Re: --version output change proposal

[EMAIL PROTECTED] (Karl Berry) writes: > rms and I have been batting around the --version output stuff. > Here's what we seem to be settling on: > > sort (GNU coreutils) 6.4 > Copyright (C) 2006 Free Software Foundation, Inc. > License GPLv2+: GNU GPL version 2 or later

Re: --version output change proposal

Simon Josefsson <[EMAIL PROTECTED]> wrote: > [EMAIL PROTECTED] (Karl Berry) writes: >> If any comments, let me know before we inscribe it into standards.texi ... > > I suppose lib/version.c in gnulib should be modified too? Yep, version-etc.c I'll wait a day or two, in case things change.

Re: warnings in some test files

Bruno Haible <[EMAIL PROTECTED]> writes: > Hi Simon, > > Got these gcc warnings: > > test-crc.c:37: warning: long unsigned int format, unsigned int arg (arg 2) > test-crc.c:44: warning: long unsigned int format, unsigned int arg (arg 2) > test-crc.c:51: warning: long unsigned int format, unsigned

Re: remove module iconvme?

Bruno Haible <[EMAIL PROTECTED]> writes: > Hi Simon, > > The iconvme module gives compilation warnings on some platforms: > > iconvme.c:142: warning: passing arg 2 of `libiconv' from incompatible pointer > type > > Before spending time fixing this: This module has already been superseded > by mod

minor performance improvement to getcwd.c

I installed this: 2007-03-19 Paul Eggert <[EMAIL PROTECTED]> * lib/getcwd.c (__getcwd): Remove redundant comparison of buf to NULL. If size is zero here, buf must be zero. --- lib/getcwd.c21 Feb 2007 20:50:00 - 1.23 +++ lib/getcwd.c19 Mar 2007 17:30:04

Re: adding bootstrap and bootstrap.conf

Bruno Haible <[EMAIL PROTECTED]> writes: > Can you try to solve both problems through overrides in > coreutils/gl/modules/gettext.diff and coreutils/gl/m4/* ? > If you get that working, I can see what needs to be changed in gnulib's > gettext module. In the current state, I don't even have a preci

[patch] doc/gnulib.texi typofix

--- gnulib.texi 15 Mar 2007 22:58:36 - 1.33 +++ gnulib.texi 19 Mar 2007 10:42:14 - @@ -202,7 +202,7 @@ handling happens in rare situations, but performing the out of memory error handling after almost all API function invocations pollute your source code and might make it harder to

[patch] doc/gnulib.texi typofixes

ChangeLog entry: * doc/gnulib.texi (Out of memory handling): Fix typos. diff: --- gnulib.texi 15 Mar 2007 22:58:36 - 1.33 +++ gnulib.texi 19 Mar 2007 10:46:02 - @@ -202,16 +202,16 @@ handling happens in rare situations, but performing the out of memory error handling after

[patch] doc/relocatable-maint.texi example sveltening

ChangeLog entry: * doc/relocatable-maint.texi (Supporting Relocation): Refill the script example, eliding one sed invocation. diff attached relocatable-maint.texi.diff Description: Binary data

Re: [patch] doc/gnulib.texi typofix

Thien-Thi Nguyen <[EMAIL PROTECTED]> writes: > ChangeLog entry: > > * doc/gnulib.texi (Out of memory handling): Fix typos. Thanks, I committed those typo fixes. -- "Sanity is not statistical." --George Orwell

add limited support for Solaris 10 ZFS-style ACLs to gnulib

Sun has seriously rethought ACLs for Solaris 10 and ZFS. I installed this first step at supporting them in gnulib; it's just enough to get GNU 'ls' to work. I have tested this on Solaris 8, Solaris 10, and Debian stable; I haven't tested it on a Linux system supporting ACLs. 2007-03-19 Paul Egg

Re: add limited support for Solaris 10 ZFS-style ACLs to gnulib

Paul Eggert <[EMAIL PROTECTED]> wrote: > Sun has seriously rethought ACLs for Solaris 10 and ZFS. I installed > this first step at supporting them in gnulib; it's just enough to get > GNU 'ls' to work. I have tested this on Solaris 8, Solaris 10, and > Debian stable; I haven't tested it on a Linu

Re: --version output change proposal

However, some legal nitpickers might say that the line "you are free to change and redistribute it" establishes different licensing You are right, it might happen, but I just can't get too excited about catering to such people. It would be a truly outrageous interpretation of a simple exp

Re: [patch] doc/relocatable-maint.texi example sveltening

Thien-Thi Nguyen wrote: > ChangeLog entry: > > * doc/relocatable-maint.texi (Supporting Relocation): > Refill the script example, eliding one sed invocation. Thanks. I don't feel that eliding the second sed invocation improves clarity, therefore I'm applying only a reformatting:

Re: adding bootstrap and bootstrap.conf

Jim Meyring wrote: > My impression is that few (if any) of the autogen scripts use gnulib The one in gettext does. > there's another definition (the first one when I type "dict bootstrap") > that is more evocative: to load and initialize the operating system on > a computer. Normally abbreviated

Re: adding bootstrap and bootstrap.conf

Jim Meyering wrote: > Another feature: bootstrap adds .cvsignore and .gitignore > entries for files it supplies. Good idea. It has annoyed me as well, but never reached the necessary pain level. Find attached a patch that implements the first part of it. The second part will be the handling of CLE

Re: *printf declarations in stdio.h?

Simon Josefsson wrote: > How about this? > > /Simon > > 2007-03-13 Simon Josefsson <[EMAIL PROTECTED]> > > * m4/stdio_h.m4: Add stubs for vasprintf too. > > * modules/stdio: Support vasprintf in sed command. > > * modules/vasprintf: Depend on stdio for prototypes, and remo

remove module iconvme?

Hi Simon, The iconvme module gives compilation warnings on some platforms: iconvme.c:142: warning: passing arg 2 of `libiconv' from incompatible pointer type Before spending time fixing this: This module has already been superseded by module 'striconv' and 'xstriconv', the migration path has al

Re: remove module iconvme?

Simon Josefsson wrote: > Yes, that's ok with me. OK, I'm removing it: 2007-03-19 Bruno Haible <[EMAIL PROTECTED]> * modules/iconvme: Remove file. * lib/iconvme.h: Remove file. * lib/iconvme.c: Remove file. * m4/iconvme.m4: Remove file. *** NEWS17 Mar 20

warning in at-func.c

Hi Jim, On MacOS X, we have this warning during the compilation of fchownat.c: at-func.c:39: warning: implicit declaration of function `lchown' The reason is that lchown() is not available on this system and therefore not declared in . This proposed patch fixes it: 2007-03-17 Bruno Haible <[

mkstemp declaration fix

Hello, On MacOS X 10.3, one gets this warning: mkstemp-safer.c:34: warning: implicit declaration of function `mkstemp' The reason is that mkstemp is declared in , not , on this platform. But for POSIX compatibility, only gnulib's should be sufficient to declare the function. I'm applying this:

gcc warning in dirname-tests

Got these warnings from "gcc -Wall": test-dirname.c:158: warning: int format, long unsigned int arg (arg 6) test-dirname.c:173: warning: int format, long unsigned int arg (arg 6) Fixing it like this: 2007-03-17 Bruno Haible <[EMAIL PROTECTED]> * tests/test-dirname.c (main): Make print

warnings in some test files

Hi Simon, Got these gcc warnings: test-crc.c:37: warning: long unsigned int format, unsigned int arg (arg 2) test-crc.c:44: warning: long unsigned int format, unsigned int arg (arg 2) test-crc.c:51: warning: long unsigned int format, unsigned int arg (arg 2) test-crc.c:58: warning: long unsigned

m4/locale-fr.m4: workaround NetBSD bug

Some of the *printf-posix tests fail on NetBSD, because NetBSD pretends to have a french locale (fr_FR.ISO8859-1). But it has a wrong decimal-point character, so is unusable for the tests. 2007-03-17 Bruno Haible <[EMAIL PROTECTED]> * m4/locale-fr.m4 (gt_LOCALE_FR, gt_LOCALE_FR_UTF8): C

argp test failure on NetBSD

Hello Sergey, The 'argp' unit test fails on NetBSD 3.0: stdin argp.12775 differ: char 118, line 2 stdin argp.12775 differ: char 106, line 2 stdin argp.12775 differ: char 252, line 12 FAIL: test-argp-2.sh Since this output is not very informative, I'm changing the test to make the comparison thro

another vasnprintf fix

Performing the unit tests on NetBSD 3.0 uncovered this: When the "size of given buffer" argument passed to vasnprintf() was > INT_MAX, the function could loop endlessly instead of failing with EOVERFLOW. Note: NetBSD also has a bug here: snprintf() returns -1 with errno = EINVAL, where POSIX says

Re: another vasnprintf fix

Eric Blake wrote: > since vasnprintf has the > nice API property that, unlike all other *printf functions, it does not > return an int, it should not need to be limited by INT_MAX; otherwise you > are introducing an artificial limitation You're right; I forgot about this argumentation. My patch co

Re: --version output change proposal

Hello Karl, > sort (GNU coreutils) 6.4 > Copyright (C) 2006 Free Software Foundation, Inc. > License GPLv2+: GNU GPL version 2 or later > This is free software: you are free to change and redistribute it. > There is NO WARRANTY, to the extent permitted by law. My gu

Re: add limited support for Solaris 10 ZFS-style ACLs to gnulib

Paul Eggert wrote: > +#if ENABLE_NLS > +# include > +# define _(Text) gettext (Text) > +#else > +# define _(Text) Text > +#endif This can be abbreviated to #include "gettext.h" admitting a dependency to the 'gettext' module. (People who don't want the gettext module can override it with a dum

progname.h contradicts gnu coding standards

The instructions in progname.h say: /* Programs using this file should do the following in main(): set_program_name (argv[0]); */ /* Set program_name, based on argv[0]. */ extern void set_program_name (const char *argv0); but the GNU coding standards say: "The program's name should be

Re: progname.h contradicts gnu coding standards

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to John Darrington on 3/19/2007 9:58 PM: > The instructions in progname.h say: > > /* Programs using this file should do the following in main(): > set_program_name (argv[0]); > */ > > /* Set program_name, based on argv[0]. */ > exte

Re: add limited support for Solaris 10 ZFS-style ACLs to gnulib

Jim Meyering <[EMAIL PROTECTED]> writes: > FYI, cp in particular needs working ACL-detection support, too. > The problem is when a Linux system mounts a ZFS file system. > Then, there is no acl_trivial function, yet the number of > ACL entries is 4, and the ACL *is* trivial. Yes. A related issue

Re: add limited support for Solaris 10 ZFS-style ACLs to gnulib

Bruno Haible <[EMAIL PROTECTED]> writes: > This can be abbreviated to > #include "gettext.h" > ... > Here I would parenthesize (Errno). Thanks, I installed the patch enclosed below. > It may be faster to use the libc's optimized strchr() function. (Untested.) Thanks. For the typical case I s