Re: AC_TYPE_LONG_LONG_INT and AC_TYPE_UNSIGNED_LONG_LONG_INT

Hi Paul, On Wednesday, I proposed a patch to disable both 'unsigned long long int' and 'long long int' on Sun C 5.[0-8] in 32-bit mode, instead of only one of them. > 2007-10-24 Bruno Haible <[EMAIL PROTECTED]> > > * m4/longlong.m4 (_AC_TYPE_LONG_LONG_SNIPPET): New macro, extracted >

borrow AC_C_RESTRICT from autoconf

Hi, The autoconf macro for 'restrict', as modified in fixes problems with mixed use of C and C++. gettext needs it, maybe others too. I'm therefore adding an override to gnulib. We need can remove it once autoconf 2.62 can be

Re: 'strerror' test fails on HP-UX 11.00

Eric Blake wrote: > Yes, since the Austin group discussed rather recently that returning the > empty string is just as broken as returning NULL. OK, applied. > Does this need any m4 touchups? No, strerror.m4 already tests against an empty string. Bruno

Re: quotation conventions

Karl Berry wrote: > When cutting and pasting such things, quotes just get in the way. If you're using xterm, you can fix its copying-extents determination by adding a line like this to your ~/.Xdefaults: XTerm*charClass:43:48,45-47:48,126:48 Bruno

Re: module suggestion: realpath

Sam Steingold wrote: > CLISP comes with a replacement realpath implementation for platforms > which lacks it (are there still such platforms?). If someone provides a correct implementation of realpath for gnulib, we will accept it. But it's better to use canonicalize_file_name(), because - real

Re: [PATCH] Added note for translators to gai_strerror.c, so they don't translate param names.

invalid directory: '%HOME%' notinvalid directory: %HOME% I am not aware that this is any particular rule for GNU, or anything else. Personally I find it much more convenient to have such filenames included without the quotes. And I think that's what most error msgs (every one

Re: getopt: How to know whether to pass char ** or char *const*?

-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 Bruno Haible wrote: > Micah Cowan wrote: >> How best do I determine in my code whether I'm using gnulib's >> getopt_long, or the system library (potentially glibc)'s? That is, I >> want to know whether I ought to be passing in a char ** or a >> char*

Re: bootstrap sync from coreutils

"Sergey Poznyakoff" <[EMAIL PROTECTED]> wrote: > Jim Meyering <[EMAIL PROTECTED]> ha escrit: > >> +* build-aux/bootstrap: Sync from coreutils: >> +2007-10-24 Jim Meyering <[EMAIL PROTECTED]> >> +Get gnulib from the git repository, not from an obsolete cvs one. Thanks. I've just sync'

Re: bootstrap sync from coreutils

Jim Meyering <[EMAIL PROTECTED]> ha escrit: > + * build-aux/bootstrap: Sync from coreutils: > + 2007-10-24 Jim Meyering <[EMAIL PROTECTED]> > + Get gnulib from the git repository, not from an obsolete cvs one. There are some cvs leftovers, though: diff --git a/build-aux/bootstrap b

Re: module suggestion: realpath

On 10/26/07, Ben Pfaff <[EMAIL PROTECTED]> wrote: > > ... realpath implementation for platforms > > which lacks it (are there still such platforms?). > > gnulib has an implementation of canonicalize_file_name. Is that > sufficient? 1. Normally, long names don't bother me too much. Nevertheless

Re: module suggestion: realpath

Sam Steingold <[EMAIL PROTECTED]> writes: > CLISP comes with a replacement realpath implementation for platforms > which lacks it (are there still such platforms?). gnulib has an implementation of canonicalize_file_name. Is that sufficient? -- "Unix... is not so much a product as it is a pains

module suggestion: realpath

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 CLISP comes with a replacement realpath implementation for platforms which lacks it (are there still such platforms?). I think gnulib is a good home for the code (or, rather, CLISP is not). Sam. -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/

Re: 'strerror' test fails on HP-UX 11.00

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Bruno Haible on 10/26/2007 4:08 AM: Hi Bruno, > > This patch fixes it for me. OK to commit? > > > 2007-10-26 Bruno Haible <[EMAIL PROTECTED]> > > * lib/strerror.c (rpl_strerror): Return "Unknown error ..." also if > str

'strerror' test fails on HP-UX 11.00

Hi Eric, Seen on HP-UX 11.00: test-strerror.c:52: assertion failed sh[9]: 4155 Abort FAIL: test-strerror In config.status: GNULIB_STRERROR = 1 HAVE_DECL_STRERROR = 1 REPLACE_STRERROR = 1 In config.h: /* Define to 1 if you have the declaration of `strerror', and to 0 if you #define HAVE_DE

Re: getopt: How to know whether to pass char ** or char *const*?

Hi, Micah Cowan wrote: > How best do I determine in my code whether I'm using gnulib's > getopt_long, or the system library (potentially glibc)'s? That is, I > want to know whether I ought to be passing in a char ** or a > char*const*, whilst avoiding warnings and the like. I would pass a 'char *