Re: new modules for grapheme cluster breaking

Bruno Haible writes: > Wow! New modules in such quality, 99% perfect, this is rare. > > Please squeeze in the attached patch ("git commit --amend"), > then you can push. Thank you. Done. (I also had to convert GraphemeBreakTest.txt from DOS to Unix line-ends and remove trailing white space, b

Re: linkat, LINK_FOLLOWS_SYMLINKS, and Solaris

On 12/27/2010 05:03 AM, Bruno Haible wrote: > Your proposed answer "don't do that" would imply that every library > is distributed in different variants, one for each standards compliance. No, I was thinking more along the lines "don't use -xc99=all". It's clearly a problem to do that, or anything

Re: linkat, LINK_FOLLOWS_SYMLINKS, and Solaris

On 12/27/2010 05:20 AM, Bruno Haible wrote: > OK, to reduce the runtime cost, here's a new proposed patch. It passes the > tests on Solaris 10, with and without "-xc99=all". Yes, thanks, that fixes the performance problem I was worried about, and it makes sense to install that patch regardless of

Re: POSIX replament functions on HP-NonStop

Joachim Schmitz wrote: > OK, thanks. > > The alignof problem is still not solved, but I managed to run a 'make check' > and here's where it fails: Good! At least the entire testdir now compiled without errors. > FAIL: test-alignof ignore for now I'm out o

Re: linkat on Solaris 11

> > Relaxing the testsuite to permit ENOTDIR is fine with me (and certainly > > cheaper). No need to make linkat() give a specific errno unless we can > > prove it helps other code to have that exact failure. > > Adjusting only the test sounds best to me, too. OK, I pushed the test change. Brun

Re: NFS timestamps

On 12/27/2010 06:20 AM, Bruno Haible wrote: > If I remove the "&& !HAVE_WORKING_UTIMES" from the condition, the three > tests pass, except for a small part with symlinks that is skipped: Thanks for trying it out, and for going the extra mile and fixing that bug. I installed this: >From dd4243aa

Re: linkat on Solaris 11

On 12/27/2010 10:50 AM, Jim Meyering wrote: > Thanks for testing on Solaris 11 and fixing that, Bruno. > Adjusting only the test sounds best to me, too. Likewise (both for the thanks, and for adjusting only the test).

Re: linkat on Solaris 11

Eric Blake wrote: > On 12/27/2010 11:05 AM, Bruno Haible wrote: >> Hi Eric, Jim, Paul, >> >> On Solaris 11 from 2010-11, I get this test failure: >> >> test-linkat.c:194: assertion failed >> /bin/sh: line 10: 4267: Abort(coredump) >> FAIL: test-linkat >> >> The reason is that this expression

Re: linkat on Solaris 11

On 12/27/2010 11:05 AM, Bruno Haible wrote: > Hi Eric, Jim, Paul, > > On Solaris 11 from 2010-11, I get this test failure: > > test-linkat.c:194: assertion failed > /bin/sh: line 10: 4267: Abort(coredump) > FAIL: test-linkat > > The reason is that this expression > linkat (dfd, BASE "lin

Re: [PATCH 0/7] contents of topic/libposix for merge to master

Hi Gary, > >     sparc Solaris 7   Sun Studio 8   (mbsnrtowcs2.sh, mbsrtowcs2.sh, > >                                       poll, ptsname) > >                       gcc 4.2.4      (pt_chown link failed: > >                                       can't open pt_chown.o) > >     sparc Solaris 8   gcc

linkat on Solaris 11

Hi Eric, Jim, Paul, On Solaris 11 from 2010-11, I get this test failure: test-linkat.c:194: assertion failed /bin/sh: line 10: 4267: Abort(coredump) FAIL: test-linkat The reason is that this expression linkat (dfd, BASE "link1", dfd, BASE "sub1/", 0) fails with error ENOTDIR, not EEXIST.

Re: more select-tests updates

Hi Paolo, Thanks for the review. I pushed the 4 patches. Bruno

Re: more select-tests updates

On 12/25/2010 01:49 AM, Bruno Haible wrote: 1) The code for pipe() on Win32 is now in module 'pipe-posix'. And 'pipe-posix' depends on 'unistd'. Ok. 2) The arguments passed to function connect_to_socket() are true and false. So it's actually a 'bool' argument. Ok. 3) Because at least some

Re: determining the program_invocation_name

On 12/24/2010 02:41 PM, Bruno Haible wrote: > Hi, > > How can some library code determine the name of the running program, > for error message and display purposes, if the program's main() function > has not stored argv[0] in a particular place? > > Let's be clear about two things: > > 1) This

Re: new modules for grapheme cluster breaking

Hello Ben, Wow! New modules in such quality, 99% perfect, this is rare. Please squeeze in the attached patch ("git commit --amend"), then you can push. The patch contains: - Reordering: In my feeling, grapheme break facilities should come after line break and word break, because they are le

Re: socket functions on HP-NonStop

Joachim Schmitz wrote: > OK, fine. Can I get a new testdir, please? You fine a new one at http://www.haible.de/bruno/gnu/testdir-posix.tar.gz Bruno

Re: linkat, LINK_FOLLOWS_SYMLINKS, and Solaris

On Mon, 27 Dec 2010, Bruno Haible wrote: This is not realistic: People are not distributing libraries in this way, and are not even aware for which standard a library was built and tested for. ("file libfoo.so" does not tell. You need "nm libfoo.so | grep values".) Since it seems that evidence

Re: NFS timestamps

Hi Paul, > I just now pushed the following to try to fix the main problem. > Can you please try this on your host? The same test failures persist: test-utimens.h:101: assertion failed FAIL: test-fdutimensat test-futimens.h:108: assertion failed FAIL: test-futimens test-utimens.h:101:

mbrtowc on Solaris 7

On Solaris 7, I'm seeing test failures: test-mbsnrtowcs.c:162: assertion failed Abort - core dumped FAIL: test-mbsnrtowcs2.sh test-mbsrtowcs.c:162: assertion failed Abort - core dumped FAIL: test-mbsrtowcs2.sh The reason turns out to be that mbrtowc() does not support a pwc = NULL argument. This

Re: linkat, LINK_FOLLOWS_SYMLINKS, and Solaris

[dropped autoconf and libtool mailing list CCs] Paul Eggert wrote: > the proposed change would appear to place a > significant performance penalty for the (presumably more common) case > of compiling and linking in the default mode.  I would suggest something > like the following patch instead, wi

Re: linkat, LINK_FOLLOWS_SYMLINKS, and Solaris

Paul Eggert wrote: > Given the other problems that ensue on Solaris when one compiles and > links to different standards, the simplest answer may be just "don't > do that". It's not just the __xpg4 and __xpg6 stuff; it's also the > _lib_version stuff: scanf behaves differently depending on which >

RE: socket functions on HP-NonStop

OK, fine. Can I get a new testdir, please? -Original Message- From: Bruno Haible [mailto:br...@clisp.org] Sent: Monday, December 27, 2010 1:32 AM To: Joachim Schmitz Cc: bug-gnulib@gnu.org Subject: Re: socket functions on HP-NonStop Joachim Schmitz wrote: > This seems to be indeed new an

RE: socket functions on HP-NonStop

I did, in another message, didn't I. Anyway, in case it got lost: extern const char *inet_ntop (int af, const void *restrict src, ^ "/usr/local/Floss/gnulib/testdir-posix/gltests/./../gllib/arpa/inet.h", line 129: error(235): declaration is incompatible with

[PATCH] read-file.c: tweak syntax

>From c2f75941ea1a5002bd1a7c5eed1b1d283bb9967d Mon Sep 17 00:00:00 2001 From: Jim Meyering Date: Mon, 27 Dec 2010 11:05:00 +0100 Subject: [PATCH] read-file.c: tweak syntax * lib/read-file.c (fread_file): Remove space after "*" in function definitions. --- ChangeLog |6 ++ lib/rea

Re: times test warnings

Bruno Haible writes: > This patch fixes it. OK to push? Yes, please do. Thanks, /Simon

Re: NFS timestamps

On 12/26/2010 02:46 AM, Bruno Haible wrote: >>> . >> > With your patch, I still get "checking whether the utimes function works... > yes" > and the same failures: > test-futimens.h:108: assertion failed > FAIL: test-futimens