let* ((dir (make-temp-file "dir" t))
+ (custom-theme-load-path (list dir)))
+ (should (file-directory-p dir))
+ (should (equal (custom-theme--load-path) custom-theme-load-path)))
+
+ ;; Expand `custom-theme-directory' path element.
+ (let ((custom-theme-load-path '(custom-theme-directory)))
+ (let ((custom-theme-directory (make-temp-name tmpdir)))
+ (should (not (file-exists-p custom-theme-directory)))
+ (should (null (custom-theme--load-path))))
+ (let ((custom-theme-directory (make-temp-file "file")))
+ (should (file-exists-p custom-theme-directory))
+ (should (not (file-directory-p custom-theme-directory)))
+ (should (null (custom-theme--load-path))))
+ (let ((custom-theme-directory (make-temp-file "dir" t)))
+ (should (file-directory-p custom-theme-directory))
+ (should (equal (custom-theme--load-path)
+ (list custom-theme-directory)))))
+
+ ;; Expand t path element.
+ (let ((custom-theme-load-path '(t)))
+ (let ((data-directory (make-temp-name tmpdir)))
+ (should (not (file-exists-p data-directory)))
+ (should (null (custom-theme--load-path))))
+ (let ((data-directory tmpdir)
+ (themedir (expand-file-name "themes" tmpdir)))
+ (should (not (file-exists-p themedir)))
+ (should (null (custom-theme--load-path)))
+ (with-temp-file themedir)
+ (should (file-exists-p themedir))
+ (should (not (file-directory-p themedir)))
+ (should (null (custom-theme--load-path)))
+ (delete-file themedir)
+ (make-directory themedir)
+ (should (file-directory-p themedir))
+ (should (equal (custom-theme--load-path) (list themedir))))))
+ (when (file-directory-p tmpdir)
+ (delete-directory tmpdir t)))))
+
+;;; custom-tests.el ends here
diff --git a/test/lisp/emacs-lisp/backtrace-tests.el
b/test/lisp/emacs-lisp/backtrace-tests.el
new file mode 100644
index 0000000..edd45c7
--- /dev/null
+++ b/test/lisp/emacs-lisp/backtrace-tests.el
@@ -0,0 +1,436 @@
+;;; backtrace-tests.el --- Tests for backtraces -*- lexical-binding: t; -*-
+
+;; Copyright (C) 2018 Free Software Foundation, Inc.
+
+;; Author: Gemini Lasswell
+
+;; This file is part of GNU Emacs.
+
+;; GNU Emacs is free software: you can redistribute it and/or modify
+;; it under the terms of the GNU General Public License as published by
+;; the Free Software Foundation, either version 3 of the License, or
+;; (at your option) any later version.
+
+;; GNU Emacs is distributed in the hope that it will be useful,
+;; but WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+;; GNU General Public License for more details.
+
+;; You should have received a copy of the GNU General Public License
+;; along with GNU Emacs. If not, see <https://www.gnu.org/licenses/>.
+
+;;; Code:
+
+(require 'backtrace)
+(require 'ert)
+(require 'ert-x)
+(require 'seq)
+
+;; Delay evaluation of the backtrace-creating functions until
+;; load so that the backtraces are the same whether this file
+;; is compiled or not.
+
+(eval-and-compile
+ (defconst backtrace-tests--uncompiled-functions
+ '(progn
+ (defun backtrace-tests--make-backtrace (arg)
+ (backtrace-tests--setup-buffer))
+
+ (defun backtrace-tests--setup-buffer ()
+ "Set up the current buffer in backtrace mode."
+ (backtrace-mode)
+ (setq backtrace-frames (backtrace-get-frames))
+ (let ((this-index))
+ ;; Discard all past `backtrace-tests-make-backtrace'.
+ (dotimes (index (length backtrace-frames))
+ (when (eq (backtrace-frame-fun (nth index backtrace-frames))
+ 'backtrace-tests--make-backtrace)
+ (setq this-index index)))
+ (setq backtrace-frames (seq-subseq backtrace-frames 0 (1+
this-index))))
+ (backtrace-print))))
+
+ (eval backtrace-tests--uncompiled-functions))
+
+(defun backtrace-tests--backtrace-lines ()
+ (if debugger-stack-frame-as-list
+ '(" (backtrace-get-frames)\n"
+ " (setq backtrace-frames (backtrace-get-frames))\n"
+ " (backtrace-tests--setup-buffer)\n"
+ " (backtrace-tests--make-backtrace %s)\n")
+ '(" backtrace-get-frames()\n"
+ " (setq backtrace-frames (backtrace-get-frames))\n"
+ " backtrace-tests--setup-buffer()\n"
+ " backtrace-tests--make-backtrace(%s)\n")))
+
+(defconst backtrace-tests--line-count (length
(backtrace-tests--backtrace-lines)))
+
+(defun backtrace-tests--backtrace-lines-with-locals ()
+ (let ((lines (backtrace-tests--backtrace-lines))
+ (locals '(" [no locals]\n"
+ " [no locals]\n"
+ " [no locals]\n"
+ " arg = %s\n")))
+ (apply #'append (cl-mapcar #'list lines locals))))
+
+(defun backtrace-tests--result (value)
+ (format (apply #'concat (backtrace-tests--backtrace-lines))
+ (cl-prin1-to-string value)))
+
+(defun backtrace-tests--result-with-locals (value)
+ (let ((str (cl-prin1-to-string value)))
+ (format (apply #'concat (backtrace-tests--backtrace-lines-with-locals))
+ str str)))
+
+;; TODO check that debugger-batch-max-lines still works
+
+(defconst backtrace-tests--header "Test header\n")
+(defun backtrace-tests--insert-header ()
+ (insert backtrace-tests--header))
+
+;;; Tests
+
+(ert-deftest backtrace-tests--variables ()
+ "Backtrace buffers can show and hide local variables."
+ (ert-with-test-buffer (:name "variables")
+ (let ((results (concat backtrace-tests--header
+ (backtrace-tests--result 'value)))
+ (last-frame (format (nth (1- backtrace-tests--line-count)
+ (backtrace-tests--backtrace-lines)) 'value))
+ (last-frame-with-locals
+ (format (apply #'concat (nthcdr (* 2 (1-
backtrace-tests--line-count))
+
(backtrace-tests--backtrace-lines-with-locals)))
+ 'value 'value)))
+ (backtrace-tests--make-backtrace 'value)
+ (setq backtrace-insert-header-function #'backtrace-tests--insert-header)
+ (backtrace-print)
+ (should (string= (backtrace-tests--get-substring (point-min) (point-max))
+ results))
+ ;; Go to the last frame.
+ (goto-char (point-max))
+ (forward-line -1)
+ ;; Turn on locals for that frame.
+ (backtrace-toggle-locals)
+ (should (string= (backtrace-tests--get-substring (point) (point-max))
+ last-frame-with-locals))
+ (should (string= (backtrace-tests--get-substring (point-min) (point-max))
+ (concat results
+ (format (car (last
(backtrace-tests--backtrace-lines-with-locals)))
+ 'value))))
+ ;; Turn off locals for that frame.
+ (backtrace-toggle-locals)
+ (should (string= (backtrace-tests--get-substring (point) (point-max))
+ last-frame))
+ (should (string= (backtrace-tests--get-substring (point-min) (point-max))
+ results))
+ ;; Turn all locals on.
+ (backtrace-toggle-locals '(4))
+ (should (string= (backtrace-tests--get-substring (point) (point-max))
+ last-frame-with-locals))
+ (should (string= (backtrace-tests--get-substring (point-min) (point-max))
+ (concat backtrace-tests--header
+ (backtrace-tests--result-with-locals 'value))))
+ ;; Turn all locals off.
+ (backtrace-toggle-locals '(4))
+ (should (string= (backtrace-tests--get-substring
+ (point) (+ (point) (length last-frame)))
+ last-frame))
+ (should (string= (backtrace-tests--get-substring (point-min) (point-max))
+ results)))))
+
+(ert-deftest backtrace-tests--backward-frame ()
+ "`backtrace-backward-frame' moves backward to the start of a frame."
+ (ert-with-test-buffer (:name "backward")
+ (let ((results (concat backtrace-tests--header
+ (backtrace-tests--result nil))))
+ (backtrace-tests--make-backtrace nil)
+ (setq backtrace-insert-header-function #'backtrace-tests--insert-header)
+ (backtrace-print)
+ (should (string= (backtrace-tests--get-substring (point-min) (point-max))
+ results))
+
+ ;; Try to move backward from header.
+ (goto-char (+ (point-min) (/ (length backtrace-tests--header) 2)))
+ (let ((pos (point)))
+ (should-error (backtrace-backward-frame))
+ (should (= pos (point))))
+
+ ;; Try to move backward from start of first line.
+ (forward-line)
+ (let ((pos (point)))
+ (should-error (backtrace-backward-frame))
+ (should (= pos (point))))
+
+ ;; Move backward from middle of line.
+ (let ((start (point)))
+ (forward-char (/ (length (nth 0 (backtrace-tests--backtrace-lines)))
2))
+ (backtrace-backward-frame)
+ (should (= start (point))))
+
+ ;; Move backward from end of buffer.
+ (goto-char (point-max))
+ (backtrace-backward-frame)
+ (let* ((last (format (car (last (backtrace-tests--backtrace-lines)))
nil))
+ (len (length last)))
+ (should (string= (buffer-substring-no-properties (point) (+ (point)
len))
+ last)))
+
+ ;; Move backward from start of line.
+ (backtrace-backward-frame)
+ (let* ((line (car (last (backtrace-tests--backtrace-lines) 2)))
+ (len (length line)))
+ (should (string= (buffer-substring-no-properties (point) (+ (point)
len))
+ line))))))
+
+(ert-deftest backtrace-tests--forward-frame ()
+ "`backtrace-forward-frame' moves forward to the start of a frame."
+ (ert-with-test-buffer (:name "forward")
+ (let* ((arg '(1 2 3))
+ (results (concat backtrace-tests--header
+ (backtrace-tests--result arg)))
+ (first-line (nth 0 (backtrace-tests--backtrace-lines))))
+ (backtrace-tests--make-backtrace arg)
+ (setq backtrace-insert-header-function #'backtrace-tests--insert-header)
+ (backtrace-print)
+ (should (string= (backtrace-tests--get-substring (point-min) (point-max))
+ results))
+ ;; Move forward from header.
+ (goto-char (+ (point-min) (/ (length backtrace-tests--header) 2)))
+ (backtrace-forward-frame)
+ (should (string= (backtrace-tests--get-substring
+ (point) (+ (point) (length first-line)))
+ first-line))
+
+ (let ((start (point))
+ (offset (/ (length first-line) 2))
+ (second-line (nth 1 (backtrace-tests--backtrace-lines))))
+ ;; Move forward from start of first frame.
+ (backtrace-forward-frame)
+ (should (string= (backtrace-tests--get-substring
+ (point) (+ (point) (length second-line)))
+ second-line))
+ ;; Move forward from middle of first frame.
+ (goto-char (+ start offset))
+ (backtrace-forward-frame)
+ (should (string= (backtrace-tests--get-substring
+ (point) (+ (point) (length second-line)))
+ second-line)))
+ ;; Try to move forward from middle of last frame.
+ (goto-char (- (point-max)
+ (/ 2 (length (car (last
(backtrace-tests--backtrace-lines)))))))
+ (should-error (backtrace-forward-frame))
+ ;; Try to move forward from end of buffer.
+ (goto-char (point-max))
+ (should-error (backtrace-forward-frame)))))
+
+(ert-deftest backtrace-tests--single-and-multi-line ()
+ "Forms in backtrace frames can be on a single line or on multiple lines."
+ (ert-with-test-buffer (:name "single-multi-line")
+ (let* ((arg '(lambda (x) ; Quote this so it isn't made into a closure.
+ (let ((number (1+ x)))
+ (+ x number))))
+ (header-string "Test header: ")
+ (header (format "%s%s\n" header-string arg))
+ (insert-header-function (lambda ()
+ (insert header-string)
+ (insert (backtrace-print-to-string arg))
+ (insert "\n")))
+ (results (concat header (backtrace-tests--result arg)))
+ (last-line (format (nth (1- backtrace-tests--line-count)
+ (backtrace-tests--backtrace-lines))
+ arg))
+ (last-line-locals (format (nth (1- (* 2
backtrace-tests--line-count))
+
(backtrace-tests--backtrace-lines-with-locals))
+ arg)))
+
+ (backtrace-tests--make-backtrace arg)
+ (setq backtrace-insert-header-function insert-header-function)
+ (backtrace-print)
+ (should (string= (backtrace-tests--get-substring (point-min) (point-max))
+ results))
+ ;; Check pp and collapse for the form in the header.
+ (goto-char (point-min))
+ (backtrace-tests--verify-single-and-multi-line header)
+ ;; Check pp and collapse for the last frame.
+ (goto-char (point-max))
+ (backtrace-backward-frame)
+ (backtrace-tests--verify-single-and-multi-line last-line)
+ ;; Check pp and collapse for local variables in the last line.
+ (goto-char (point-max))
+ (backtrace-backward-frame)
+ (backtrace-toggle-locals)
+ (forward-line)
+ (backtrace-tests--verify-single-and-multi-line last-line-locals))))
+
+(defun backtrace-tests--verify-single-and-multi-line (line)
+ "Verify that `backtrace-single-line' and `backtrace-multi-line' work at
point.
+Point should be at the beginning of a line, and LINE should be a
+string containing the text of the line at point. Assume that the
+line contains the strings \"lambda\" and \"number\"."
+ (let ((pos (point)))
+ (backtrace-multi-line)
+ ;; Verify point is still at the start of the line.
+ (should (= pos (point))))
+
+ ;; Verify the form now spans multiple lines.
+ (let ((pos (point)))
+ (search-forward "number")
+ (should-not (= pos (point-at-bol))))
+ ;; Collapse the form.
+ (backtrace-single-line)
+ ;; Verify that the form is now back on one line,
+ ;; and that point is at the same place.
+ (should (string= (backtrace-tests--get-substring
+ (- (point) 6) (point)) "number"))
+ (should-not (= (point) (point-at-bol)))
+ (should (string= (backtrace-tests--get-substring
+ (point-at-bol) (1+ (point-at-eol)))
+ line)))
+
+(ert-deftest backtrace-tests--print-circle ()
+ "Backtrace buffers can toggle `print-circle' syntax."
+ (ert-with-test-buffer (:name "print-circle")
+ (let* ((print-circle nil)
+ (arg (let ((val (make-list 5 'a))) (nconc val val) val))
+ (results (backtrace-tests--make-regexp
+ (backtrace-tests--result arg)))
+ (results-circle (regexp-quote (let ((print-circle t))
+ (backtrace-tests--result arg))))
+ (last-frame (backtrace-tests--make-regexp
+ (format (nth (1- backtrace-tests--line-count)
+ (backtrace-tests--backtrace-lines))
+ arg)))
+ (last-frame-circle (regexp-quote
+ (let ((print-circle t))
+ (format (nth (1- backtrace-tests--line-count)
+
(backtrace-tests--backtrace-lines))
+ arg)))))
+ (backtrace-tests--make-backtrace arg)
+ (backtrace-print)
+ (should (string-match-p results
+ (backtrace-tests--get-substring (point-min)
(point-max))))
+ ;; Go to the last frame.
+ (goto-char (point-max))
+ (forward-line -1)
+ ;; Turn on print-circle for that frame.
+ (backtrace-toggle-print-circle)
+ (should (string-match-p last-frame-circle
+ (backtrace-tests--get-substring (point)
(point-max))))
+ ;; Turn off print-circle for the frame.
+ (backtrace-toggle-print-circle)
+ (should (string-match-p last-frame
+ (backtrace-tests--get-substring (point)
(point-max))))
+ (should (string-match-p results
+ (backtrace-tests--get-substring (point-min)
(point-max))))
+ ;; Turn print-circle on for the buffer.
+ (backtrace-toggle-print-circle '(4))
+ (should (string-match-p last-frame-circle
+ (backtrace-tests--get-substring (point)
(point-max))))
+ (should (string-match-p results-circle
+ (backtrace-tests--get-substring (point-min)
(point-max))))
+ ;; Turn print-circle off.
+ (backtrace-toggle-print-circle '(4))
+ (should (string-match-p last-frame
+ (backtrace-tests--get-substring
+ (point) (+ (point) (length last-frame)))))
+ (should (string-match-p results
+ (backtrace-tests--get-substring (point-min)
(point-max)))))))
+
+(defun backtrace-tests--make-regexp (str)
+ "Make regexp from STR for `backtrace-tests--print-circle'.
+Used for results of printing circular objects without
+`print-circle' on. Look for #n in string STR where n is any
+digit and replace with #[0-9]."
+ (let ((regexp (regexp-quote str)))
+ (with-temp-buffer
+ (insert regexp)
+ (goto-char (point-min))
+ (while (re-search-forward "#[0-9]" nil t)
+ (replace-match "#[0-9]")))
+ (buffer-string)))
+
+(ert-deftest backtrace-tests--expand-ellipsis ()
+ "Backtrace buffers ellipsify large forms as buttons which expand the
ellipses."
+ ;; make a backtrace with an ellipsis
+ ;; expand the ellipsis
+ (ert-with-test-buffer (:name "variables")
+ (let* ((print-level nil)
+ (print-length nil)
+ (backtrace-line-length 300)
+ (arg (make-list 40 (make-string 10 ?a)))
+ (results (backtrace-tests--result arg)))
+ (backtrace-tests--make-backtrace arg)
+ (backtrace-print)
+
+ ;; There should be an ellipsis. Find and expand it.
+ (goto-char (point-min))
+ (search-forward "...")
+ (backward-char)
+ (push-button)
+
+ (should (string= (backtrace-tests--get-substring (point-min) (point-max))
+ results)))))
+
+(ert-deftest backtrace-tests--expand-ellipses ()
+ "Backtrace buffers ellipsify large forms and can expand the ellipses."
+ (ert-with-test-buffer (:name "variables")
+ (let* ((print-level nil)
+ (print-length nil)
+ (backtrace-line-length 300)
+ (arg (let ((outer (make-list 40 (make-string 10 ?a)))
+ (nested (make-list 40 (make-string 10 ?b))))
+ (setf (nth 39 nested) (make-list 40 (make-string 10 ?c)))
+ (setf (nth 39 outer) nested)
+ outer))
+ (results (backtrace-tests--result-with-locals arg)))
+
+ ;; Make a backtrace with local variables visible.
+ (backtrace-tests--make-backtrace arg)
+ (backtrace-print)
+ (backtrace-toggle-locals '(4))
+
+ ;; There should be two ellipses.
+ (goto-char (point-min))
+ (should (search-forward "..."))
+ (should (search-forward "..."))
+ (should-error (search-forward "..."))
+
+ ;; Expanding the last frame without argument should expand both
+ ;; ellipses, but the expansions will contain one ellipsis each.
+ (let ((buffer-len (- (point-max) (point-min))))
+ (goto-char (point-max))
+ (backtrace-backward-frame)
+ (backtrace-expand-ellipses)
+ (should (> (- (point-max) (point-min)) buffer-len))
+ (goto-char (point-min))
+ (should (search-forward "..."))
+ (should (search-forward "..."))
+ (should-error (search-forward "...")))
+
+ ;; Expanding with argument should remove all ellipses.
+ (goto-char (point-max))
+ (backtrace-backward-frame)
+ (backtrace-expand-ellipses '(4))
+ (goto-char (point-min))
+
+ (should-error (search-forward "..."))
+ (should (string= (backtrace-tests--get-substring (point-min) (point-max))
+ results)))))
+
+
+(ert-deftest backtrace-tests--to-string ()
+ "Backtraces can be produced as strings."
+ (let ((frames (ert-with-test-buffer (:name nil)
+ (backtrace-tests--make-backtrace "string")
+ backtrace-frames)))
+ (should (string= (backtrace-to-string frames)
+ (backtrace-tests--result "string")))))
+
+(defun backtrace-tests--get-substring (beg end)
+ "Return the visible text between BEG and END.
+Strip the string properties because it makes failed test results
+easier to read."
+ (substring-no-properties (filter-buffer-substring beg end)))
+
+(provide 'backtrace-tests)
+
+;;; backtrace-tests.el ends here
diff --git a/test/lisp/emacs-lisp/cl-print-tests.el
b/test/lisp/emacs-lisp/cl-print-tests.el
index 404d323..a469b55 100644
--- a/test/lisp/emacs-lisp/cl-print-tests.el
+++ b/test/lisp/emacs-lisp/cl-print-tests.el
@@ -56,19 +56,30 @@
(let ((long-list (make-list 5 'a))
(long-vec (make-vector 5 'b))
(long-struct (cl-print-tests-con))
+ (long-string (make-string 5 ?a))
(print-length 4))
(should (equal "(a a a a ...)" (cl-prin1-to-string long-list)))
(should (equal "[b b b b ...]" (cl-prin1-to-string long-vec)))
(should (equal "#s(cl-print-tests-struct :a nil :b nil :c nil :d nil ...)"
- (cl-prin1-to-string long-struct)))))
+ (cl-prin1-to-string long-struct)))
+ (should (equal "\"aaaa...\"" (cl-prin1-to-string long-string)))))
(ert-deftest cl-print-tests-4 ()
"CL printing observes `print-level'."
- (let ((deep-list '(a (b (c (d (e))))))
- (deep-struct (cl-print-tests-con))
- (print-level 4))
+ (let* ((deep-list '(a (b (c (d (e))))))
+ (buried-vector '(a (b (c (d [e])))))
+ (deep-struct (cl-print-tests-con))
+ (buried-struct `(a (b (c (d ,deep-struct)))))
+ (buried-string '(a (b (c (d #("hello" 0 5 (cl-print-test t)))))))
+ (buried-simple-string '(a (b (c (d "hello")))))
+ (print-level 4))
(setf (cl-print-tests-struct-a deep-struct) deep-list)
(should (equal "(a (b (c (d ...))))" (cl-prin1-to-string deep-list)))
+ (should (equal "(a (b (c (d ...))))" (cl-prin1-to-string buried-vector)))
+ (should (equal "(a (b (c (d ...))))" (cl-prin1-to-string buried-struct)))
+ (should (equal "(a (b (c (d ...))))" (cl-prin1-to-string buried-string)))
+ (should (equal "(a (b (c (d \"hello\"))))"
+ (cl-prin1-to-string buried-simple-string)))
(should (equal "#s(cl-print-tests-struct :a (a (b (c ...))) :b nil :c nil
:d nil :e nil)"
(cl-prin1-to-string deep-struct)))))
@@ -82,6 +93,129 @@
(should (equal "((quote a) (function b) (\\` ((\\, c) (\\,@ d))))"
(cl-prin1-to-string quoted-stuff))))))
+(ert-deftest cl-print-tests-strings ()
+ "CL printing prints strings and propertized strings."
+ (let* ((str1 "abcdefghij")
+ (str2 #("abcdefghij" 3 6 (bold t) 7 9 (italic t)))
+ (str3 #("abcdefghij" 0 10 (test t)))
+ (obj '(a b))
+ ;; Since the byte compiler reuses string literals,
+ ;; and the put-text-property call is destructive, use
+ ;; copy-sequence to make a new string.
+ (str4 (copy-sequence "abcdefghij")))
+ (put-text-property 0 5 'test obj str4)
+ (put-text-property 7 10 'test obj str4)
+
+ (should (equal "\"abcdefghij\"" (cl-prin1-to-string str1)))
+ (should (equal "#(\"abcdefghij\" 3 6 (bold t) 7 9 (italic t))"
+ (cl-prin1-to-string str2)))
+ (should (equal "#(\"abcdefghij\" 0 10 (test t))"
+ (cl-prin1-to-string str3)))
+ (let ((print-circle nil))
+ (should
+ (equal
+ "#(\"abcdefghij\" 0 5 (test (a b)) 7 10 (test (a b)))"
+ (cl-prin1-to-string str4))))
+ (let ((print-circle t))
+ (should
+ (equal
+ "#(\"abcdefghij\" 0 5 (test #1=(a b)) 7 10 (test #1#))"
+ (cl-prin1-to-string str4))))))
+
+(ert-deftest cl-print-tests-ellipsis-cons ()
+ "Ellipsis expansion works in conses."
+ (let ((print-length 4)
+ (print-level 3))
+ (cl-print-tests-check-ellipsis-expansion
+ '(0 1 2 3 4 5) "(0 1 2 3 ...)" "4 5")
+ (cl-print-tests-check-ellipsis-expansion
+ '(0 1 2 3 4 5 6 7 8 9) "(0 1 2 3 ...)" "4 5 6 7 ...")
+ (cl-print-tests-check-ellipsis-expansion
+ '(a (b (c (d (e))))) "(a (b (c ...)))" "(d (e))")
+ (cl-print-tests-check-ellipsis-expansion
+ (let ((x (make-list 6 'b)))
+ (setf (nthcdr 6 x) 'c)
+ x)
+ "(b b b b ...)" "b b . c")))
+
+(ert-deftest cl-print-tests-ellipsis-vector ()
+ "Ellipsis expansion works in vectors."
+ (let ((print-length 4)
+ (print-level 3))
+ (cl-print-tests-check-ellipsis-expansion
+ [0 1 2 3 4 5] "[0 1 2 3 ...]" "4 5")
+ (cl-print-tests-check-ellipsis-expansion
+ [0 1 2 3 4 5 6 7 8 9] "[0 1 2 3 ...]" "4 5 6 7 ...")
+ (cl-print-tests-check-ellipsis-expansion
+ [a [b [c [d [e]]]]] "[a [b [c ...]]]" "[d [e]]")))
+
+(ert-deftest cl-print-tests-ellipsis-string ()
+ "Ellipsis expansion works in strings."
+ (let ((print-length 4)
+ (print-level 3))
+ (cl-print-tests-check-ellipsis-expansion
+ "abcdefg" "\"abcd...\"" "efg")
+ (cl-print-tests-check-ellipsis-expansion
+ "abcdefghijk" "\"abcd...\"" "efgh...")
+ (cl-print-tests-check-ellipsis-expansion
+ '(1 (2 (3 #("abcde" 0 5 (test t)))))
+ "(1 (2 (3 ...)))" "#(\"abcd...\" 0 5 (test t))")
+ (cl-print-tests-check-ellipsis-expansion
+ #("abcd" 0 1 (bold t) 1 2 (invisible t) 3 4 (italic t))
+ "#(\"abcd\" 0 1 (bold t) ...)" "1 2 (invisible t) ...")))
+
+(ert-deftest cl-print-tests-ellipsis-struct ()
+ "Ellipsis expansion works in structures."
+ (let ((print-length 4)
+ (print-level 3)
+ (struct (cl-print-tests-con)))
+ (cl-print-tests-check-ellipsis-expansion
+ struct "#s(cl-print-tests-struct :a nil :b nil :c nil :d nil ...)" ":e
nil")
+ (let ((print-length 2))
+ (cl-print-tests-check-ellipsis-expansion
+ struct "#s(cl-print-tests-struct :a nil :b nil ...)" ":c nil :d nil
..."))
+ (cl-print-tests-check-ellipsis-expansion
+ `(a (b (c ,struct)))
+ "(a (b (c ...)))"
+ "#s(cl-print-tests-struct :a nil :b nil :c nil :d nil ...)")))
+
+(ert-deftest cl-print-tests-ellipsis-circular ()
+ "Ellipsis expansion works with circular objects."
+ (let ((wide-obj (list 0 1 2 3 4))
+ (deep-obj `(0 (1 (2 (3 (4))))))
+ (print-length 4)
+ (print-level 3))
+ (setf (nth 4 wide-obj) wide-obj)
+ (setf (car (cadadr (cadadr deep-obj))) deep-obj)
+ (let ((print-circle nil))
+ (cl-print-tests-check-ellipsis-expansion-rx
+ wide-obj (regexp-quote "(0 1 2 3 ...)") "\\`#[0-9]\\'")
+ (cl-print-tests-check-ellipsis-expansion-rx
+ deep-obj (regexp-quote "(0 (1 (2 ...)))") "\\`(3 (#[0-9]))\\'"))
+ (let ((print-circle t))
+ (cl-print-tests-check-ellipsis-expansion
+ wide-obj "#1=(0 1 2 3 ...)" "#1#")
+ (cl-print-tests-check-ellipsis-expansion
+ deep-obj "#1=(0 (1 (2 ...)))" "(3 (#1#))"))))
+
+(defun cl-print-tests-check-ellipsis-expansion (obj expected expanded)
+ (let* ((result (cl-prin1-to-string obj))
+ (pos (next-single-property-change 0 'cl-print-ellipsis result))
+ value)
+ (should pos)
+ (setq value (get-text-property pos 'cl-print-ellipsis result))
+ (should (equal expected result))
+ (should (equal expanded (with-output-to-string (cl-print-expand-ellipsis
+ value nil))))))
+
+(defun cl-print-tests-check-ellipsis-expansion-rx (obj expected expanded)
+ (let* ((result (cl-prin1-to-string obj))
+ (pos (next-single-property-change 0 'cl-print-ellipsis result))
+ (value (get-text-property pos 'cl-print-ellipsis result)))
+ (should (string-match expected result))
+ (should (string-match expanded (with-output-to-string
+ (cl-print-expand-ellipsis value nil))))))
+
(ert-deftest cl-print-circle ()
(let ((x '(#1=(a . #1#) #1#)))
(let ((print-circle nil))
@@ -99,5 +233,41 @@
(let ((print-circle t))
(should (equal "(0 . #1=(0 . #1#))" (cl-prin1-to-string x))))))
+(ert-deftest cl-print-tests-print-to-string-with-limit ()
+ (let* ((thing10 (make-list 10 'a))
+ (thing100 (make-list 100 'a))
+ (thing10x10 (make-list 10 thing10))
+ (nested-thing (let ((val 'a))
+ (dotimes (_i 20)
+ (setq val (list val)))
+ val))
+ ;; Make a consistent environment for this test.
+ (print-circle nil)
+ (print-level nil)
+ (print-length nil))
+
+ ;; Print something that fits in the space given.
+ (should (string= (cl-prin1-to-string thing10)
+ (cl-print-to-string-with-limit #'cl-prin1 thing10 100)))
+
+ ;; Print something which needs to be abbreviated and which can be.
+ (should (< (length (cl-print-to-string-with-limit #'cl-prin1 thing100 100))
+ 100
+ (length (cl-prin1-to-string thing100))))
+
+ ;; Print something resistant to easy abbreviation.
+ (should (string= (cl-prin1-to-string thing10x10)
+ (cl-print-to-string-with-limit #'cl-prin1 thing10x10
100)))
+
+ ;; Print something which should be abbreviated even if the limit is large.
+ (should (< (length (cl-print-to-string-with-limit #'cl-prin1 nested-thing
1000))
+ (length (cl-prin1-to-string nested-thing))))
+
+ ;; Print with no limits.
+ (dolist (thing (list thing10 thing100 thing10x10 nested-thing))
+ (let ((rep (cl-prin1-to-string thing)))
+ (should (string= rep (cl-print-to-string-with-limit #'cl-prin1 thing
0)))
+ (should (string= rep (cl-print-to-string-with-limit #'cl-prin1 thing
nil)))))))
+
;;; cl-print-tests.el ends here.
diff --git a/test/lisp/emacs-lisp/edebug-resources/edebug-test-code.el
b/test/lisp/emacs-lisp/edebug-resources/edebug-test-code.el
index e86c2f1..97dead0 100644
--- a/test/lisp/emacs-lisp/edebug-resources/edebug-test-code.el
+++ b/test/lisp/emacs-lisp/edebug-resources/edebug-test-code.el
@@ -41,7 +41,7 @@
(defun edebug-test-code-range (num)
!start!(let ((index 0)
(result nil))
- (while (< index num)!test!
+ (while !lt!(< index num)!test!
(push index result)!loop!
(cl-incf index))!end-loop!
(nreverse result)))
@@ -130,5 +130,12 @@
(let ((two 2) (three 3))
(cl-destructuring-bind (x . y) (cons two three) (+ x!x! y!y!))))
+(defun edebug-test-code-use-cl-macrolet (x)
+ (cl-macrolet ((wrap (func &rest args)
+ `(format "The result of applying %s to %s is %S"
+ ',func!func! ',args
+ ,(cons func args))))
+ (wrap + 1 x)))
+
(provide 'edebug-test-code)
;;; edebug-test-code.el ends here
diff --git a/test/lisp/emacs-lisp/edebug-tests.el
b/test/lisp/emacs-lisp/edebug-tests.el
index 85f6bd4..7880aaf 100644
--- a/test/lisp/emacs-lisp/edebug-tests.el
+++ b/test/lisp/emacs-lisp/edebug-tests.el
@@ -432,9 +432,11 @@ test and possibly others should be updated."
(verify-keybinding "P" 'edebug-view-outside) ;; same as v
(verify-keybinding "W" 'edebug-toggle-save-windows)
(verify-keybinding "?" 'edebug-help)
- (verify-keybinding "d" 'edebug-backtrace)
+ (verify-keybinding "d" 'edebug-pop-to-backtrace)
(verify-keybinding "-" 'negative-argument)
- (verify-keybinding "=" 'edebug-temp-display-freq-count)))
+ (verify-keybinding "=" 'edebug-temp-display-freq-count)
+ (should (eq (lookup-key backtrace-mode-map "n") 'backtrace-forward-frame))
+ (should (eq (lookup-key backtrace-mode-map "s") 'backtrace-goto-source))))
(ert-deftest edebug-tests-stop-point-at-start-of-first-instrumented-function ()
"Edebug stops at the beginning of an instrumented function."
@@ -913,5 +915,28 @@ test and possibly others should be updated."
"g"
(should (equal address@hidden 5)))))
+(ert-deftest edebug-tests-cl-macrolet ()
+ "Edebug can instrument `cl-macrolet' expressions. (Bug#29919)"
+ (edebug-tests-with-normal-env
+ (edebug-tests-setup-@ "use-cl-macrolet" '(10) t)
+ (edebug-tests-run-kbd-macro
+ "@ SPC SPC"
+ (edebug-tests-should-be-at "use-cl-macrolet" "func")
+ (edebug-tests-should-match-result-in-messages "+")
+ "g"
+ (should (equal address@hidden "The result of applying + to (1 x) is
11")))))
+
+(ert-deftest edebug-tests-backtrace-goto-source ()
+ "Edebug can jump to instrumented source from its *Edebug-Backtrace* buffer."
+ (edebug-tests-with-normal-env
+ (edebug-tests-setup-@ "range" '(2) t)
+ (edebug-tests-run-kbd-macro
+ "@ SPC SPC"
+ (edebug-tests-should-be-at "range" "lt")
+ "dns" ; Pop to backtrace, next frame, goto source.
+ (edebug-tests-should-be-at "range" "start")
+ "g"
+ (should (equal address@hidden '(0 1))))))
+
(provide 'edebug-tests)
;;; edebug-tests.el ends here
diff --git a/test/lisp/emacs-lisp/ert-tests.el
b/test/lisp/emacs-lisp/ert-tests.el
index cb957bd..1fe5b79 100644
--- a/test/lisp/emacs-lisp/ert-tests.el
+++ b/test/lisp/emacs-lisp/ert-tests.el
@@ -376,7 +376,7 @@ This macro is used to test if macroexpansion in `should'
works."
(test (make-ert-test :body test-body))
(result (ert-run-test test)))
(should (ert-test-failed-p result))
- (should (eq (nth 1 (car (ert-test-failed-backtrace result)))
+ (should (eq (backtrace-frame-fun (car (ert-test-failed-backtrace result)))
'signal))))
(ert-deftest ert-test-messages ()
diff --git a/test/lisp/emacs-lisp/lisp-mode-tests.el
b/test/lisp/emacs-lisp/lisp-mode-tests.el
index 8598d41..30f606d 100644
--- a/test/lisp/emacs-lisp/lisp-mode-tests.el
+++ b/test/lisp/emacs-lisp/lisp-mode-tests.el
@@ -113,6 +113,29 @@ noindent\" 3
;; we're indenting ends on the previous line.
(should (equal (buffer-string) original)))))
+(ert-deftest indent-sexp-go ()
+ "Make sure `indent-sexp' doesn't stop after #s."
+ ;; See https://debbugs.gnu.org/cgi/bugreport.cgi?bug=31984.
+ (with-temp-buffer
+ (emacs-lisp-mode)
+ (insert "#s(foo\nbar)\n")
+ (goto-char (point-min))
+ (indent-sexp)
+ (should (equal (buffer-string) "\
+#s(foo
+ bar)\n"))))
+
+(ert-deftest indent-sexp-cant-go ()
+ "`indent-sexp' shouldn't error before a sexp."
+ ;; See https://debbugs.gnu.org/cgi/bugreport.cgi?bug=31984#32.
+ (with-temp-buffer
+ (emacs-lisp-mode)
+ (insert "(())")
+ (goto-char (1+ (point-min)))
+ ;; Paredit calls `indent-sexp' from this position.
+ (indent-sexp)
+ (should (equal (buffer-string) "(())"))))
+
(ert-deftest lisp-indent-region ()
"Test basics of `lisp-indent-region'."
(with-temp-buffer
diff --git a/test/lisp/emacs-lisp/package-tests.el
b/test/lisp/emacs-lisp/package-tests.el
index db6d103..f08bc92 100644
--- a/test/lisp/emacs-lisp/package-tests.el
+++ b/test/lisp/emacs-lisp/package-tests.el
@@ -112,7 +112,7 @@
upload-base)
&rest body)
"Set up temporary locations and variables for testing."
- (declare (indent 1))
+ (declare (indent 1) (debug (([&rest form]) body)))
`(let* ((package-test-user-dir (make-temp-file "pkg-test-user-dir-" t))
(process-environment (cons (format "HOME=%s" package-test-user-dir)
process-environment))
@@ -158,6 +158,7 @@
(defmacro with-fake-help-buffer (&rest body)
"Execute BODY in a temp buffer which is treated as the \"*Help*\" buffer."
+ (declare (debug body))
`(with-temp-buffer
(help-mode)
;; Trick `help-buffer' into using the temp buffer.
@@ -467,15 +468,23 @@ Must called from within a `tar-mode' buffer."
(ert-deftest package-test-signed ()
"Test verifying package signature."
- (skip-unless (ignore-errors
- (let ((homedir (make-temp-file "package-test" t)))
- (unwind-protect
- (let ((process-environment
- (cons (format "HOME=%s" homedir)
- process-environment)))
- (epg-check-configuration
- (epg-find-configuration 'OpenPGP)))
- (delete-directory homedir t)))))
+ (skip-unless (let ((homedir (make-temp-file "package-test" t)))
+ (unwind-protect
+ (let ((process-environment
+ (cons (concat "HOME=" homedir)
+ process-environment)))
+ (epg-find-configuration
+ 'OpenPGP nil
+ ;; By default we require gpg2 2.1+ due to some
+ ;; practical problems with pinentry. But this
+ ;; test works fine with 2.0 as well.
+ (let ((prog-alist (copy-tree
epg-config--program-alist)))
+ (setf (alist-get "gpg2"
+ (alist-get 'OpenPGP prog-alist)
+ nil nil #'equal)
+ "2.0")
+ prog-alist)))
+ (delete-directory homedir t))))
(let* ((keyring (expand-file-name "key.pub" package-test-data-dir))
(package-test-data-dir
(expand-file-name "package-resources/signed" package-test-file-dir)))
@@ -506,7 +515,7 @@ Must called from within a `tar-mode' buffer."
(with-fake-help-buffer
(describe-package 'signed-good)
(goto-char (point-min))
- (should (re-search-forward "signed-good is an? \\(\\S-+\\) package."
nil t))
+ (should (re-search-forward "Package signed-good is \\(\\S-+\\)\\." nil
t))
(should (string-equal (match-string-no-properties 1) "installed"))
(should (re-search-forward
"Status: Installed in ['`‘]signed-good-1.0/['’]."
diff --git a/test/lisp/emacs-lisp/subr-x-tests.el
b/test/lisp/emacs-lisp/subr-x-tests.el
index f7f0ef3..81467ba 100644
--- a/test/lisp/emacs-lisp/subr-x-tests.el
+++ b/test/lisp/emacs-lisp/subr-x-tests.el
@@ -532,6 +532,53 @@
(format "abs sum is: %s"))
"abs sum is: 15")))
+
+;; Substring tests
+
+(ert-deftest subr-x-test-string-trim-left ()
+ "Test `string-trim-left' behavior."
+ (should (equal (string-trim-left "") ""))
+ (should (equal (string-trim-left " \t\n\r") ""))
+ (should (equal (string-trim-left " \t\n\ra") "a"))
+ (should (equal (string-trim-left "a \t\n\r") "a \t\n\r"))
+ (should (equal (string-trim-left "" "") ""))
+ (should (equal (string-trim-left "a" "") "a"))
+ (should (equal (string-trim-left "aa" "a*") ""))
+ (should (equal (string-trim-left "ba" "a*") "ba"))
+ (should (equal (string-trim-left "aa" "a*?") "aa"))
+ (should (equal (string-trim-left "aa" "a+?") "a")))
+
+(ert-deftest subr-x-test-string-trim-right ()
+ "Test `string-trim-right' behavior."
+ (should (equal (string-trim-right "") ""))
+ (should (equal (string-trim-right " \t\n\r") ""))
+ (should (equal (string-trim-right " \t\n\ra") " \t\n\ra"))
+ (should (equal (string-trim-right "a \t\n\r") "a"))
+ (should (equal (string-trim-right "" "") ""))
+ (should (equal (string-trim-right "a" "") "a"))
+ (should (equal (string-trim-right "aa" "a*") ""))
+ (should (equal (string-trim-right "ab" "a*") "ab"))
+ (should (equal (string-trim-right "aa" "a*?") "")))
+
+(ert-deftest subr-x-test-string-remove-prefix ()
+ "Test `string-remove-prefix' behavior."
+ (should (equal (string-remove-prefix "" "") ""))
+ (should (equal (string-remove-prefix "" "a") "a"))
+ (should (equal (string-remove-prefix "a" "") ""))
+ (should (equal (string-remove-prefix "a" "b") "b"))
+ (should (equal (string-remove-prefix "a" "a") ""))
+ (should (equal (string-remove-prefix "a" "aa") "a"))
+ (should (equal (string-remove-prefix "a" "ab") "b")))
+
+(ert-deftest subr-x-test-string-remove-suffix ()
+ "Test `string-remove-suffix' behavior."
+ (should (equal (string-remove-suffix "" "") ""))
+ (should (equal (string-remove-suffix "" "a") "a"))
+ (should (equal (string-remove-suffix "a" "") ""))
+ (should (equal (string-remove-suffix "a" "b") "b"))
+ (should (equal (string-remove-suffix "a" "a") ""))
+ (should (equal (string-remove-suffix "a" "aa") "a"))
+ (should (equal (string-remove-suffix "a" "ba") "b")))
(provide 'subr-x-tests)
;;; subr-x-tests.el ends here
diff --git a/test/lisp/epg-tests.el b/test/lisp/epg-tests.el
index 7efe04b..c1e98a6 100644
--- a/test/lisp/epg-tests.el
+++ b/test/lisp/epg-tests.el
@@ -32,17 +32,26 @@
(defconst epg-tests--config-program-alist
;; The default `epg-config--program-alist' requires gpg2 2.1 or
- ;; greater due to some practical problems with pinentry. But the
- ;; tests here all work fine with 2.0 as well.
- (let ((prog-alist (copy-sequence epg-config--program-alist)))
+ ;; greater due to some practical problems with pinentry. But most
+ ;; tests here work fine with 2.0 as well.
+ (let ((prog-alist (copy-tree epg-config--program-alist)))
(setf (alist-get "gpg2"
(alist-get 'OpenPGP prog-alist)
nil nil #'equal)
"2.0")
prog-alist))
-(defun epg-tests-find-usable-gpg-configuration (&optional _require-passphrase)
- (epg-find-configuration 'OpenPGP 'no-cache epg-tests--config-program-alist))
+(defun epg-tests-find-usable-gpg-configuration
+ (&optional require-passphrase require-public-key)
+ ;; Clear config cache because we may be using a different
+ ;; program-alist. We do want to update the cache, so that
+ ;; `epg-make-context' can use our result.
+ (setq epg--configurations nil)
+ (epg-find-configuration 'OpenPGP nil
+ ;; The symmetric operations fail on Hydra
+ ;; with gpg 2.0.
+ (if (or (not require-passphrase) require-public-key)
+ epg-tests--config-program-alist)))
(defun epg-tests-passphrase-callback (_c _k _d)
;; Need to create a copy here, since the string will be wiped out
@@ -62,12 +71,14 @@
(format "GNUPGHOME=%s" epg-tests-home-directory))
process-environment)))
(unwind-protect
- (let ((context (epg-make-context 'OpenPGP)))
+ ;; GNUPGHOME is needed to find a usable gpg, so we can't
+ ;; check whether to skip any earlier (Bug#23561).
+ (let ((epg-config (or (epg-tests-find-usable-gpg-configuration
+ ,require-passphrase ,require-public-key)
+ (ert-skip "No usable gpg config")))
+ (context (epg-make-context 'OpenPGP)))
(setf (epg-context-program context)
- (alist-get 'program
- (epg-tests-find-usable-gpg-configuration
- ,(if require-passphrase
- `'require-passphrase))))
+ (alist-get 'program epg-config))
(setf (epg-context-home-directory context)
epg-tests-home-directory)
,(if require-passphrase
@@ -96,7 +107,6 @@
(delete-directory epg-tests-home-directory t)))))
(ert-deftest epg-decrypt-1 ()
- (skip-unless (epg-tests-find-usable-gpg-configuration 'require-passphrase))
(with-epg-tests (:require-passphrase t)
(should (equal "test"
(epg-decrypt-string epg-tests-context "\
@@ -108,14 +118,12 @@
jA0EAwMCE19JBLTvvmhgyRrGGglRbnKkK9PJG8fDwO5ccjysrR7IcdNcnA==
-----END PGP MESSAGE-----")))))
(ert-deftest epg-roundtrip-1 ()
- (skip-unless (epg-tests-find-usable-gpg-configuration 'require-passphrase))
(with-epg-tests (:require-passphrase t)
(let ((cipher (epg-encrypt-string epg-tests-context "symmetric" nil)))
(should (equal "symmetric"
(epg-decrypt-string epg-tests-context cipher))))))
(ert-deftest epg-roundtrip-2 ()
- (skip-unless (epg-tests-find-usable-gpg-configuration 'require-passphrase))
(with-epg-tests (:require-passphrase t
:require-public-key t
:require-secret-key t)
@@ -126,7 +134,6 @@ jA0EAwMCE19JBLTvvmhgyRrGGglRbnKkK9PJG8fDwO5ccjysrR7IcdNcnA==
(epg-decrypt-string epg-tests-context cipher))))))
(ert-deftest epg-sign-verify-1 ()
- (skip-unless (epg-tests-find-usable-gpg-configuration 'require-passphrase))
(with-epg-tests (:require-passphrase t
:require-public-key t
:require-secret-key t)
@@ -140,7 +147,6 @@ jA0EAwMCE19JBLTvvmhgyRrGGglRbnKkK9PJG8fDwO5ccjysrR7IcdNcnA==
(should (eq 'good (epg-signature-status (car verify-result)))))))
(ert-deftest epg-sign-verify-2 ()
- (skip-unless (epg-tests-find-usable-gpg-configuration 'require-passphrase))
(with-epg-tests (:require-passphrase t
:require-public-key t
:require-secret-key t)
@@ -156,7 +162,6 @@ jA0EAwMCE19JBLTvvmhgyRrGGglRbnKkK9PJG8fDwO5ccjysrR7IcdNcnA==
(should (eq 'good (epg-signature-status (car verify-result)))))))
(ert-deftest epg-sign-verify-3 ()
- (skip-unless (epg-tests-find-usable-gpg-configuration 'require-passphrase))
(with-epg-tests (:require-passphrase t
:require-public-key t
:require-secret-key t)
@@ -171,7 +176,6 @@ jA0EAwMCE19JBLTvvmhgyRrGGglRbnKkK9PJG8fDwO5ccjysrR7IcdNcnA==
(should (eq 'good (epg-signature-status (car verify-result)))))))
(ert-deftest epg-import-1 ()
- (skip-unless (epg-tests-find-usable-gpg-configuration 'require-passphrase))
(with-epg-tests (:require-passphrase nil)
(should (= 0 (length (epg-list-keys epg-tests-context))))
(should (= 0 (length (epg-list-keys epg-tests-context nil t)))))
diff --git a/test/lisp/filenotify-tests.el b/test/lisp/filenotify-tests.el
index 56403f4..612ea8c 100644
--- a/test/lisp/filenotify-tests.el
+++ b/test/lisp/filenotify-tests.el
@@ -891,9 +891,9 @@ delivered."
;; Modify file. We wait for two seconds, in order to
;; have another timestamp. One second seems to be too
- ;; short.
+ ;; short. And Cygwin sporadically requires more than two.
(ert-with-message-capture captured-messages
- (sleep-for 2)
+ (sleep-for (if (eq system-type 'cygwin) 3 2))
(write-region
"foo bla" nil file-notify--test-tmpfile nil 'no-message)
diff --git a/test/lisp/net/secrets-tests.el b/test/lisp/net/secrets-tests.el
index 9aa79da..de3ce73 100644
--- a/test/lisp/net/secrets-tests.el
+++ b/test/lisp/net/secrets-tests.el
@@ -92,7 +92,8 @@
(should (secrets-open-session))
;; There must be at least the collections "Login" and "session".
- (should (member "Login" (secrets-list-collections)))
+ (should (or (member "Login" (secrets-list-collections))
+ (member "login" (secrets-list-collections))))
(should (member "session" (secrets-list-collections)))
;; Create a random collection. This asks for a password
@@ -160,7 +161,8 @@
;; There shall be no items in the "session" collection.
(should-not (secrets-list-items "session"))
;; There shall be items in the "Login" collection.
- (should (secrets-list-items "Login"))
+ (should (or (secrets-list-items "Login")
+ (secrets-list-items "login")))
;; Create a new item.
(should (setq item-path (secrets-create-item "session" "foo" "secret")))
diff --git a/test/lisp/net/tramp-archive-tests.el
b/test/lisp/net/tramp-archive-tests.el
index 0a8716b..e759786 100644
--- a/test/lisp/net/tramp-archive-tests.el
+++ b/test/lisp/net/tramp-archive-tests.el
@@ -311,6 +311,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(ert-deftest tramp-archive-test07-file-exists-p ()
"Check `file-exist-p', `write-region' and `delete-file'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
(unwind-protect
@@ -333,6 +334,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(ert-deftest tramp-archive-test08-file-local-copy ()
"Check `file-local-copy'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
(let (tmp-name)
@@ -359,6 +361,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(ert-deftest tramp-archive-test09-insert-file-contents ()
"Check `insert-file-contents'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
(let ((tmp-name (expand-file-name "bar/bar" tramp-archive-test-archive)))
@@ -385,6 +388,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(ert-deftest tramp-archive-test11-copy-file ()
"Check `copy-file'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
;; Copy simple file.
@@ -450,6 +454,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(ert-deftest tramp-archive-test15-copy-directory ()
"Check `copy-directory'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
(let* ((tmp-name1 (expand-file-name "bar" tramp-archive-test-archive))
@@ -504,6 +509,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(ert-deftest tramp-archive-test16-directory-files ()
"Check `directory-files'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
(let ((tmp-name tramp-archive-test-archive)
@@ -527,6 +533,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(ert-deftest tramp-archive-test17-insert-directory ()
"Check `insert-directory'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
(let (;; We test for the summary line. Keyword "total" could be localized.
@@ -569,6 +576,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(ert-deftest tramp-archive-test18-file-attributes ()
"Check `file-attributes'.
This tests also `file-readable-p' and `file-regular-p'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
(let ((tmp-name1 (expand-file-name "foo.txt" tramp-archive-test-archive))
@@ -619,6 +627,7 @@ This tests also `file-readable-p' and `file-regular-p'."
(ert-deftest tramp-archive-test19-directory-files-and-attributes ()
"Check `directory-files-and-attributes'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
(let ((tmp-name (expand-file-name "bar" tramp-archive-test-archive))
@@ -644,6 +653,7 @@ This tests also `file-readable-p' and `file-regular-p'."
(ert-deftest tramp-archive-test20-file-modes ()
"Check `file-modes'.
This tests also `file-executable-p', `file-writable-p' and `set-file-modes'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
(let ((tmp-name1 (expand-file-name "foo.txt" tramp-archive-test-archive))
@@ -673,6 +683,7 @@ This tests also `file-executable-p', `file-writable-p' and
`set-file-modes'."
(ert-deftest tramp-archive-test21-file-links ()
"Check `file-symlink-p' and `file-truename'"
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
;; We must use `file-truename' for the file archive, because it
@@ -711,6 +722,7 @@ This tests also `file-executable-p', `file-writable-p' and
`set-file-modes'."
(ert-deftest tramp-archive-test26-file-name-completion ()
"Check `file-name-completion' and `file-name-all-completions'."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
(let ((tmp-name tramp-archive-test-archive))
@@ -802,8 +814,9 @@ This tests also `file-executable-p', `file-writable-p' and
`set-file-modes'."
(zerop (nth 1 fsi))
(zerop (nth 2 fsi))))))
-(ert-deftest tramp-archive-test43-auto-load ()
+(ert-deftest tramp-archive-test44-auto-load ()
"Check that `tramp-archive' autoloads properly."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
;; Autoloading tramp-archive works since Emacs 27.1.
(skip-unless (tramp-archive--test-emacs27-p))
@@ -832,8 +845,9 @@ This tests also `file-executable-p', `file-writable-p' and
`set-file-modes'."
(mapconcat 'shell-quote-argument load-path " -L ")
(shell-quote-argument (format code file)))))))))
-(ert-deftest tramp-archive-test43-delay-load ()
+(ert-deftest tramp-archive-test44-delay-load ()
"Check that `tramp-archive' is loaded lazily, only when needed."
+ :tags '(:expensive-test)
(skip-unless tramp-archive-enabled)
;; Autoloading tramp-archive works since Emacs 27.1.
(skip-unless (tramp-archive--test-emacs27-p))
diff --git a/test/lisp/net/tramp-tests.el b/test/lisp/net/tramp-tests.el
index 5c5eff8..293a005 100644
--- a/test/lisp/net/tramp-tests.el
+++ b/test/lisp/net/tramp-tests.el
@@ -2182,7 +2182,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(unwind-protect
;; FIXME: This fails on my QNAP server, see
;; /share/Web/owncloud/data/owncloud.log
- (unless (tramp--test-owncloud-p)
+ (unless (tramp--test-nextcloud-p)
(write-region "foo" nil source)
(should (file-exists-p source))
(make-directory target)
@@ -2205,7 +2205,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(unwind-protect
;; FIXME: This fails on my QNAP server, see
;; /share/Web/owncloud/data/owncloud.log
- (unless (and (tramp--test-owncloud-p)
+ (unless (and (tramp--test-nextcloud-p)
(or (not (file-remote-p source))
(not (file-remote-p target))))
(make-directory source)
@@ -2231,7 +2231,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
;; FIXME: This fails on my QNAP server, see
;; /share/Web/owncloud/data/owncloud.log
(unless
- (and (tramp--test-owncloud-p) (not (file-remote-p source)))
+ (and (tramp--test-nextcloud-p) (not (file-remote-p source)))
(make-directory source)
(should (file-directory-p source))
(write-region "foo" nil (expand-file-name "foo" source))
@@ -2320,7 +2320,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(unwind-protect
;; FIXME: This fails on my QNAP server, see
;; /share/Web/owncloud/data/owncloud.log
- (unless (tramp--test-owncloud-p)
+ (unless (tramp--test-nextcloud-p)
(make-directory source)
(should (file-directory-p source))
(write-region "foo" nil (expand-file-name "foo" source))
@@ -2344,7 +2344,7 @@ This checks also `file-name-as-directory',
`file-name-directory',
(unwind-protect
;; FIXME: This fails on my QNAP server, see
;; /share/Web/owncloud/data/owncloud.log
- (unless (tramp--test-owncloud-p)
+ (unless (tramp--test-nextcloud-p)
(make-directory source)
(should (file-directory-p source))
(write-region "foo" nil (expand-file-name "foo" source))
@@ -4427,10 +4427,10 @@ This does not support external Emacs calls."
(string-equal
"mock" (file-remote-p tramp-test-temporary-file-directory 'method)))
-(defun tramp--test-owncloud-p ()
- "Check, whether the owncloud method is used."
+(defun tramp--test-nextcloud-p ()
+ "Check, whether the nextcloud method is used."
(string-equal
- "owncloud" (file-remote-p tramp-test-temporary-file-directory 'method)))
+ "nextcloud" (file-remote-p tramp-test-temporary-file-directory 'method)))
(defun tramp--test-rsync-p ()
"Check, whether the rsync method is used.
@@ -5056,6 +5056,8 @@ process sentinels. They shall not disturb each other."
;; This test is inspired by Bug#29163.
(ert-deftest tramp-test43-auto-load ()
"Check that Tramp autoloads properly."
+ (skip-unless (tramp--test-enabled))
+
(let ((default-directory (expand-file-name temporary-file-directory))
(code
(format
@@ -5166,42 +5168,52 @@ Since it unloads Tramp, it shall be the last test to
run."
;; cannot test older Emacsen, therefore.
(skip-unless (tramp--test-emacs26-p))
- (when (featurep 'tramp)
- (unload-feature 'tramp 'force)
- ;; No Tramp feature must be left.
- (should-not (featurep 'tramp))
- (should-not (all-completions "tramp" (delq 'tramp-tests features)))
- ;; `file-name-handler-alist' must be clean.
- (should-not (all-completions "tramp" (mapcar 'cdr
file-name-handler-alist)))
- ;; There shouldn't be left a bound symbol, except buffer-local
- ;; variables, and autoload functions. We do not regard our test
- ;; symbols, and the Tramp unload hooks.
- (mapatoms
- (lambda (x)
- (and (or (and (boundp x) (null (local-variable-if-set-p x)))
- (and (functionp x) (null (autoloadp (symbol-function x)))))
- (string-match "^tramp" (symbol-name x))
- (not (string-match "^tramp--?test" (symbol-name x)))
- (not (string-match "unload-hook$" (symbol-name x)))
- (ert-fail (format "`%s' still bound" x)))))
- ;; The defstruct `tramp-file-name' and all its internal functions
- ;; shall be purged.
- (should-not (cl--find-class 'tramp-file-name))
- (mapatoms
- (lambda (x)
- (and (functionp x)
- (string-match "tramp-file-name" (symbol-name x))
- (ert-fail (format "Structure function `%s' still exists" x)))))
- ;; There shouldn't be left a hook function containing a Tramp
- ;; function. We do not regard the Tramp unload hooks.
- (mapatoms
- (lambda (x)
- (and (boundp x)
- (string-match "-\\(hook\\|function\\)s?$" (symbol-name x))
- (not (string-match "unload-hook$" (symbol-name x)))
- (consp (symbol-value x))
- (ignore-errors (all-completions "tramp" (symbol-value x)))
- (ert-fail (format "Hook `%s' still contains Tramp function" x)))))))
+ ;; We have autoloaded objects from tramp.el and tramp-archive.el.
+ ;; In order to remove them, we first need to load both packages.
+ (require 'tramp)
+ (require 'tramp-archive)
+ (should (featurep 'tramp))
+ (should (featurep 'tramp-archive))
+ ;; This unloads also tramp-archive.el and tramp-theme.el if needed.
+ (unload-feature 'tramp 'force)
+ ;; No Tramp feature must be left.
+ (should-not (featurep 'tramp))
+ (should-not (featurep 'tramp-archive))
+ (should-not (featurep 'tramp-theme))
+ (should-not
+ (all-completions
+ "tramp" (delq 'tramp-tests (delq 'tramp-archive-tests features))))
+ ;; `file-name-handler-alist' must be clean.
+ (should-not (all-completions "tramp" (mapcar 'cdr file-name-handler-alist)))
+ ;; There shouldn't be left a bound symbol, except buffer-local
+ ;; variables, and autoload functions. We do not regard our test
+ ;; symbols, and the Tramp unload hooks.
+ (mapatoms
+ (lambda (x)
+ (and (or (and (boundp x) (null (local-variable-if-set-p x)))
+ (and (functionp x) (null (autoloadp (symbol-function x)))))
+ (string-match "^tramp" (symbol-name x))
+ (not (string-match "^tramp\\(-archive\\)?--?test" (symbol-name x)))
+ (not (string-match "unload-hook$" (symbol-name x)))
+ (ert-fail (format "`%s' still bound" x)))))
+ ;; The defstruct `tramp-file-name' and all its internal functions
+ ;; shall be purged.
+ (should-not (cl--find-class 'tramp-file-name))
+ (mapatoms
+ (lambda (x)
+ (and (functionp x)
+ (string-match "tramp-file-name" (symbol-name x))
+ (ert-fail (format "Structure function `%s' still exists" x)))))
+ ;; There shouldn't be left a hook function containing a Tramp
+ ;; function. We do not regard the Tramp unload hooks.
+ (mapatoms
+ (lambda (x)
+ (and (boundp x)
+ (string-match "-\\(hook\\|function\\)s?$" (symbol-name x))
+ (not (string-match "unload-hook$" (symbol-name x)))
+ (consp (symbol-value x))
+ (ignore-errors (all-completions "tramp" (symbol-value x)))
+ (ert-fail (format "Hook `%s' still contains Tramp function" x))))))
(defun tramp-test-all (&optional interactive)
"Run all tests for \\[tramp]."
@@ -5222,7 +5234,7 @@ Since it unloads Tramp, it shall be the last test to run."
;; * Fix `tramp-test05-expand-file-name-relative' in `expand-file-name'.
;; * Fix `tramp-test06-directory-file-name' for `ftp'.
;; * Investigate, why `tramp-test11-copy-file' and `tramp-test12-rename-file'
-;; do not work properly for `owncloud'.
+;; do not work properly for `nextcloud'.
;; * Fix `tramp-test29-start-file-process' on MS Windows (`process-send-eof'?).
;; * Fix `tramp-test30-interrupt-process', timeout doesn't work reliably.
;; * Fix Bug#16928 in `tramp-test42-asynchronous-requests'.
diff --git a/test/lisp/progmodes/compile-tests.el
b/test/lisp/progmodes/compile-tests.el
index a106030..4e2dc86 100644
--- a/test/lisp/progmodes/compile-tests.el
+++ b/test/lisp/progmodes/compile-tests.el
@@ -343,6 +343,29 @@ meaning a range of columns starting on LINE and ending on
END-LINE, if that matched. TYPE can be left out, in which case
any message type is accepted.")
+(defconst compile-tests--grep-regexp-testcases
+ ;; Bug#32051.
+ '(("c:/Users/my.name/src/project\\src\\kbhit.hpp\0\ 29:#include <termios.h>"
+ 1 nil 29 "c:/Users/my.name/src/project\\src\\kbhit.hpp")
+ ("d:/gnu/emacs/branch/src/callproc.c\0\ 214:#ifdef DOS_NT"
+ 1 nil 214 "d:/gnu/emacs/branch/src/callproc.c")
+ ("/gnu/emacs/branch/src/callproc.c\0\ 214:#ifdef DOS_NT"
+ 1 nil 214 "/gnu/emacs/branch/src/callproc.c"))
+ "List of tests for `grep-regexp-list'.
+The format is the same as `compile-tests--test-regexps-data', but
+the match is expected to be the same when NUL bytes are replaced
+with colon.")
+
+(defconst compile-tests--grep-regexp-tricky-testcases
+ ;; Bug#7378.
+
'(("./x11-libs---nx/3.4.0:0:C.30253.1289557929.792611.C/nx-3.4.0.exheres-0\0\
42:some text"
+ 1 nil 42
"./x11-libs---nx/3.4.0:0:C.30253.1289557929.792611.C/nx-3.4.0.exheres-0")
+ ("2011-08-31_11:57:03_1\0\ 7:Date: Wed, 31 Aug 2011 11:57:03 +0000"
+ 1 nil 7 "2011-08-31_11:57:03_1"))
+ "List of tricky tests for `grep-regexp-list'.
+Same as `compile-tests--grep-regexp-testcases', but these cases
+can only work with the NUL byte to disambiguate colons.")
+
(defun compile--test-error-line (test)
(erase-buffer)
(setq compilation-locs (make-hash-table))
@@ -370,7 +393,8 @@ any message type is accepted.")
(should (equal (car (nth 2 (compilation--loc->file-struct loc)))
(or end-line line)))
(when type
- (should (equal type (compilation--message->type msg)))))))
+ (should (equal type (compilation--message->type msg)))))
+ msg))
(ert-deftest compile-test-error-regexps ()
"Test the `compilation-error-regexp-alist' regexps.
@@ -379,4 +403,24 @@ The test data is in `compile-tests--test-regexps-data'."
(font-lock-mode -1)
(mapc #'compile--test-error-line compile-tests--test-regexps-data)))
+(ert-deftest compile-test-grep-regexps ()
+ "Test the `grep-regexp-alist' regexps.
+The test data is in `compile-tests--grep-regexp-testcases'."
+ (with-temp-buffer
+ (grep-mode)
+ (setq buffer-read-only nil)
+ (font-lock-mode -1)
+ (dolist (testcase compile-tests--grep-regexp-testcases)
+ (let (msg1 msg2)
+ (setq msg1 (ert-info ((format "%S" testcase) :prefix "testcase: ")
+ (compile--test-error-line testcase)))
+ ;; Make sure replacing the NUL character with a colon still matches.
+ (setf (car testcase) (replace-regexp-in-string "\0" ":" (car
testcase)))
+ (setq msg2 (ert-info ((format "%S" testcase) :prefix "testcase: ")
+ (compile--test-error-line testcase)))
+ (should (equal msg1 msg2))))
+ (dolist (testcase compile-tests--grep-regexp-tricky-testcases)
+ (ert-info ((format "%S" testcase) :prefix "testcase: ")
+ (compile--test-error-line testcase)))))
+
;;; compile-tests.el ends here
diff --git a/test/lisp/shadowfile-tests.el b/test/lisp/shadowfile-tests.el
new file mode 100644
index 0000000..22f7b2d
--- /dev/null
+++ b/test/lisp/shadowfile-tests.el
@@ -0,0 +1,945 @@
+;;; shadowfile-tests.el --- Tests of shadowfile
+
+;; Copyright (C) 2018 Free Software Foundation, Inc.
+
+;; Author: Michael Albinus <address@hidden>
+
+;; This program is free software: you can redistribute it and/or
+;; modify it under the terms of the GNU General Public License as
+;; published by the Free Software Foundation, either version 3 of the
+;; License, or (at your option) any later version.
+;;
+;; This program is distributed in the hope that it will be useful, but
+;; WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+;; General Public License for more details.
+;;
+;; You should have received a copy of the GNU General Public License
+;; along with this program. If not, see `https://www.gnu.org/licenses/'.
+
+;;; Commentary:
+
+;; Some of the tests require access to a remote host files. Since
+;; this could be problematic, a mock-up connection method "mock" is
+;; used. Emulating a remote connection, it simply calls "sh -i".
+;; Tramp's file name handlers still run, so this test is sufficient
+;; except for connection establishing.
+
+;; If you want to test a real Tramp connection, set
+;; $REMOTE_TEMPORARY_FILE_DIRECTORY to a suitable value in order to
+;; overwrite the default value. If you want to skip tests accessing a
+;; remote host, set this environment variable to "/dev/null" or
+;; whatever is appropriate on your system.
+
+;; A whole test run can be performed calling the command `shadowfile-test-all'.
+
+;;; Code:
+
+(require 'ert)
+(require 'shadowfile)
+(require 'tramp)
+
+;; There is no default value on w32 systems, which could work out of the box.
+(defconst shadow-test-remote-temporary-file-directory
+ (cond
+ ((getenv "REMOTE_TEMPORARY_FILE_DIRECTORY"))
+ ((eq system-type 'windows-nt) null-device)
+ (t (add-to-list
+ 'tramp-methods
+ '("mock"
+ (tramp-login-program "sh")
+ (tramp-login-args (("-i")))
+ (tramp-remote-shell "/bin/sh")
+ (tramp-remote-shell-args ("-c"))
+ (tramp-connection-timeout 10)))
+ (add-to-list
+ 'tramp-default-host-alist
+ `("\\`mock\\'" nil ,(system-name)))
+ ;; Emacs' Makefile sets $HOME to a nonexistent value. Needed in
+ ;; batch mode only, therefore. It cannot be
+ ;; `temporary-directory', because the tests with "~" would fail.
+ (unless (and (null noninteractive) (file-directory-p "~/"))
+ (setenv "HOME" invocation-directory))
+ (format "/mock::%s" temporary-file-directory)))
+ "Temporary directory for Tramp tests.")
+
+(defconst shadow-test-info-file
+ (expand-file-name "shadows_test" temporary-file-directory)
+ "File to keep shadow information in during tests.")
+
+(defconst shadow-test-todo-file
+ (expand-file-name "shadow_todo_test" temporary-file-directory)
+ "File to store the list of uncopied shadows in during tests.")
+
+(ert-deftest shadow-test00-clusters ()
+ "Check cluster definitions.
+Per definition, all files are identical on the different hosts of
+a cluster (or site). This is not tested here; it must be
+guaranteed by the originator of a cluster definition."
+ (skip-unless (not (memq system-type '(windows-nt ms-dos))))
+ (skip-unless (file-remote-p shadow-test-remote-temporary-file-directory))
+
+ (let ((text-quoting-style 'grave) ;; We inspect the *Messages* buffer!
+ (inhibit-message t)
+ (shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file)
+ shadow-clusters
+ cluster primary regexp mocked-input)
+ (unwind-protect
+ ;; We must mock `read-from-minibuffer' and `read-string', in
+ ;; order to avoid interactive arguments.
+ (cl-letf* (((symbol-function 'read-from-minibuffer)
+ (lambda (&rest args) (pop mocked-input)))
+ ((symbol-function 'read-string)
+ (lambda (&rest args) (pop mocked-input))))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+
+ ;; Define a cluster.
+ (setq cluster "cluster"
+ primary shadow-system-name
+ regexp (shadow-regexp-superquote primary)
+ mocked-input `(,cluster ,primary ,regexp))
+ (call-interactively 'shadow-define-cluster)
+ (should
+ (string-equal
+ (shadow-cluster-name (shadow-get-cluster cluster)) cluster))
+ (should
+ (string-equal
+ (shadow-cluster-primary (shadow-get-cluster cluster)) primary))
+ (should
+ (string-equal
+ (shadow-cluster-regexp (shadow-get-cluster cluster)) regexp))
+ (should-not (shadow-get-cluster "non-existent-cluster-name"))
+
+ ;; Test `shadow-set-cluster' and `make-shadow-cluster'.
+ (shadow-set-cluster cluster primary regexp)
+ (should
+ (equal (shadow-get-cluster cluster)
+ (make-shadow-cluster
+ :name cluster :primary primary :regexp regexp)))
+
+ ;; The primary must be either `shadow-system-name', or a remote file.
+ (setq ;; The second "cluster" is wrong.
+ mocked-input `(,cluster ,cluster ,primary ,regexp))
+ (with-current-buffer (messages-buffer)
+ (narrow-to-region (point-max) (point-max)))
+ (call-interactively 'shadow-define-cluster)
+ (should
+ (string-match
+ (regexp-quote "Not a valid primary!")
+ (with-current-buffer (messages-buffer) (buffer-string))))
+ ;; The first cluster definition is still valid.
+ (should
+ (string-equal
+ (shadow-cluster-name (shadow-get-cluster cluster)) cluster))
+ (should
+ (string-equal
+ (shadow-cluster-primary (shadow-get-cluster cluster)) primary))
+ (should
+ (string-equal
+ (shadow-cluster-regexp (shadow-get-cluster cluster)) regexp))
+
+ ;; The regexp must match the primary name.
+ (setq ;; The second "cluster" is wrong.
+ mocked-input `(,cluster ,primary ,cluster ,regexp))
+ (with-current-buffer (messages-buffer)
+ (narrow-to-region (point-max) (point-max)))
+ (call-interactively 'shadow-define-cluster)
+ (should
+ (string-match
+ (regexp-quote "Regexp doesn't include the primary host!")
+ (with-current-buffer (messages-buffer) (buffer-string))))
+ ;; The first cluster definition is still valid.
+ (should
+ (string-equal
+ (shadow-cluster-name (shadow-get-cluster cluster)) cluster))
+ (should
+ (string-equal
+ (shadow-cluster-primary (shadow-get-cluster cluster)) primary))
+ (should
+ (string-equal
+ (shadow-cluster-regexp (shadow-get-cluster cluster)) regexp))
+
+ ;; Redefine the cluster.
+ (setq primary
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ regexp (shadow-regexp-superquote primary)
+ mocked-input `(,cluster ,primary ,regexp))
+ (call-interactively 'shadow-define-cluster)
+ (should
+ (string-equal
+ (shadow-cluster-name (shadow-get-cluster cluster)) cluster))
+ (should
+ (string-equal
+ (shadow-cluster-primary (shadow-get-cluster cluster)) primary))
+ (should
+ (string-equal
+ (shadow-cluster-regexp (shadow-get-cluster cluster)) regexp))
+
+ ;; Test `shadow-set-cluster' and `make-shadow-cluster'.
+ (shadow-set-cluster cluster primary regexp)
+ (should
+ (equal (shadow-get-cluster cluster)
+ (make-shadow-cluster
+ :name cluster :primary primary :regexp regexp))))
+
+ ;; Cleanup.
+ (with-current-buffer (messages-buffer) (widen))
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file)))))
+
+(ert-deftest shadow-test01-sites ()
+ "Check site definitions.
+Per definition, all files are identical on the different hosts of
+a cluster (or site). This is not tested here; it must be
+guaranteed by the originator of a cluster definition."
+ (skip-unless (not (memq system-type '(windows-nt ms-dos))))
+ (skip-unless (file-remote-p shadow-test-remote-temporary-file-directory))
+
+ (let ((shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file)
+ shadow-clusters
+ cluster1 cluster2 primary1 primary2 regexp1 regexp2 mocked-input)
+ (unwind-protect
+ ;; We must mock `read-from-minibuffer' and `read-string', in
+ ;; order to avoid interactive arguments.
+ (cl-letf* (((symbol-function 'read-from-minibuffer)
+ (lambda (&rest args) (pop mocked-input)))
+ ((symbol-function 'read-string)
+ (lambda (&rest args) (pop mocked-input))))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+
+ ;; Define a cluster.
+ (setq cluster1 "cluster1"
+ primary1 shadow-system-name
+ regexp1 (shadow-regexp-superquote primary1))
+ (shadow-set-cluster cluster1 primary1 regexp1)
+
+ ;; A site is either a cluster identification, or a primary host.
+ (should (string-equal cluster1 (shadow-site-name cluster1)))
+ (should (string-equal primary1 (shadow-name-site primary1)))
+ (should
+ (string-equal (format "/%s:" cluster1) (shadow-name-site cluster1)))
+ (should (string-equal (system-name) (shadow-site-name primary1)))
+ (should
+ (string-equal
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ (shadow-name-site
+ (file-remote-p shadow-test-remote-temporary-file-directory))))
+ (should
+ (string-equal
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ (shadow-site-name
+ (file-remote-p shadow-test-remote-temporary-file-directory))))
+
+ (should (equal (shadow-site-cluster cluster1)
+ (shadow-get-cluster cluster1)))
+ (should (equal (shadow-site-cluster (shadow-name-site cluster1))
+ (shadow-get-cluster cluster1)))
+ (should (equal (shadow-site-cluster primary1)
+ (shadow-get-cluster cluster1)))
+ (should (equal (shadow-site-cluster (shadow-site-name primary1))
+ (shadow-get-cluster cluster1)))
+ (should (string-equal (shadow-site-primary cluster1) primary1))
+ (should (string-equal (shadow-site-primary primary1) primary1))
+
+ ;; `shadow-read-site' accepts "cluster", "/cluster:",
+ ;; "system", "/system:". It shall reject bad site names.
+ (setq mocked-input
+ `(,cluster1 ,(shadow-name-site cluster1)
+ ,primary1 ,(shadow-site-name primary1)
+ ,shadow-system-name "" "bad" "/bad:"))
+ (should (string-equal (shadow-read-site) cluster1))
+ (should (string-equal (shadow-read-site) (shadow-name-site cluster1)))
+ (should (string-equal (shadow-read-site) primary1))
+ (should (string-equal (shadow-read-site) (shadow-site-name primary1)))
+ (should (string-equal (shadow-read-site) shadow-system-name))
+ (should-not (shadow-read-site)) ; ""
+ (should-not (shadow-read-site)) ; "bad"
+ (should-not (shadow-read-site)) ; "/bad:"
+ (should-error (shadow-read-site)) ; no input at all
+
+ ;; Define a second cluster.
+ (setq cluster2 "cluster2"
+ primary2
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ regexp2 (format "^\\(%s\\|%s\\)$" shadow-system-name primary2))
+ (shadow-set-cluster cluster2 primary2 regexp2)
+
+ ;; `shadow-site-match' shall know all different kind of site names.
+ (should (shadow-site-match cluster1 cluster1))
+ (should (shadow-site-match primary1 primary1))
+ (should (shadow-site-match cluster1 primary1))
+ (should (shadow-site-match primary1 cluster1))
+ (should (shadow-site-match cluster2 cluster2))
+ (should (shadow-site-match primary2 primary2))
+ (should (shadow-site-match cluster2 primary2))
+ (should (shadow-site-match primary2 cluster2))
+
+ ;; The regexp of `cluster2' matches the primary of
+ ;; `cluster1'. Not vice versa.
+ (should (shadow-site-match cluster2 cluster1))
+ (should-not (shadow-site-match cluster1 cluster2))
+
+ ;; If we use the primaries of a cluster, it doesn't match.
+ (should-not
+ (shadow-site-match (shadow-site-primary cluster2) cluster1))
+ (should-not
+ (shadow-site-match (shadow-site-primary cluster1) cluster2)))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file)))))
+
+(ert-deftest shadow-test02-files ()
+ "Check file manipulation functions."
+ (skip-unless (not (memq system-type '(windows-nt ms-dos))))
+ (skip-unless (file-remote-p shadow-test-remote-temporary-file-directory))
+
+ (let ((shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file)
+ shadow-clusters
+ cluster primary regexp file hup)
+ (unwind-protect
+ (progn
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+
+ ;; Define a cluster.
+ (setq cluster "cluster"
+ primary shadow-system-name
+ regexp (shadow-regexp-superquote primary)
+ file (make-temp-name
+ (expand-file-name
+ "shadowfile-tests" temporary-file-directory)))
+ (shadow-set-cluster cluster primary regexp)
+
+ ;; The constant structure to compare with.
+ (setq hup (make-tramp-file-name :host (system-name) :localname file))
+
+ ;; The structure a local file is transformed in.
+ (should (equal (shadow-parse-name file) hup))
+ (should (equal (shadow-parse-name (concat "/" cluster ":" file))
hup))
+ (should (equal (shadow-parse-name (concat primary file)) hup))
+
+ ;; A local file name is kept.
+ (should
+ (string-equal (shadow-local-file file) file))
+ ;; A file on this cluster is also local.
+ (should
+ (string-equal
+ (shadow-local-file (concat "/" cluster ":" file)) file))
+ ;; A file on the primary host is also local.
+ (should
+ (string-equal (shadow-local-file (concat primary file)) file))
+
+ ;; Redefine the cluster.
+ (setq primary
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster primary regexp)
+
+ ;; The structure of the local file is still the same.
+ (should (equal (shadow-parse-name file) hup))
+ ;; The cluster name must be used.
+ (setf (tramp-file-name-host hup) cluster)
+ (should (equal (shadow-parse-name (concat "/" cluster ":" file))
hup))
+ ;; The structure of a remote file is different.
+ (should
+ (equal (shadow-parse-name (concat primary file))
+ (tramp-dissect-file-name (concat primary file))))
+
+ ;; A local file is still local.
+ (should (shadow-local-file file))
+ ;; A file on this cluster is not local.
+ (should-not (shadow-local-file (concat "/" cluster ":" file)))
+ ;; A file on the primary host is not local.
+ (should-not (shadow-local-file (concat primary file)))
+ ;; There's no error on wrong FILE.
+ (should-not (shadow-local-file nil)))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file)))))
+
+(ert-deftest shadow-test03-expand-cluster-in-file-name ()
+ "Check canonical file name of a cluster or site."
+ (skip-unless (not (memq system-type '(windows-nt ms-dos))))
+ (skip-unless (file-remote-p shadow-test-remote-temporary-file-directory))
+
+ (let ((shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file)
+ shadow-clusters
+ cluster primary regexp file1 file2)
+ (unwind-protect
+ (progn
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+
+ ;; Define a cluster.
+ (setq cluster "cluster"
+ primary shadow-system-name
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster primary regexp)
+
+ (setq file1
+ (make-temp-name
+ (expand-file-name "shadowfile-tests" temporary-file-directory))
+ file2
+ (make-temp-name
+ (expand-file-name
+ "shadowfile-tests"
+ shadow-test-remote-temporary-file-directory)))
+
+ ;; A local file name is kept.
+ (should
+ (string-equal (shadow-expand-cluster-in-file-name file1) file1))
+ ;; A remote file is kept.
+ (should
+ (string-equal (shadow-expand-cluster-in-file-name file2) file2))
+ ;; A cluster name is expanded to the primary name.
+ (should
+ (string-equal
+ (shadow-expand-cluster-in-file-name (format "/%s:%s" cluster
file1))
+ (shadow-expand-cluster-in-file-name (concat primary file1))))
+ ;; A primary name is expanded if it is a local file name.
+ (should
+ (string-equal
+ (shadow-expand-cluster-in-file-name (concat primary file1)) file1))
+
+ ;; Redefine the cluster.
+ (setq primary
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster primary regexp)
+
+ ;; A cluster name is expanded to the primary name.
+ (should
+ (string-equal
+ (shadow-expand-cluster-in-file-name (format "/%s:%s" cluster
file1))
+ (shadow-expand-cluster-in-file-name (concat primary file1))))
+ ;; A primary name is not expanded if it isn't is a local file name.
+ (should
+ (string-equal
+ (shadow-expand-cluster-in-file-name (concat primary file1))
+ (concat primary file1))))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file)))))
+
+(ert-deftest shadow-test04-contract-file-name ()
+ "Check canonical file name of a cluster or site."
+ (skip-unless (not (memq system-type '(windows-nt ms-dos))))
+ (skip-unless (file-remote-p shadow-test-remote-temporary-file-directory))
+
+ (let ((shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file)
+ shadow-clusters
+ cluster primary regexp file)
+ (unwind-protect
+ (progn
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+
+ ;; Define a cluster.
+ (setq cluster "cluster"
+ primary shadow-system-name
+ regexp (shadow-regexp-superquote primary)
+ file (make-temp-name
+ (expand-file-name
+ "shadowfile-tests" temporary-file-directory)))
+ (shadow-set-cluster cluster primary regexp)
+
+ ;; The cluster name is prepended for local files.
+ (should
+ (string-equal
+ (shadow-contract-file-name file) (concat "/cluster:" file)))
+ ;; A cluster file name is preserved.
+ (should
+ (string-equal
+ (shadow-contract-file-name (concat "/cluster:" file))
+ (concat "/cluster:" file)))
+ ;; `shadow-system-name' is mapped to the cluster.
+ (should
+ (string-equal
+ (shadow-contract-file-name (concat shadow-system-name file))
+ (concat "/cluster:" file)))
+
+ ;; Redefine the cluster.
+ (setq primary
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster primary regexp)
+
+ ;; A remote file name is mapped to the cluster.
+ (should
+ (string-equal
+ (shadow-contract-file-name
+ (concat
+ (file-remote-p shadow-test-remote-temporary-file-directory)
file))
+ (concat "/cluster:" file))))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file)))))
+
+(ert-deftest shadow-test05-file-match ()
+ "Check `shadow-same-site' and `shadow-file-match'."
+ (skip-unless (not (memq system-type '(windows-nt ms-dos))))
+ (skip-unless (file-remote-p shadow-test-remote-temporary-file-directory))
+
+ (let ((shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file)
+ shadow-clusters
+ cluster primary regexp file)
+ (unwind-protect
+ (progn
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+
+ ;; Define a cluster.
+ (setq cluster "cluster"
+ primary shadow-system-name
+ regexp (shadow-regexp-superquote primary)
+ file (make-temp-name
+ (expand-file-name
+ "shadowfile-tests" temporary-file-directory)))
+ (shadow-set-cluster cluster primary regexp)
+
+ (should (shadow-same-site (shadow-parse-name "/cluster:") file))
+ (should
+ (shadow-same-site (shadow-parse-name shadow-system-name) file))
+ (should (shadow-same-site (shadow-parse-name file) file))
+
+ (should
+ (shadow-file-match
+ (shadow-parse-name (concat "/cluster:" file)) file))
+ (should
+ (shadow-file-match
+ (shadow-parse-name (concat shadow-system-name file)) file))
+ (should (shadow-file-match (shadow-parse-name file) file))
+
+ ;; Redefine the cluster.
+ (setq primary
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster primary regexp)
+
+ (should
+ (shadow-file-match
+ (shadow-parse-name
+ (concat
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ file))
+ file)))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file)))))
+
+(ert-deftest shadow-test06-literal-groups ()
+ "Check literal group definitions."
+ (skip-unless (not (memq system-type '(windows-nt ms-dos))))
+ (skip-unless (file-remote-p shadow-test-remote-temporary-file-directory))
+
+ (let ((shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file)
+ shadow-clusters shadow-literal-groups
+ cluster1 cluster2 primary regexp file1 file2 mocked-input)
+ (unwind-protect
+ ;; We must mock `read-from-minibuffer' and `read-string', in
+ ;; order to avoid interactive arguments.
+ (cl-letf* (((symbol-function 'read-from-minibuffer)
+ (lambda (&rest args) (pop mocked-input)))
+ ((symbol-function 'read-string)
+ (lambda (&rest args) (pop mocked-input))))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+
+ ;; Define clusters.
+ (setq cluster1 "cluster1"
+ primary shadow-system-name
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster1 primary regexp)
+
+ (setq cluster2 "cluster2"
+ primary
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ regexp (format "^\\(%s\\|%s\\)$" shadow-system-name primary))
+ (shadow-set-cluster cluster2 primary regexp)
+
+ ;; Define a literal group.
+ (setq file1
+ (make-temp-name
+ (expand-file-name "shadowfile-tests" temporary-file-directory))
+ file2
+ (make-temp-name
+ (expand-file-name
+ "shadowfile-tests"
+ shadow-test-remote-temporary-file-directory))
+ mocked-input `(,cluster1 ,file1 ,cluster2 ,file2 ,(kbd "RET")))
+ (with-temp-buffer
+ (setq-local buffer-file-name file1)
+ (call-interactively 'shadow-define-literal-group))
+
+ ;; `shadow-literal-groups' is a list of lists.
+ (should (consp shadow-literal-groups))
+ (should (consp (car shadow-literal-groups)))
+ (should-not (cdr shadow-literal-groups))
+
+ (should (member (format "/%s:%s" cluster1 (file-local-name file1))
+ (car shadow-literal-groups)))
+ (should (member (format "/%s:%s" cluster2 (file-local-name file2))
+ (car shadow-literal-groups))))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file)))))
+
+(ert-deftest shadow-test07-regexp-groups ()
+ "Check regexp group definitions."
+ (skip-unless (not (memq system-type '(windows-nt ms-dos))))
+ (skip-unless (file-remote-p shadow-test-remote-temporary-file-directory))
+
+ (let ((shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file)
+ shadow-clusters shadow-regexp-groups
+ cluster1 cluster2 primary regexp file mocked-input)
+ (unwind-protect
+ ;; We must mock `read-from-minibuffer' and `read-string', in
+ ;; order to avoid interactive arguments.
+ (cl-letf* (((symbol-function 'read-from-minibuffer)
+ (lambda (&rest args) (pop mocked-input)))
+ ((symbol-function 'read-string)
+ (lambda (&rest args) (pop mocked-input))))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+
+ ;; Define clusters.
+ (setq cluster1 "cluster1"
+ primary shadow-system-name
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster1 primary regexp)
+
+ (setq cluster2 "cluster2"
+ primary
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ regexp (format "^\\(%s\\|%s\\)$" shadow-system-name primary))
+ (shadow-set-cluster cluster2 primary regexp)
+
+ ;; Define a regexp group.
+ (setq file
+ (make-temp-name
+ (expand-file-name "shadowfile-tests" temporary-file-directory))
+ mocked-input `(,(shadow-regexp-superquote file)
+ ,cluster1 ,cluster2 ,(kbd "RET")))
+ (with-temp-buffer
+ (setq-local buffer-file-name nil)
+ (call-interactively 'shadow-define-regexp-group))
+
+ ;; `shadow-regexp-groups' is a list of lists.
+ (should (consp shadow-regexp-groups))
+ (should (consp (car shadow-regexp-groups)))
+ (should-not (cdr shadow-regexp-groups))
+
+ (should
+ (member
+ (concat
+ (shadow-site-primary cluster1) (shadow-regexp-superquote file))
+ (car shadow-regexp-groups)))
+ (should
+ (member
+ (concat
+ (shadow-site-primary cluster2) (shadow-regexp-superquote file))
+ (car shadow-regexp-groups))))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file)))))
+
+(ert-deftest shadow-test08-shadow-todo ()
+ "Check that needed shadows are added to todo."
+ (skip-unless (not (memq system-type '(windows-nt ms-dos))))
+ (skip-unless (file-remote-p shadow-test-remote-temporary-file-directory))
+ (skip-unless (file-writable-p shadow-test-remote-temporary-file-directory))
+
+ (let ((backup-inhibited t)
+ (shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file)
+ (shadow-inhibit-message t)
+ shadow-clusters shadow-literal-groups shadow-regexp-groups
+ shadow-files-to-copy
+ cluster1 cluster2 primary regexp file)
+ (unwind-protect
+ (condition-case err
+ (progn
+ (require 'trace)
+ (dolist (elt (all-completions "shadow-" obarray 'functionp))
+ (trace-function-background (intern elt)))
+ (trace-function-background 'save-buffer)
+ (dolist (elt write-file-functions)
+ (trace-function-background elt))
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+
+ (message "Point 1")
+ ;; Define clusters.
+ (setq cluster1 "cluster1"
+ primary shadow-system-name
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster1 primary regexp)
+
+ (setq cluster2 "cluster2"
+ primary
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster2 primary regexp)
+
+ (message "Point 2")
+ ;; Define a literal group.
+ (setq file
+ (make-temp-name
+ (expand-file-name "shadowfile-tests" temporary-file-directory))
+ shadow-literal-groups
+ `((,(concat "/cluster1:" file) ,(concat "/cluster2:" file))))
+
+ (message "Point 3")
+ ;; Save file from "cluster1" definition.
+ (with-temp-buffer
+ (setq buffer-file-name file)
+ (insert "foo")
+ (save-buffer))
+ (message "%s" file)
+ (message "%s" (shadow-contract-file-name (concat "/cluster2:" file)))
+ (message "%s" shadow-files-to-copy)
+ (should
+ (member
+ (cons file (shadow-contract-file-name (concat "/cluster2:" file)))
+ shadow-files-to-copy))
+
+ (message "Point 4")
+ ;; Save file from "cluster2" definition.
+ (with-temp-buffer
+ (message "Point 4.1")
+ (message "%s" file)
+ (message "%s" (shadow-site-primary cluster2))
+ (setq buffer-file-name (concat (shadow-site-primary cluster2)
file))
+ (message "Point 4.2")
+ (insert "foo")
+ (message "%s" buffer-file-name)
+ (message "%s" write-file-functions)
+ (setenv "BUG_32226" "1")
+ (save-buffer))
+ (setenv "BUG_32226")
+ (message "Point 4.3")
+ (message "%s" (shadow-site-primary cluster2))
+ (message "%s" (shadow-contract-file-name (concat "/cluster1:" file)))
+ (message "%s" shadow-files-to-copy)
+ (should
+ (member
+ (cons
+ (concat (shadow-site-primary cluster2) file)
+ (shadow-contract-file-name (concat "/cluster1:" file)))
+ shadow-files-to-copy))
+
+ (message "Point 5")
+ ;; Define a regexp group.
+ (setq shadow-files-to-copy nil
+ shadow-regexp-groups
+ `((,(concat (shadow-site-primary cluster1)
+ (shadow-regexp-superquote file))
+ ,(concat (shadow-site-primary cluster2)
+ (shadow-regexp-superquote file)))))
+
+ (message "Point 6")
+ ;; Save file from "cluster1" definition.
+ (with-temp-buffer
+ (setq buffer-file-name file)
+ (insert "foo")
+ (save-buffer))
+ (should
+ (member
+ (cons file (shadow-contract-file-name (concat "/cluster2:" file)))
+ shadow-files-to-copy))
+
+ (message "Point 7")
+ ;; Save file from "cluster2" definition.
+ (with-temp-buffer
+ (setq buffer-file-name (concat (shadow-site-primary cluster2)
file))
+ (insert "foo")
+ (save-buffer))
+ (should
+ (member
+ (cons
+ (concat (shadow-site-primary cluster2) file)
+ (shadow-contract-file-name (concat "/cluster1:" file)))
+ shadow-files-to-copy)))
+ (error (message "Error: %s" err) (signal (car err) (cdr err))))
+
+ (setenv "BUG_32226")
+ (untrace-all)
+ (message "%s" (with-current-buffer trace-buffer (buffer-string)))
+
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+ (ignore-errors
+ (when (file-exists-p file)
+ (delete-file file)))
+ (ignore-errors
+ (when (file-exists-p (concat (shadow-site-primary cluster2) file))
+ (delete-file (concat (shadow-site-primary cluster2) file)))))))
+
+(ert-deftest shadow-test09-shadow-copy-files ()
+ "Check that needed shadow files are copied."
+ (skip-unless (not (memq system-type '(windows-nt ms-dos))))
+ (skip-unless (file-remote-p shadow-test-remote-temporary-file-directory))
+
+ (let ((backup-inhibited t)
+ (shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file)
+ (shadow-inhibit-message t)
+ (shadow-noquery t)
+ shadow-clusters shadow-files-to-copy
+ cluster1 cluster2 primary regexp file mocked-input)
+ (unwind-protect
+ (progn
+ ;; Cleanup.
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+ (when (buffer-live-p shadow-todo-buffer)
+ (with-current-buffer shadow-todo-buffer (erase-buffer)))
+
+ ;; Define clusters.
+ (setq cluster1 "cluster1"
+ primary shadow-system-name
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster1 primary regexp)
+
+ (setq cluster2 "cluster2"
+ primary
+ (file-remote-p shadow-test-remote-temporary-file-directory)
+ regexp (shadow-regexp-superquote primary))
+ (shadow-set-cluster cluster2 primary regexp)
+
+ ;; Define files to copy.
+ (setq file
+ (make-temp-name
+ (expand-file-name "shadowfile-tests" temporary-file-directory))
+ shadow-literal-groups
+ `((,(concat "/cluster1:" file) ,(concat "/cluster2:" file)))
+ shadow-regexp-groups
+ `((,(concat (shadow-site-primary cluster1)
+ (shadow-regexp-superquote file))
+ ,(concat (shadow-site-primary cluster2)
+ (shadow-regexp-superquote file))))
+ mocked-input `(,(concat (shadow-site-primary cluster2) file)
+ ,file))
+
+ ;; Save files.
+ (with-temp-buffer
+ (setq buffer-file-name file)
+ (insert "foo")
+ (save-buffer))
+ (with-temp-buffer
+ (setq buffer-file-name (concat (shadow-site-primary cluster2)
file))
+ (insert "foo")
+ (save-buffer))
+
+ ;; We must mock `write-region', in order to check proper
+ ;; action.
+ (add-function
+ :before (symbol-function 'write-region)
+ (lambda (&rest args)
+ (when (and (buffer-file-name) mocked-input)
+ (should (equal (buffer-file-name) (pop mocked-input)))))
+ '((name . "write-region-mock")))
+
+ ;; Copy the files.
+ (shadow-copy-files 'noquery)
+ (should-not shadow-files-to-copy)
+ (with-current-buffer shadow-todo-buffer
+ (goto-char (point-min))
+ (should
+ (looking-at (regexp-quote "(setq shadow-files-to-copy nil)")))))
+
+ ;; Cleanup.
+ (remove-function (symbol-function 'write-region) "write-region-mock")
+ (when (file-exists-p shadow-info-file)
+ (delete-file shadow-info-file))
+ (when (file-exists-p shadow-todo-file)
+ (delete-file shadow-todo-file))
+ (ignore-errors
+ (when (file-exists-p file)
+ (delete-file file)))
+ (ignore-errors
+ (when (file-exists-p (concat (shadow-site-primary cluster2) file))
+ (delete-file (concat (shadow-site-primary cluster2) file)))))))
+
+(defun shadowfile-test-all (&optional interactive)
+ "Run all tests for \\[shadowfile]."
+ (interactive "p")
+ (if interactive
+ (ert-run-tests-interactively "^shadowfile-")
+ (ert-run-tests-batch "^shadowfile-")))
+
+(let ((shadow-info-file shadow-test-info-file)
+ (shadow-todo-file shadow-test-todo-file))
+ (shadow-initialize))
+
+(provide 'shadowfile-tests)
+;;; shadowfile-tests.el ends here
diff --git a/test/lisp/wdired-tests.el b/test/lisp/wdired-tests.el
new file mode 100644
index 0000000..b4ef4ab
--- /dev/null
+++ b/test/lisp/wdired-tests.el
@@ -0,0 +1,129 @@
+;;; wdired-tests.el --- tests for wdired.el -*- lexical-binding: t;
-*-
+
+;; Copyright (C) 2018 Free Software Foundation, Inc.
+
+;; This file is part of GNU Emacs.
+
+;; This program is free software; you can redistribute it and/or modify
+;; it under the terms of the GNU General Public License as published by
+;; the Free Software Foundation, either version 3 of the License, or
+;; (at your option) any later version.
+
+;; This program is distributed in the hope that it will be useful,
+;; but WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+;; GNU General Public License for more details.
+
+;; You should have received a copy of the GNU General Public License
+;; along with this program. If not, see <https://www.gnu.org/licenses/>.
+
+;;; Code:
+
+(require 'ert)
+(require 'dired)
+(require 'wdired)
+
+(defvar dired-query) ; Pacify byte compiler.
+
+(ert-deftest wdired-test-bug32173-01 ()
+ "Test using non-nil wdired-use-interactive-rename.
+Partially modifying a file name should succeed."
+ (let* ((test-dir (make-temp-file "test-dir-" t))
+ (test-file (concat (file-name-as-directory test-dir) "foo.c"))
+ (replace "bar")
+ (new-file (replace-regexp-in-string "foo" replace test-file))
+ (wdired-use-interactive-rename t))
+ (write-region "" nil test-file nil 'silent)
+ (advice-add 'dired-query ; Don't ask confirmation to overwrite a file.
+ :override
+ (lambda (_sym _prompt &rest _args) (setq dired-query t))
+ '((name . "advice-dired-query")))
+ (let ((buf (find-file-noselect test-dir)))
+ (unwind-protect
+ (with-current-buffer buf
+ (should (equal (dired-file-name-at-point) test-file))
+ (dired-toggle-read-only)
+ (kill-region (point) (progn (search-forward ".")
+ (forward-char -1) (point)))
+ (insert replace)
+ (wdired-finish-edit)
+ (should (equal (dired-file-name-at-point) new-file)))
+ (if buf (kill-buffer buf))
+ (delete-directory test-dir t)))))
+
+(ert-deftest wdired-test-bug32173-02 ()
+ "Test using non-nil wdired-use-interactive-rename.
+Aborting an edit should leaving original file name unchanged."
+ (let* ((test-dir (make-temp-file "test-dir-" t))
+ (test-file (concat (file-name-as-directory test-dir) "foo.c"))
+ (wdired-use-interactive-rename t))
+ (write-region "" nil test-file nil 'silent)
+ ;; Make dired-do-create-files-regexp a noop to mimic typing C-g
+ ;; at its prompt before wdired-finish-edit returns.
+ (advice-add 'dired-do-create-files-regexp
+ :override
+ (lambda (&rest _) (ignore))
+ '((name . "advice-dired-do-create-files-regexp")))
+ (let ((buf (find-file-noselect test-dir)))
+ (unwind-protect
+ (with-current-buffer buf
+ (should (equal (dired-file-name-at-point) test-file))
+ (dired-toggle-read-only)
+ (kill-region (point) (progn (search-forward ".")
+ (forward-char -1) (point)))
+ (insert "bar")
+ (wdired-finish-edit)
+ (should (equal (dired-get-filename) test-file)))
+ (if buf (kill-buffer buf))
+ (delete-directory test-dir t)))))
+
+(ert-deftest wdired-test-symlink-name ()
+ "Test the file name of a symbolic link.
+The Dired and WDired functions returning the name should include
+only the name before the link arrow."
+ (let* ((test-dir (make-temp-file "test-dir-" t))
+ (link-name "foo"))
+ (let ((buf (find-file-noselect test-dir)))
+ (unwind-protect
+ (with-current-buffer buf
+ (make-symbolic-link "./bar/baz" link-name)
+ (revert-buffer)
+ (let* ((file-name (dired-get-filename))
+ (dir-part (file-name-directory file-name))
+ (lf-name (concat dir-part link-name)))
+ (should (equal file-name lf-name))
+ (dired-toggle-read-only)
+ (should (equal (wdired-get-filename) lf-name))
+ (dired-toggle-read-only)))
+ (if buf (kill-buffer buf))
+ (delete-directory test-dir t)))))
+
+(ert-deftest wdired-test-unfinished-edit-01 ()
+ "Test editing a file name without saving the change.
+Finding the new name should be possible while still in
+wdired-mode."
+ :expected-result (if (< emacs-major-version 27) :failed :passed)
+ (let* ((test-dir (make-temp-file "test-dir-" t))
+ (test-file (concat (file-name-as-directory test-dir) "foo.c"))
+ (replace "bar")
+ (new-file (replace-regexp-in-string "foo" replace test-file)))
+ (write-region "" nil test-file nil 'silent)
+ (let ((buf (find-file-noselect test-dir)))
+ (unwind-protect
+ (with-current-buffer buf
+ (should (equal (dired-file-name-at-point) test-file))
+ (dired-toggle-read-only)
+ (kill-region (point) (progn (search-forward ".")
+ (forward-char -1) (point)))
+ (insert replace)
+ (should (equal (dired-get-filename) new-file))))
+ (when buf
+ (with-current-buffer buf
+ ;; Prevent kill-buffer-query-functions from chiming in.
+ (set-buffer-modified-p nil)
+ (kill-buffer buf)))
+ (delete-directory test-dir t))))
+
+
+(provide 'wdired-tests)
+;;; wdired-tests.el ends here
diff --git a/test/src/editfns-tests.el b/test/src/editfns-tests.el
index 501e0d8..8dee4bd 100644
--- a/test/src/editfns-tests.el
+++ b/test/src/editfns-tests.el
@@ -169,7 +169,13 @@
(should (eq (type-of (read (format "#o%o" most-negative-fixnum)))
'integer))
(should (eq (type-of (read (format "#32rG%x" most-positive-fixnum)))
- 'integer)))
+ 'integer))
+ (let ((binary-as-unsigned nil))
+ (dolist (fmt '("%d" "%s" "#o%o" "#x%x"))
+ (dolist (val (list most-negative-fixnum (1+ most-negative-fixnum)
+ -1 0 1
+ (1- most-positive-fixnum) most-positive-fixnum))
+ (should (eq val (read (format fmt val))))))))
(ert-deftest format-%o-invalid-float ()
(should-error (format "%o" -1e-37)
diff --git a/test/src/fns-tests.el b/test/src/fns-tests.el
index d560f0b..f722ed6 100644
--- a/test/src/fns-tests.el
+++ b/test/src/fns-tests.el
@@ -23,6 +23,17 @@
(require 'cl-lib)
+;; Test that equality predicates work correctly on NaNs when combined
+;; with hash tables based on those predicates. This was not the case
+;; for eql in Emacs 26.
+(ert-deftest fns-tests-equality-nan ()
+ (dolist (test (list #'eq #'eql #'equal))
+ (let* ((h (make-hash-table :test test))
+ (nan 0.0e+NaN)
+ (-nan (- nan)))
+ (puthash nan t h)
+ (should (eq (funcall test nan -nan) (gethash -nan h))))))
+
(ert-deftest fns-tests-reverse ()
(should-error (reverse))
(should-error (reverse 1))
diff --git a/test/src/regex-tests.el b/test/src/regex-emacs-tests.el
similarity index 99%
rename from test/src/regex-tests.el
rename to test/src/regex-emacs-tests.el
index 083ed5c..7a07590 100644
--- a/test/src/regex-tests.el
+++ b/test/src/regex-emacs-tests.el
@@ -1,4 +1,4 @@
-;;; regex-tests.el --- tests for regex.c functions -*- lexical-binding: t -*-
+;;; regex-emacs-tests.el --- tests for regex-emacs.c -*- lexical-binding: t -*-
;; Copyright (C) 2015-2018 Free Software Foundation, Inc.
@@ -24,7 +24,7 @@
(defvar regex-tests--resources-dir
(concat (concat (file-name-directory (or load-file-name buffer-file-name))
"/regex-resources/"))
- "Path to regex-resources directory next to the \"regex-tests.el\" file.")
+ "Path to regex-resources directory next to the \"regex-emacs-tests.el\"
file.")
(ert-deftest regex-word-cc-fallback-test ()
"Test that \"[[:cc:]]*x\" matches \"x\" (bug#24020).
@@ -683,4 +683,4 @@ This evaluates the TESTS test cases from glibc."
(should-not (string-match "\\`x\\{65535\\}" (make-string 65534 ?x)))
(should-error (string-match "\\`x\\{65536\\}" "X") :type 'invalid-regexp))
-;;; regex-tests.el ends here
+;;; regex-emacs-tests.el ends here
diff --git a/test/src/thread-tests.el b/test/src/thread-tests.el
index 0e909d3..364f6d6 100644
--- a/test/src/thread-tests.el
+++ b/test/src/thread-tests.el
@@ -19,38 +19,64 @@
;;; Code:
+;; Declare the functions in case Emacs has been configured --without-threads.
+(declare-function all-threads "thread.c" ())
+(declare-function condition-mutex "thread.c" (cond))
+(declare-function condition-name "thread.c" (cond))
+(declare-function condition-notify "thread.c" (cond &optional all))
+(declare-function condition-wait "thread.c" (cond))
+(declare-function current-thread "thread.c" ())
+(declare-function make-condition-variable "thread.c" (mutex &optional name))
+(declare-function make-mutex "thread.c" (&optional name))
+(declare-function make-thread "thread.c" (function &optional name))
+(declare-function mutex-lock "thread.c" (mutex))
+(declare-function mutex-unlock "thread.c" (mutex))
+(declare-function thread--blocker "thread.c" (thread))
+(declare-function thread-alive-p "thread.c" (thread))
+(declare-function thread-join "thread.c" (thread))
+(declare-function thread-last-error "thread.c" (&optional cleanup))
+(declare-function thread-name "thread.c" (thread))
+(declare-function thread-signal "thread.c" (thread error-symbol data))
+(declare-function thread-yield "thread.c" ())
+(defvar main-thread)
+
(ert-deftest threads-is-one ()
"Test for existence of a thread."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should (current-thread)))
(ert-deftest threads-threadp ()
"Test of threadp."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should (threadp (current-thread))))
(ert-deftest threads-type ()
"Test of thread type."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should (eq (type-of (current-thread)) 'thread)))
(ert-deftest threads-name ()
"Test for name of a thread."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(string= "hi bob" (thread-name (make-thread #'ignore "hi bob")))))
(ert-deftest threads-alive ()
"Test for thread liveness."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(thread-alive-p (make-thread #'ignore))))
(ert-deftest threads-all-threads ()
"Simple test for all-threads."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should (listp (all-threads))))
+(ert-deftest threads-main-thread ()
+ "Simple test for all-threads."
+ (skip-unless (featurep 'threads))
+ (should (eq main-thread (car (all-threads)))))
+
(defvar threads-test-global nil)
(defun threads-test-thread1 ()
@@ -58,7 +84,7 @@
(ert-deftest threads-basic ()
"Basic thread test."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(progn
(setq threads-test-global nil)
@@ -69,20 +95,29 @@
(ert-deftest threads-join ()
"Test of `thread-join'."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(progn
(setq threads-test-global nil)
(let ((thread (make-thread #'threads-test-thread1)))
- (thread-join thread)
- (and threads-test-global
- (not (thread-alive-p thread)))))))
+ (and (= (thread-join thread) 23)
+ (= threads-test-global 23)
+ (not (thread-alive-p thread)))))))
(ert-deftest threads-join-self ()
"Cannot `thread-join' the current thread."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should-error (thread-join (current-thread))))
+(ert-deftest threads-join-error ()
+ "Test of error signalling from `thread-join'."
+ :tags '(:unstable)
+ (skip-unless (featurep 'threads))
+ (let ((thread (make-thread #'threads-call-error)))
+ (while (thread-alive-p thread)
+ (thread-yield))
+ (should-error (thread-join thread))))
+
(defvar threads-test-binding nil)
(defun threads-test-thread2 ()
@@ -92,7 +127,7 @@
(ert-deftest threads-let-binding ()
"Simple test of threads and let bindings."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(progn
(setq threads-test-global nil)
@@ -104,22 +139,22 @@
(ert-deftest threads-mutexp ()
"Simple test of `mutexp'."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should-not (mutexp 'hi)))
(ert-deftest threads-mutexp-2 ()
"Another simple test of `mutexp'."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should (mutexp (make-mutex))))
(ert-deftest threads-mutex-type ()
"type-of mutex."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should (eq (type-of (make-mutex)) 'mutex)))
(ert-deftest threads-mutex-lock-unlock ()
"Test mutex-lock and unlock."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(let ((mx (make-mutex)))
(mutex-lock mx)
@@ -128,7 +163,7 @@
(ert-deftest threads-mutex-recursive ()
"Test mutex recursion."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(let ((mx (make-mutex)))
(mutex-lock mx)
@@ -149,7 +184,7 @@
(ert-deftest threads-mutex-contention ()
"Test of mutex contention."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(progn
(setq threads-mutex (make-mutex))
@@ -170,8 +205,8 @@
(ert-deftest threads-mutex-signal ()
"Test signaling a blocked thread."
- (skip-unless (fboundp 'make-thread))
- (should
+ (skip-unless (featurep 'threads))
+ (should-error
(progn
(setq threads-mutex (make-mutex))
(setq threads-mutex-key nil)
@@ -180,15 +215,17 @@
(while (not threads-mutex-key)
(thread-yield))
(thread-signal thr 'quit nil)
- (thread-join thr))
- t)))
+ ;; `quit' is not catched by `should-error'. We must indicate it.
+ (condition-case nil
+ (thread-join thr)
+ (quit (signal 'error nil)))))))
(defun threads-test-io-switch ()
(setq threads-test-global 23))
(ert-deftest threads-io-switch ()
"Test that `accept-process-output' causes thread switch."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(progn
(setq threads-test-global nil)
@@ -199,67 +236,71 @@
(ert-deftest threads-condvarp ()
"Simple test of `condition-variable-p'."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should-not (condition-variable-p 'hi)))
(ert-deftest threads-condvarp-2 ()
"Another simple test of `condition-variable-p'."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should (condition-variable-p (make-condition-variable (make-mutex)))))
(ert-deftest threads-condvar-type ()
"type-of condvar"
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should (eq (type-of (make-condition-variable (make-mutex)))
'condition-variable)))
(ert-deftest threads-condvar-mutex ()
"Simple test of `condition-mutex'."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(let ((m (make-mutex)))
(eq m (condition-mutex (make-condition-variable m))))))
(ert-deftest threads-condvar-name ()
"Simple test of `condition-name'."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(eq nil (condition-name (make-condition-variable (make-mutex))))))
(ert-deftest threads-condvar-name-2 ()
"Another simple test of `condition-name'."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(should
(string= "hi bob"
(condition-name (make-condition-variable (make-mutex)
"hi bob")))))
-(defun call-error ()
+
+(defun threads-call-error ()
"Call `error'."
(error "Error is called"))
;; This signals an error internally; the error should be caught.
-(defun thread-custom ()
- (defcustom thread-custom-face 'highlight
+(defun threads-custom ()
+ (defcustom threads-custom-face 'highlight
"Face used for thread customizations."
:type 'face
:group 'widget-faces))
-(ert-deftest thread-errors ()
+(ert-deftest threads-errors ()
"Test what happens when a thread signals an error."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(let (th1 th2)
- (setq th1 (make-thread #'call-error "call-error"))
+ (setq th1 (make-thread #'threads-call-error "call-error"))
(should (threadp th1))
(while (thread-alive-p th1)
(thread-yield))
(should (equal (thread-last-error)
'(error "Error is called")))
- (setq th2 (make-thread #'thread-custom "thread-custom"))
+ (should (equal (thread-last-error 'cleanup)
+ '(error "Error is called")))
+ (should-not (thread-last-error))
+ (setq th2 (make-thread #'threads-custom "threads-custom"))
(should (threadp th2))))
-(ert-deftest thread-sticky-point ()
+(ert-deftest threads-sticky-point ()
"Test bug #25165 with point movement in cloned buffer."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(with-temp-buffer
(insert "Lorem ipsum dolor sit amet, consectetur adipiscing elit.")
(goto-char (point-min))
@@ -268,9 +309,9 @@
(sit-for 1)
(should (= (point) 21))))
-(ert-deftest thread-signal-early ()
+(ert-deftest threads-signal-early ()
"Test signaling a thread as soon as it is started by the OS."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(let ((thread
(make-thread #'(lambda ()
(while t (thread-yield))))))
@@ -291,7 +332,7 @@
(ert-deftest threads-condvar-wait ()
"Test waiting on conditional variable."
- (skip-unless (fboundp 'make-thread))
+ (skip-unless (featurep 'threads))
(let ((cv-mutex (make-mutex))
new-thread)
;; We could have spurious threads from the previous tests still