[PATCH 1/3] grep: kill GREP_PROGRAM/EGREP_PROGRAM/FGREP_PROGRAM

[Paolo Bonzini]

* NEWS: Document slight semantic change.
* TODO: #ifdefs are gone.
* po/POTFILES.in: Update.
* src/Makefile.am (grep_SOURCES, egrep_SOURCES, fgrep_SOURCES): Remove
grep.c/egrep.c/fgrep.c.
(noinst_LIBRARIES): Change libsearch.a to libgrep.a.
(libsearch_a_SOURCES): Rename to libgrep_a_SOURCES, add grep.c
(LDADD): Change libsearch.a to libgrep.a.
* src/esearch.c: Add before_options and after_options.
* src/fsearch.c: Likewise.
* src/gsearch.c: Likewise.
* src/grep.c (short_options, long_options): Remove GREP_PROGRAM
special-casing.
(usage): Use before_options and after_options, look at matchers.
(setmatcher): Merge with install_matcher.
(main): Call setmatcher (NULL) before getopt instead of install_matcher after.
* src/grep.h (GREP_PROGRAM): Remove.
(before_options, after_options): Add.
---
 .x-sc_require_config_h       |    2 -
 .x-sc_require_config_h_first |    2 -
 NEWS                         |    6 ++
 TODO                         |    2 -
 po/POTFILES.in               |    3 +
 src/Makefile.am              |   14 +++---
 src/egrep.c                  |    2 -
 src/esearch.c                |    4 ++
 src/fgrep.c                  |    2 -
 src/fsearch.c                |    4 ++
 src/grep.c                   |  107 +++++++++++++++++-------------------------
 src/grep.h                   |   11 +---
 src/gsearch.c                |    6 ++
 13 files changed, 76 insertions(+), 89 deletions(-)
 delete mode 100644 .x-sc_require_config_h
 delete mode 100644 .x-sc_require_config_h_first
 delete mode 100644 src/egrep.c
 delete mode 100644 src/fgrep.c

diff --git a/.x-sc_require_config_h b/.x-sc_require_config_h
deleted file mode 100644
index c11a39b..0000000
--- a/.x-sc_require_config_h
+++ /dev/null
@@ -1,2 +0,0 @@
-src/egrep.c
-src/fgrep.c
diff --git a/.x-sc_require_config_h_first b/.x-sc_require_config_h_first
deleted file mode 100644
index c11a39b..0000000
--- a/.x-sc_require_config_h_first
+++ /dev/null
@@ -1,2 +0,0 @@
-src/egrep.c
-src/fgrep.c
diff --git a/NEWS b/NEWS
index 42611fa..8c61e03 100644
--- a/NEWS
+++ b/NEWS
@@ -40,6 +40,12 @@ GNU grep NEWS                                    -*- outline 
-*-
   --no-group-separator, enable further customization of the output
   when -A, -B or -C is being used.
 
+** Other changes
+
+  egrep accepts the -E option and fgrep accepts the -F option.  If egrep
+  and fgrep are given another of the -E/-F/-G options, they print a more
+  meaningful error message.
+
 * Noteworthy changes in release 2.5.4 (2009-02-10) [stable]
 
   - This is a bugfix release. No new features.
diff --git a/TODO b/TODO
index 2cfd0ce..bb2a7d4 100644
--- a/TODO
+++ b/TODO
@@ -28,8 +28,6 @@ Multithreading?
 GNU grep does 32-bit arithmetic, it needs to move to 64-bit (i.e.
 size_t/ptrdiff_t).
 
-Clean up, too many #ifdefs!
-
 Lazy dynamic linking of libpcre.
 
 Check FreeBSD's integration of zgrep (-Z) and bzgrep (-J) in one
diff --git a/po/POTFILES.in b/po/POTFILES.in
index e2454a0..b047712 100644
--- a/po/POTFILES.in
+++ b/po/POTFILES.in
@@ -27,4 +27,7 @@ src/dfa.c
 src/grep.c
 src/kwset.c
 src/dfasearch.c
+src/esearch.c
+src/fsearch.c
+src/gsearch.c
 src/pcresearch.c
diff --git a/src/Makefile.am b/src/Makefile.am
index 6571529..a4f686f 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -19,16 +19,16 @@ LN = ln
 AM_CFLAGS = $(WARN_CFLAGS) $(WERROR_CFLAGS)
 
 bin_PROGRAMS = grep egrep fgrep
-grep_SOURCES  =  grep.c gsearch.c
-egrep_SOURCES = egrep.c esearch.c
-fgrep_SOURCES = fgrep.c fsearch.c
+grep_SOURCES  = gsearch.c
+egrep_SOURCES = esearch.c
+fgrep_SOURCES = fsearch.c
 noinst_HEADERS = grep.h dfa.h kwset.h search.h system.h mbsupport.h
 
-noinst_LIBRARIES = libsearch.a
-libsearch_a_SOURCES = kwset.c dfa.c searchutils.c dfasearch.c kwsearch.c \
-       pcresearch.c
+noinst_LIBRARIES = libgrep.a
+libgrep_a_SOURCES = kwset.c dfa.c searchutils.c dfasearch.c kwsearch.c \
+       pcresearch.c grep.c
 
-LDADD = $(LIBINTL) libsearch.a ../lib/libgreputils.a
+LDADD = $(LIBINTL) libgrep.a ../lib/libgreputils.a
 grep_LDADD = $(LDADD) $(PCRE_LIBS)
 localedir = $(datadir)/locale
 INCLUDES = -I$(top_srcdir)/lib -DLOCALEDIR=\"$(localedir)\"
diff --git a/src/egrep.c b/src/egrep.c
deleted file mode 100644
index 1cabb4d..0000000
--- a/src/egrep.c
+++ /dev/null
@@ -1,2 +0,0 @@
-#define EGREP_PROGRAM
-#include "grep.c"
diff --git a/src/esearch.c b/src/esearch.c
index 647d2cb..80595fe 100644
--- a/src/esearch.c
+++ b/src/esearch.c
@@ -12,3 +12,7 @@ struct matcher const matchers[] = {
   { NULL, NULL, NULL },
 };
 
+const char before_options[] =
+N_("PATTERN is an extended regular expression (ERE).\n");
+const char after_options[] =
+N_("Invocation as `egrep' is deprecated; use `grep -E' instead.\n");
diff --git a/src/fgrep.c b/src/fgrep.c
deleted file mode 100644
index 4332310..0000000
--- a/src/fgrep.c
+++ /dev/null
@@ -1,2 +0,0 @@
-#define FGREP_PROGRAM
-#include "grep.c"
diff --git a/src/fsearch.c b/src/fsearch.c
index a524487..f91e626 100644
--- a/src/fsearch.c
+++ b/src/fsearch.c
@@ -6,3 +6,7 @@ struct matcher const matchers[] = {
   { NULL, NULL, NULL },
 };
 
+const char before_options[] =
+N_("PATTERN is a set of newline-separated fixed strings.\n");
+const char after_options[] =
+N_("Invocation as `fgrep' is deprecated; use `grep -F' instead.\n");
diff --git a/src/grep.c b/src/grep.c
index 225706b..e398adc 100644
--- a/src/grep.c
+++ b/src/grep.c
@@ -270,11 +270,7 @@ static struct exclude *included_patterns;
 static struct exclude *excluded_directory_patterns;
 /* Short options.  */
 static char const short_options[] =
-"0123456789A:B:C:D:HITUVabcd:e:f:hiKLlm:noqRrsuvwxyZz"
-#ifdef GREP_PROGRAM
-"EFGPX:"
-#endif
-;
+"0123456789A:B:C:D:EFGHIPTUVX:abcd:e:f:hiKLlm:noqRrsuvwxyZz";
 
 /* Non-boolean long options that have no corresponding short equivalents.  */
 enum
@@ -293,13 +289,11 @@ enum
 /* Long options equivalences. */
 static struct option const long_options[] =
 {
-#ifdef GREP_PROGRAM
   {"basic-regexp",    no_argument, NULL, 'G'},
   {"extended-regexp", no_argument, NULL, 'E'},
   {"fixed-regexp",    no_argument, NULL, 'F'},
   {"fixed-strings",   no_argument, NULL, 'F'},
   {"perl-regexp",     no_argument, NULL, 'P'},
-#endif
   {"after-context", required_argument, NULL, 'A'},
   {"before-context", required_argument, NULL, 'B'},
   {"binary-files", required_argument, NULL, BINARY_FILES_OPTION},
@@ -1454,28 +1448,18 @@ usage (int status)
       printf (_("Usage: %s [OPTION]... PATTERN [FILE]...\n"), program_name);
       printf (_("\
 Search for PATTERN in each FILE or standard input.\n"));
-#if defined(EGREP_PROGRAM)
-      printf (_("\
-PATTERN is an extended regular expression (ERE).\n"));
-#elif defined(FGREP_PROGRAM)
-      printf (_("\
-PATTERN is a set of newline-separated fixed strings.\n"));
-#else
-      printf (_("\
-PATTERN is, by default, a basic regular expression (BRE).\n"));
-#endif /* ?GREP_PROGRAM */
+      printf ("%s", gettext (before_options));
       printf (_("\
 Example: %s -i 'hello world' menu.h main.c\n\
 \n\
 Regexp selection and interpretation:\n"), program_name);
-#ifdef GREP_PROGRAM
-      printf (_("\
+      if (matchers[1].name)
+        printf (_("\
   -E, --extended-regexp     PATTERN is an extended regular expression (ERE)\n\
   -F, --fixed-strings       PATTERN is a set of newline-separated fixed 
strings\n\
   -G, --basic-regexp        PATTERN is a basic regular expression (BRE)\n\
   -P, --perl-regexp         PATTERN is a Perl regular expression\n"));
   /* -X is undocumented on purpose. */
-#endif /* GREP_PROGRAM */
       printf (_("\
   -e, --regexp=PATTERN      use PATTERN for matching\n\
   -f, --file=FILE           obtain PATTERN from FILE\n\
@@ -1544,17 +1528,7 @@ Context control:\n\
   -U, --binary              do not strip CR characters at EOL (MSDOS)\n\
   -u, --unix-byte-offsets   report offsets as if CRs were not there (MSDOS)\n\
 \n"));
-#if defined(EGREP_PROGRAM)
-      printf (_("\
-Invocation as `egrep' is deprecated; use `grep -E' instead.\n"));
-#elif defined(FGREP_PROGRAM)
-      printf (_("\
-Invocation as `fgrep' is deprecated; use `grep -F' instead.\n"));
-#else
-      printf (_("\
-`egrep' means `grep -E'.  `fgrep' means `grep -F'.\n\
-Direct invocation as either `egrep' or `fgrep' is deprecated.\n"));
-#endif /* ?GREP_PROGRAM */
+      printf ("%s", after_options);
       printf (_("\
 With no FILE, or when FILE is -, read standard input.  If less than two 
FILEs\n\
 are given, assume -h.  Exit status is 0 if any line was selected, 1 
otherwise;\n\
@@ -1569,34 +1543,48 @@ if any error occurs and -q was not given, the exit 
status is 2.\n"));
   exit (status);
 }
 
-static char const *matcher;
-
-#ifdef GREP_PROGRAM
-/* Set the matcher to M, reporting any conflicts.  */
+/* If M is NULL, initialize the matcher to the default.  Otherwise set the
+   matcher to M if available.  Exit in case of conflicts or if M is not
+   available.  */
 static void
 setmatcher (char const *m)
 {
-  if (matcher && strcmp (matcher, m) != 0)
-    error (EXIT_TROUBLE, 0, _("conflicting matchers specified"));
-  matcher = m;
-}
-#endif
+  static char const *matcher;
+  unsigned int i;
 
-/* Go through the matchers vector and look for the specified matcher.
-   If we find it, install it in compile and execute, and return 1.  */
-static int
-install_matcher (char const *name)
-{
-  int i;
+  if (!m)
+    {
+      compile = matchers[0].compile;
+      execute = matchers[0].execute;
+      if (!matchers[1].name)
+       matcher = matchers[0].name;
+    }
 
-  for (i = 0; matchers[i].name; i++)
-    if (strcmp (name, matchers[i].name) == 0)
-      {
-       compile = matchers[i].compile;
-       execute = matchers[i].execute;
-       return 1;
-      }
-  return 0;
+  else if (matcher)
+    {
+      if (matcher && strcmp (matcher, m) == 0)
+        ;
+
+      else if (!matchers[1].name)
+        error (EXIT_TROUBLE, 0, _("%s can only use the %s pattern syntax"),
+               program_name, matcher);
+      else
+        error (EXIT_TROUBLE, 0, _("conflicting matchers specified"));
+    }
+
+  else
+    {
+      for (i = 0; matchers[i].name; i++)
+        if (strcmp (m, matchers[i].name) == 0)
+          {
+            compile = matchers[i].compile;
+            execute = matchers[i].execute;
+            matcher = m;
+            return;
+          }
+
+      error (EXIT_TROUBLE, 0, _("invalid matcher %s"), m);
+    }
 }
 
 static void
@@ -1865,6 +1853,7 @@ main (int argc, char **argv)
   atexit (close_stdout);
 
   prepend_default_options (getenv ("GREP_OPTIONS"), &argc, &argv);
+  setmatcher (NULL);
 
   while ((opt = get_nondigit_option (argc, argv, &default_context)) != -1)
     switch (opt)
@@ -1892,7 +1881,6 @@ main (int argc, char **argv)
          error (EXIT_TROUBLE, 0, _("unknown devices method"));
        break;
 
-#ifdef GREP_PROGRAM
       case 'E':
        setmatcher ("egrep");
        break;
@@ -1912,7 +1900,6 @@ main (int argc, char **argv)
       case 'X': /* undocumented on purpose */
        setmatcher (optarg);
        break;
-#endif /* GREP_PROGRAM */
 
       case 'H':
        with_filenames = 1;
@@ -2228,15 +2215,7 @@ There is NO WARRANTY, to the extent permitted by 
law.\n"),
     else
       usage (EXIT_TROUBLE);
 
-  if (matcher && install_matcher (matcher))
-    ;
-  else if (install_matcher (matchers[0].name))
-    ;
-  else
-    abort ();
-
   set_limits();
-
   compile(keys, keycc);
   free (keys);
 
diff --git a/src/grep.h b/src/grep.h
index 8d46a69..67ea793 100644
--- a/src/grep.h
+++ b/src/grep.h
@@ -21,14 +21,6 @@
 # define __attribute__(x)
 #endif
 
-/* We build specialized legacy "egrep" and "fgrep" programs.
-   No program adjusts its behavior according to its argv[0].
-   No scripts are provided as an alternative.  Distributors
-   are free to do otherwise, but it is their burden to do so.  */
-#if !defined(GREP_PROGRAM) && !defined(EGREP_PROGRAM) && 
!defined(FGREP_PROGRAM)
-# define GREP_PROGRAM
-#endif
-
 /* Function pointer types.  */
 typedef void (*compile_fp_t) (char const *, size_t);
 typedef size_t (*execute_fp_t) (char const *, size_t, size_t *, char const *);
@@ -42,6 +34,9 @@ extern struct matcher
   execute_fp_t execute;
 } const matchers[];
 
+extern const char before_options[];
+extern const char after_options[];
+
 /* The following flags are exported from grep for the matchers
    to look at. */
 extern int match_icase;                /* -i */
diff --git a/src/gsearch.c b/src/gsearch.c
index b71d3b7..8ef4968 100644
--- a/src/gsearch.c
+++ b/src/gsearch.c
@@ -30,3 +30,9 @@ struct matcher const matchers[] = {
   { "perl",    Pcompile, Pexecute },
   { NULL, NULL, NULL },
 };
+
+const char before_options[] =
+N_("PATTERN is, by default, a basic regular expression (BRE).\n");
+const char after_options[] =
+N_("`egrep' means `grep -E'.  `fgrep' means `grep -F'.\n\
+Direct invocation as either `egrep' or `fgrep' is deprecated.\n");
-- 
1.6.6.1