Following up on Bruno's comments, here's a revised proposed
patch to gnulib, which I've tested with coreutils and with Emacs.
This patch doesn't affect areadlinkat-with-size and areadlink-with-size,
which I'd like to obsolete; that can wait.
diff --git a/ChangeLog b/ChangeLog
index 9f58471..248bf
Hi Bruno,
On Thu, Mar 31, 2011 at 5:53 PM, Bruno Haible wrote:
> Bruce Korb wrote:
>> > rm -f stdlib.h-t stdlib.h && \
>> > { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */' && \
>> > sed -e 's|@''INCLUDE_NEXT''@||g' \
>> > -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g'
Bruce Korb wrote:
> > rm -f stdlib.h-t stdlib.h && \
> > { echo '/* DO NOT EDIT! GENERATED AUTOMATICALLY! */' && \
> > sed -e 's|@''INCLUDE_NEXT''@||g' \
> > -e 's|@''PRAGMA_SYSTEM_HEADER''@|@PRAGMA_SYSTEM_HEADER@|g' \
> > -e 's|@''PRAGMA_COLUMNS''@|@PRAGMA_COLUMNS@|g' \
> >
Hi,
I used gnulib-tool to pull some stuff into a new project directory:
local opts='--symlink --import'
local glt=$(which gnulib-tool)
test -f "${glt}" || {
glt=$(echo ~gnu/proj/gnulib)/gnulib-tool
test -f "${glt}" || die "Please put gnulib into PATH"
}
local
I don't have time to fix this today, but test-getcwd is failing on mingw
when compiler warnings are fatal, since mingw declares getcwd as
(char*,int) rather than the required (char*,size_t). I suppose the
solution is the same as other problematic function signatures, such as
our wrapper around ioc
Actually testing on mingw uncovered a few more problems.
* lib/nonblocking.c (set_nonblocking_flag): Succeed when clearing
non-blocking flag on regular file.
(get_nonblocking_flag): Set errno on invalid fd.
* tests/test-nonblocking.c (main): Avoid test failure on
directories if fchdir is not activ
Hi Simon,
> However if I remove the following from all gnulib-common.m4 (there are
> three of them):
>
> > +AC_DEFUN([gl_ASSERT_NO_GNULIB_POSIXCHECK],
> > +[
> > + dnl Override gl_WARN_ON_USE_PREPARE.
> > + AC_DEFUN([gl_WARN_ON_USE_PREPARE], [])
> > +])
>
> Then building works f
Paul Eggert writes:
>diff --git a/m4/gnulib-common.m4 b/m4/gnulib-common.m4
>index c4f41f1..05a66f3 100644
>--- a/m4/gnulib-common.m4
>+++ b/m4/gnulib-common.m4
>@@ -102,6 +102,16 @@ AC_DEFUN([gl_MODULE_INDICATOR_FOR_TESTS],
> [Define to 1 when the gnulib module $1 shoul
On 31/03/11 15:20, Eric Blake wrote:
> Sounds to me like the gnulib fallback should be made smarter. Which
> systems lack linkat() but have the capability to set timestamps? BSD?
> But the original report was about opensuse, which is Linux based, and my
> recollection is that Linux handles hardli
[adding bug-gnulib]
On 03/31/2011 08:11 AM, Pádraig Brady wrote:
>> So probably -al is broken since you fixed -a in 7.5.
>
> Hmm it looks now like we're creating symlinks (with wrong timestamps),
> but in fact we should be creating hardlinks to symlinks.
>
> This seems to have been changed with:
Eric Blake wrote:
> I've touched up your proposal, and done more testing with this.
Great, thanks!
I've done three more touch-ups:
- In the module description, prefer a lib_SOURCES augmentation to an
AC_LIBOBJ invocation when possible. In this case both are equivalent,
but AC_LIBOBJ has
Hi Eric,
Thanks for taking up and improving on my proposal.
> > - use select() or poll() followed by read() or write() if the
> > descriptor
> >is ready,
> > - call read() or write() in separate threads. */
>
> - use aio_*, although that is not as widely portable
Yes, good p
12 matches
Mail list logo
