Re: pwrite module

2010-05-10 Thread Peter O'Gorman
On Tue, May 11, 2010 at 12:57:20AM +0200, Bruno Haible wrote: > Hi Peter, > > > Attached. Also noticed that I declare read_buf, but never use it > > Thanks, I applied it, only fixing up the ChangeLog entries (wrong file names). Oh, bah! Thank you! > > > (good job, really, as it's one byte too

Re: pwrite module

2010-05-10 Thread Bruno Haible
Hi Peter, > Attached. Also noticed that I declare read_buf, but never use it Thanks, I applied it, only fixing up the ChangeLog entries (wrong file names). > (good job, really, as it's one byte too short). Huh? No, it had the right size. When you call read() with argument N, the buffer has to b

Re: libvirt vs. gettext-0.18

2010-05-10 Thread Bruno Haible
Jim Meyering wrote: > I've just realized that making all libvirt developers install > gettext-0.18 will not be an option, so libvirt will have to > be careful, somehow, not to pull in this new requirement. > libvirt currently requires only gettext-0.14.1. This can be achieved if libvirt does not u

libvirt vs. gettext-0.18 [Re: gettext: update to 0.18

2010-05-10 Thread Jim Meyering
Jim Meyering wrote: > Bruno Haible wrote: >> gettext 0.18 will be released in a couple of hours. This patch upgrades the >> gnulib macros and module. >> >> 2010-05-09 Bruno Haible >> >> Update to GNU gettext 0.18. >> * m4/gettext.m4: Update to GNU gettext 0.17. >> * m4/intl.m4: L

Re: init.sh: enable MALLOC_PERTURB_

2010-05-10 Thread Jim Meyering
Peter O'Gorman wrote: > On Sun, May 09, 2010 at 10:47:07PM +0200, Jim Meyering wrote: > >> +# Enable glibc's malloc-perturbing option. >> +# This is cheap and useful for exposing code that depends on the fact that >> +# malloc-related functions often return memory that is mostly zeroed. >> +# If y

Re: init.sh: enable MALLOC_PERTURB_

2010-05-10 Thread Jim Meyering
Peter O'Gorman wrote: > On Sun, May 09, 2010 at 10:47:07PM +0200, Jim Meyering wrote: > >> +# Enable glibc's malloc-perturbing option. >> +# This is cheap and useful for exposing code that depends on the fact that >> +# malloc-related functions often return memory that is mostly zeroed. >> +# If yo

Re: pwrite module

2010-05-10 Thread Peter O'Gorman
On Sat, May 08, 2010 at 10:04:44AM +0200, Bruno Haible wrote: > Peter O'Gorman wrote: > > The Written Word, Inc. already has an assignment for gnulib. Look for > > 'GNULIB The Written Word' in copyright.list. > > Yes, indeed. It wasn't clear to me whether the patch was owned by you or > by The Wr

Re: init.sh: enable MALLOC_PERTURB_

2010-05-10 Thread Peter O'Gorman
On Sun, May 09, 2010 at 10:47:07PM +0200, Jim Meyering wrote: > +# Enable glibc's malloc-perturbing option. > +# This is cheap and useful for exposing code that depends on the fact that > +# malloc-related functions often return memory that is mostly zeroed. > +# If you have the time and cycles, u

Re: [PATCH] Do not limit git-version-gen to work only under the repository root.

2010-05-10 Thread Giuseppe Scrivano
Bruno, Jim, thanks for your comments. Jim Meyering writes: > What is your use case? I have the git repository in myserver/, but the top_srcdir is in myserver/myserver/. > Won't this proposal cause trouble when > running git-version-gen in a non-version-controlled > pkg-top-level-directory t

Re: gettext: update to 0.18

2010-05-10 Thread Jim Meyering
Bruno Haible wrote: > gettext 0.18 will be released in a couple of hours. This patch upgrades the > gnulib macros and module. > > 2010-05-09 Bruno Haible > > Update to GNU gettext 0.18. > * m4/gettext.m4: Update to GNU gettext 0.17. > * m4/intl.m4: Likewise. > * m4/po.m4:

Re: [PATCH] Do not limit git-version-gen to work only under the repository root.

2010-05-10 Thread Jim Meyering
Giuseppe Scrivano wrote: > I would like to use git-version-gen from a directory that is not the > repository root, so "test -d .git" is not the proper check. > > What about this patch? Hi Giuseppe, What is your use case? Won't this proposal cause trouble when running git-version-gen in a non-ver

Re: [PATCH] Do not limit git-version-gen to work only under the repository root.

2010-05-10 Thread Bruno Haible
Hello Giuseppe, > +elif git describe > /dev/null 2>&1 \ If 'git' is not found in PATH, this line would yield an error message in some shells. To avoid that, run the command in a sub-shell: elif (git describe) > /dev/null 2>&1 \ > * build-aux/git-version-gen: Check if it is a git repository usi

[PATCH] Do not limit git-version-gen to work only under the repository root.

2010-05-10 Thread Giuseppe Scrivano
Hello, I would like to use git-version-gen from a directory that is not the repository root, so "test -d .git" is not the proper check. What about this patch? Cheers, Giuseppe >From d059eef04a41ddc8edf62b53ed310359ed1f3a75 Mon Sep 17 00:00:00 2001 From: Giuseppe Scrivano Date: Mon, 10 May 20

Re: [PATCH] maint.mk: prohibit definition of symbols defined by gnulib

2010-05-10 Thread Jim Meyering
Ralf Wildenhues wrote: > * Jim Meyering wrote on Sun, May 09, 2010 at 07:28:17PM CEST: >> It is unlike most other sc_ rules in that it creates a temporary >> file, .re-defmac. However, it does take care to remove it upon >> receipt of a catchable signal. > > This looks a bit odd to me: > >> +.re-d