[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: FreeBSD vs. Debian behavior discrepancy
From: |
Jim Meyering |
Subject: |
Re: FreeBSD vs. Debian behavior discrepancy |
Date: |
Wed, 02 Nov 2011 18:55:35 +0100 |
Paul Eggert wrote:
> I'm afraid that script is busted. It should use xargs -r.
> You can see the bug by running the script with stdout
> not being a terminal -- the script will send garbage to stdout.
>
> That being said, the FreeBSD behavior sounds a bit
> more useful, so I installed the following patch, and
> the new behavior should be in the next version of GNU gzip.
>
>>From 749cadea46b273be2acb50189341df1cf6a51492 Mon Sep 17 00:00:00 2001
> From: Paul Eggert <address@hidden>
> Date: Wed, 2 Nov 2011 10:35:00 -0700
> Subject: [PATCH] * gzip.c (treat_stdin): If quiet, be quiet with plain gzip
> -q.
> MIME-Version: 1.0
> Content-Type: text/plain; charset=UTF-8
> Content-Transfer-Encoding: 8bit
>
> Problem reported by Michaël Guitton in
> <http://lists.gnu.org/archive/html/bug-gzip/2011-11/msg00000.html>.
> ---
> gzip.c | 14 +++++++++-----
> 1 files changed, 9 insertions(+), 5 deletions(-)
>
> diff --git a/gzip.c b/gzip.c
> index 6164438..cb10a23 100644
> --- a/gzip.c
> +++ b/gzip.c
> @@ -617,11 +617,15 @@ local void treat_stdin()
> *
> * Here we use the --force option to get the other behavior.
> */
> - fprintf(stderr,
> - "%s: compressed data not %s a terminal. Use -f to force
> %scompression.\n",
> - program_name, decompress ? "read from" : "written to",
> - decompress ? "de" : "");
> - fprintf (stderr, "For help, type: %s -h\n", program_name);
> + if (! quiet)
> + fprintf (stderr,
> + ("%s: compressed data not %s a terminal."
> + " Use -f to force %scompression.\n"
> + "For help, type: %s -h\n"),
> + program_name,
> + decompress ? "read from" : "written to",
> + decompress ? "de" : "",
> + program_name);
> do_exit(ERROR);
> }
Thanks, Paul.
It's no big deal, but that broke "make distcheck", which happens
to run "make syntax-check" and which checks for leading TABs.
I fixed that with 1/3 below.
That made me notice I'd fixed a minor bug in a test script
back in July (2/3).
And it's been a relatively long time since the last gnulib update,
so I've done that, too (3/3).
>From dd9cda0858dbe50fa638db267bdf63c77afc2c75 Mon Sep 17 00:00:00 2001
From: Jim Meyering <address@hidden>
Date: Wed, 2 Nov 2011 18:47:50 +0100
Subject: [PATCH 1/3] maint: avoid "make syntax-check" failure
* gzip.c (treat_stdin): Indent with spaces, not TABs.
---
gzip.c | 18 +++++++++---------
1 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/gzip.c b/gzip.c
index cb10a23..b454e6d 100644
--- a/gzip.c
+++ b/gzip.c
@@ -617,15 +617,15 @@ local void treat_stdin()
*
* Here we use the --force option to get the other behavior.
*/
- if (! quiet)
- fprintf (stderr,
- ("%s: compressed data not %s a terminal."
- " Use -f to force %scompression.\n"
- "For help, type: %s -h\n"),
- program_name,
- decompress ? "read from" : "written to",
- decompress ? "de" : "",
- program_name);
+ if (! quiet)
+ fprintf (stderr,
+ ("%s: compressed data not %s a terminal."
+ " Use -f to force %scompression.\n"
+ "For help, type: %s -h\n"),
+ program_name,
+ decompress ? "read from" : "written to",
+ decompress ? "de" : "",
+ program_name);
do_exit(ERROR);
}
--
1.7.8.rc0.32.g87bf9
>From 50a1a56051c77bd16f4a5c52f650c10ac8db57c2 Mon Sep 17 00:00:00 2001
From: Jim Meyering <address@hidden>
Date: Sun, 24 Jul 2011 11:43:47 +0200
Subject: [PATCH 2/3] tests: mixed: correct size-enumeration logic
* tests/mixed (sizes): Fix misplaced "&& break" that made us test
only with a size of 0, rather than all sizes in 0..64.
---
tests/mixed | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/tests/mixed b/tests/mixed
index bd561e8..22913df 100644
--- a/tests/mixed
+++ b/tests/mixed
@@ -48,7 +48,7 @@
in_str=0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ_-+=%
for i in 0 1 2 3 4 5 6 7 8 9 a; do in_str="$in_str$in_str" ;done
# Start with some small sizes. $(seq 64)
-sizes=$(i=0; while :; do echo $i; test $i = 64; i=$(expr $i + 1) && break;
done)
+sizes=$(i=0; while :; do echo $i; test $i = 64 && break; i=$(expr $i + 1);
done)
# gzip's internal buffer size is 32KiB + 64 bytes:
sizes="$sizes 32831 32832 32833"
--
1.7.8.rc0.32.g87bf9
>From b477de776b18203341072692915a77c753db88a6 Mon Sep 17 00:00:00 2001
From: Jim Meyering <address@hidden>
Date: Wed, 2 Nov 2011 18:41:11 +0100
Subject: [PATCH 3/3] build: update gnulib submodule to latest
---
gnulib | 2 +-
lib/.gitignore | 11 +++++++++++
m4/.gitignore | 7 +++++++
3 files changed, 19 insertions(+), 1 deletions(-)
diff --git a/gnulib b/gnulib
index 5b9e0c3..afe598a 160000
--- a/gnulib
+++ b/gnulib
@@ -1 +1 @@
-Subproject commit 5b9e0c3ce734b9d9b0c0001e31b65a8a4713700d
+Subproject commit afe598a51a14e75cd697a30ac54fa9cc29c8cde0
diff --git a/lib/.gitignore b/lib/.gitignore
index 61e76b1..9820da1 100644
--- a/lib/.gitignore
+++ b/lib/.gitignore
@@ -18,6 +18,7 @@
/close-stream.c
/close-stream.h
/close.c
+/closedir.c
/closein.c
/closein.h
/closeout.c
@@ -27,6 +28,7 @@
/config.hin
/configmake.h
/creat-safer.c
+/dirent-private.h
/dirent.h
/dirent.in.h
/dirfd.c
@@ -34,6 +36,7 @@
/dirname.h
/dosname.h
/dup-safer.c
+/dup.c
/dup2.c
/errno.h
/errno.in.h
@@ -57,6 +60,7 @@
/fd-safer.c
/fdopendir.c
/fflush.c
+/filename.h
/filenamecat-lgpl.c
/filenamecat.h
/float+.h
@@ -79,6 +83,7 @@
/fseeko.c
/fseterr.c
/fseterr.h
+/fstat.c
/fstatat.c
/ftell.c
/ftello.c
@@ -105,6 +110,7 @@
/isnanl-nolibm.h
/isnanl.c
/iswblank.c
+/itold.c
/lchown.c
/libgzip.a
/localcharset.c
@@ -122,6 +128,10 @@
/memrchr.c
/mkdir.c
/mkdirat.c
+/msvc-inval.c
+/msvc-inval.h
+/msvc-nothrow.c
+/msvc-nothrow.h
/open-safer.c
/open.c
/openat-die.c
@@ -129,6 +139,7 @@
/openat-proc.c
/openat.c
/openat.h
+/opendir.c
/pathmax.h
/perror.c
/pipe-safer.c
diff --git a/m4/.gitignore b/m4/.gitignore
index a566731..b3b29dc 100644
--- a/m4/.gitignore
+++ b/m4/.gitignore
@@ -9,6 +9,7 @@
/cloexec.m4
/close-stream.m4
/close.m4
+/closedir.m4
/closein.m4
/closeout.m4
/codeset.m4
@@ -19,6 +20,7 @@
/dirname.m4
/dos.m4
/double-slash-root.m4
+/dup.m4
/dup2.m4
/errno_h.m4
/error.m4
@@ -46,6 +48,7 @@
/frexpl.m4
/fseek.m4
/fseeko.m4
+/fstat.m4
/ftell.m4
/ftello.m4
/getcwd-abort-bug.m4
@@ -96,10 +99,13 @@
/mkdir.m4
/mmap-anon.m4
/mode_t.m4
+/msvc-inval.m4
+/msvc-nothrow.m4
/multiarch.m4
/nocrash.m4
/open.m4
/openat.m4
+/opendir.m4
/pathmax.m4
/perror.m4
/printf-frexp.m4
@@ -112,6 +118,7 @@
/save-cwd.m4
/signbit.m4
/size_max.m4
+/ssize_t.m4
/stat-time.m4
/stat.m4
/stdbool.m4
--
1.7.8.rc0.32.g87bf9
- FreeBSD vs. Debian behavior discrepancy, Michaël Guitton, 2011/11/02
- Re: FreeBSD vs. Debian behavior discrepancy, Paul Eggert, 2011/11/02
- Re: FreeBSD vs. Debian behavior discrepancy,
Jim Meyering <=
- Re: FreeBSD vs. Debian behavior discrepancy, Paul Eggert, 2011/11/02
- gzip zgrep-signal failure on Fedora 15 in an Emacs shell window, Paul Eggert, 2011/11/02
- Re: gzip zgrep-signal failure on Fedora 15 in an Emacs shell window, Eric Blake, 2011/11/02
- Re: gzip zgrep-signal failure on Fedora 15 in an Emacs shell window, Paul Eggert, 2011/11/02
- Re: gzip zgrep-signal failure on Fedora 15 in an Emacs shell window, Eric Blake, 2011/11/02
- Re: gzip zgrep-signal failure on Fedora 15 in an Emacs shell window, Paul Eggert, 2011/11/02