Re: [bug-libunistring] ‘test-xalloc-die’ failure

2009-11-27 Thread Jim Meyering
Ludovic Courtès wrote: > With the help of Rob Vermaas of NixOS (TU Delft), I’ve added a > libunistring job to Hydra so that libunistring would be continuously > built from Git: > > http://hydra.nixos.org/jobset/gnu/libunistring/all > > Currently, ‘test-xalloc-die’ fails: > > --- - 2009-11-24 09:3

Re: [bug-libunistring] ‘test-xalloc-die’ failure

2009-11-27 Thread Simon Josefsson
Jim Meyering writes: > Simon, here's a proposed fix: > It passes my simple test of running like this (no libtool, though): > > ./gnulib-tool --create-testdir --with-tests --test xalloc-die > > Ok to push? I see no problem with it, please push! /Simon

Re: [PATCH 2/2] test pread (using the new init.sh framework)

2009-11-27 Thread Simon Josefsson
Jim, there seems to be some PATH issue causing build failures: ../../gltests/test-pread.sh: 6: test-pread: not found FAIL: test-pread.sh See: http://autobuild.josefsson.org/gnulib/log-200911270751109732000.txt /Simon

Error handling in Win32 + gnulib

2009-11-27 Thread Richard W.M. Jones
I'm confused by errno and error handling in gnulib, and what the strategy for this is supposed to be, versus how it seems to work now. Win32 has two methods to return an error code from a syscall: DWORD WSAGetLastError(void); // for all socket functions DWORD GetLastError(); // for al

BSD getopt bugs

2009-11-27 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I found a regression in m4 on cygwin 1.7, and traced it to the fact that getopt.m4 no longer rejects BSD getopt even though it still has bugs compared to GNU getopt_long. Cygwin 1.7 is currently using getopt.c with the following header: /* $OpenB

Re: [PATCH 2/2] test pread (using the new init.sh framework)

2009-11-27 Thread Jim Meyering
Simon Josefsson wrote: > Jim, there seems to be some PATH issue causing build failures: > > ../../gltests/test-pread.sh: 6: test-pread: not found > FAIL: test-pread.sh > > See: http://autobuild.josefsson.org/gnulib/log-200911270751109732000.txt Hi Simon, Thanks for the report. How can I reproduce

Re: [PATCH] tests/init.sh: new file to be used via most *.sh tests

2009-11-27 Thread Ralf Wildenhues
Hi Jim, * Jim Meyering wrote on Wed, Nov 25, 2009 at 02:49:59PM CET: > I've pushed tests/init.sh, as yet unused. I haven't looked at this in detail, due to time constraints, so just a couple of notes: Automake has a similar file tests/defs.in that is not as elaborate; still, you might be able to

Re: [PATCH] tests/init.sh: new file to be used via most *.sh tests

2009-11-27 Thread Jim Meyering
Ralf Wildenhues wrote: > * Jim Meyering wrote on Wed, Nov 25, 2009 at 02:49:59PM CET: >> I've pushed tests/init.sh, as yet unused. > > I haven't looked at this in detail, due to time constraints, Hi Ralf, Thanks for the feedback. FYI, the mkdtemp-related code has been in gnulib for some time, Eve

Re: [PATCH] tests/init.sh: new file to be used via most *.sh tests

2009-11-27 Thread Ralf Wildenhues
* Jim Meyering wrote on Fri, Nov 27, 2009 at 10:03:10PM CET: > > so just a couple of notes: Automake has a similar file tests/defs.in > > that is not as elaborate; still, you might be able to profit from it. > > For example, turning on VERBOSE if srcdir is not set but derived from > > $0 is very ha

Re: BSD getopt bugs

2009-11-27 Thread Eric Blake
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Eric Blake on 11/27/2009 9:24 AM: > I found a regression in m4 on cygwin 1.7, and traced it to the fact that > getopt.m4 no longer rejects BSD getopt even though it still has bugs > compared to GNU getopt_long. Committing this. It's rath