Re: lib/exclude.c calls towlower() without checking for wideline support

Eric Blake wrote: > This trips a bug in mingw. There, WEOF is defined as 0x, but towlower > (0x) returns 0x77c5. This works around it. 2009-09-08 Bruno Haible Work around towlower, towupper bug on mingw. * lib/wctype.in.h (towlower, towupper) [__MINGW32__]: New r

[PATCH] build: don't try to run autoheader if we don't use it

I wanted this change in vc-dwim, since it doesn't use autoheader: I'd rather skip autoheader than add an unnecessary AC_CONFIG_HEADERS. >From 6588bee3103aceaf840cb073ef5717d85d3e412a Mon Sep 17 00:00:00 2001 From: Jim Meyering Date: Tue, 8 Sep 2009 22:21:59 +0200 Subject: [PATCH] build: don't try

Re: lib/exclude.c calls towlower() without checking for wideline support

Bruno Haible clisp.org> writes: > + /* Check that the tow* functions map WEOF to WEOF. */ > + ASSERT (towlower (e) == e); > + ASSERT (towupper (e) == e); This trips a bug in mingw. There, WEOF is defined as 0x, but towlower (0x) returns 0x77c5. Oddly enough, towupper(0x) wo

Re: avoiding obsolete macros like AC_TRY_RUN

Hello Sergey, do you mind if I apply this patch which touches a file maintained by you? Thanks, Ralf * Ralf Wildenhues wrote on Fri, Sep 04, 2009 at 08:50:11PM CEST: [...] > Sergey Poznyakoff for > m4/argp.m4 [...] > OK to co

Re: new snapshot available: coreutils-7.5.65-61cc6

Jim Meyering meyering.net> writes: > > euidaccess.c is also broken; it refers to AT_EACCESS without using . > > > > + euidaccess: fix compilation error > > + * lib/euidaccess.c (includes): Add , for AT_EACCESS. > > For what system? Cygwin 1.7 has faccessat but (currently) lacks euidaccess.

Re: new snapshot available: coreutils-7.5.65-61cc6

Eric Blake wrote: > Eric Blake byu.net> writes: > >> According to Pádraig Brady on 9/7/2009 6:32 PM: >> > gnulib 52c658e seems to have removed #include "openat.h" from fstatat.c >> > but not replaced it with #include >> >> Yep; and only Solaris had the problem, because that is the only platform >

Re: new snapshot available: coreutils-7.5.65-61cc6

Eric Blake byu.net> writes: > According to Pádraig Brady on 9/7/2009 6:32 PM: > > gnulib 52c658e seems to have removed #include "openat.h" from fstatat.c > > but not replaced it with #include > > Yep; and only Solaris had the problem, because that is the only platform > with broken fstatat (the

Re: mkfifoat, renameat

Eric Blake wrote: > According to Jim Meyering on 9/8/2009 2:43 AM: >>> I suppose there might still be some pathological cases where renameat >>> currently fails but could be made to succeed. For example, >>> renameat(super_deep,a,root,b) will probably fail with ENAMETOOLONG when >> >> Speaking of

Re: relax strchrnul to LGPLv2+?

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Jim Meyering on 9/8/2009 6:32 AM: >>> What do you think about relaxing it? >> Done. > > Thanks! > > As I tried to use that module in libvirt, I now see that its > dependent module, rawmemchr (also LGPL) needs the same treatment. > So, if

Re: relax strchrnul to LGPLv2+?

Eric Blake wrote: > According to Jim Meyering on 9/8/2009 3:17 AM: >> Hi Eric, >> >> I would like to use strchrnul in libvirt, but its license is >> listed at the default "LGPL" (as glibc-derived, I guess?) >> and that conflicts with libvirt's requirement for LGPLv2+. >> >> What do you think about

Re: mkfifoat, renameat

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Jim Meyering on 9/8/2009 2:43 AM: >> I suppose there might still be some pathological cases where renameat >> currently fails but could be made to succeed. For example, >> renameat(super_deep,a,root,b) will probably fail with ENAMETOOLONG

Re: relax strchrnul to LGPLv2+?

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Jim Meyering on 9/8/2009 3:17 AM: > Hi Eric, > > I would like to use strchrnul in libvirt, but its license is > listed at the default "LGPL" (as glibc-derived, I guess?) > and that conflicts with libvirt's requirement for LGPLv2+. > > Wh

Re: mkfifoat, renameat

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Bruno Haible on 9/8/2009 1:50 AM: > Eric Blake wrote: >> --- a/modules/sys_stat >> +++ b/modules/sys_stat >> ... >> + -e 's|@''GNULIB_OPENAT''@|$(GNULIB_OPENAT)|g' \ > > Why this? The @GNULIB_OPENAT@ token is used only in fcnt

relax strchrnul to LGPLv2+?

Hi Eric, I would like to use strchrnul in libvirt, but its license is listed at the default "LGPL" (as glibc-derived, I guess?) and that conflicts with libvirt's requirement for LGPLv2+. What do you think about relaxing it? Jim

Re: mkfifoat, renameat

Eric Blake wrote: > Two more *at functions. mkfifoat is rather simple; the hardest part was > ensuring the test still compiles on mingw. renameat is more complex, > although I intentionally made it call into a helper function, at_func2, so > that I can reuse that function when I do linkat. The r

Re: mkfifoat, renameat

Eric Blake wrote: > --- a/modules/sys_stat > +++ b/modules/sys_stat > ... > +         -e 's|@''GNULIB_OPENAT''@|$(GNULIB_OPENAT)|g' \ Why this? The @GNULIB_OPENAT@ token is used only in fcntl.in.h, not in sys_stat.in.h. Bruno