Paul Eggert <[EMAIL PROTECTED]> writes:
> Jim Meyering <[EMAIL PROTECTED]> writes:
>
>> However, a long-term (forward-looking :-) solution should
>> also take into consideration the state of NFSv4 and ZFS.
>
> Yes, I'll see if I can pry free some time to look into that.
Here's a proposed patch to
Hello Bruno,
* Bruno Haible wrote on Tue, Oct 23, 2007 at 12:47:32AM CEST:
>
> > Reversing @LIBICONV@ and $(LDADD) fixes that (and is done everywhere
> > else in the gnulib testsuite).
> >
> > OK to apply?
>
> Yes, please.
Applied, thanks. Also the other LIBINTL patch for the unistdio tests.
> From: Paul Eggert <[EMAIL PROTECTED]>
>
> > I installed the following patch to gnulib. Basically it removes the
> > "line 104" mentioned above. Can you please verify that it fixes your
> > problem? Thanks.
> > [...]
>
>That seems to satisfy it. Thanks.
On the bright side, Sun's Stud
From: Paul Eggert <[EMAIL PROTECTED]>
> Thanks for the details. I guess 'false' is a keyword or predefined
> macro or something?
Keyword, I assume. As a macro, I've seen it only in stdbool.h.
> I installed the following patch to gnulib. Basically it removes the
> "line 104" mentioned above
Eric Blake wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Sam Steingold on 10/22/2007 8:29 AM:
I put some gnulib code into my project (in directory gllib) and now I
_sometimes_ (usually it works) get this error:
/space/sbcl-arch/autobench/+clisp/src/build-aux/missing --run
a
Hello Ralf,
Ralf Wildenhues wrote:
> Hello Jim, all,
>
> on Tru64 4.0D I got this:
>
> | cc -DHAVE_CONFIG_H -DEXEEXT=\"\" -DEXEEXT=\"\" -DNO_XMALLOC -DEXEEXT=\"\"
> -I. -I.. -I../../dummy-0/gllib -I../intl -ieee -g -c -o realloc.o
> ../../dummy-0/gllib/realloc.c
> | cc: Error: ../../dummy-0/
Hello Ralf,
> On FreeBSD 6.2 I get this kind of error for the test-u*-vs(|n)printf1
> tests: (beware, creative mixing of regex and glob syntax above ;-)
In glob syntax: test-u*-vs{,n}printf1
> gcc -std=gnu99 -g -O2 -Wl,--as-needed -o test-u16-vsprintf1
> test-u16-vsprintf1.o ../gllib/libgnu.
Hello Ralf,
> Reversing @LIBICONV@ and $(LDADD) fixes that (and is done everywhere
> else in the gnulib testsuite).
>
> OK to apply?
Yes, please. A ChangeLog entry from 2007-04-12 did the same change already
in some other modules. Somehow I keep confusing LDADD, which should come
first, with LDF
Paul Eggert wrote:
> > * m4/ulonglong.m4 (AC_TYPE_UNSIGNED_LONG_LONG_INT): Use -1ull, not
> > -1u, in preprocessor expression.
>
> Thanks for checking that. I installed that fix into gnulib, and a
> similar patch into autoconf.
Thanks. Now, since this is already the second time that we a
Ralf Wildenhues <[EMAIL PROTECTED]> writes:
> This patch should fix the C89 issues.
Thanks.
> Should I report this upstream?
We should send in a single patch that resolves the C89 issues all at
once, rather than multiple smaller patches. This would require some
work.
Jim Meyering meyering.net> writes:
> Bruno Haible clisp.org> wrote:
>
> Both sound good.
> I'll fix it tomorrow if no one beats me to it.
I did it today, along with my patch to add xvprintf.
--
Eric Blake
"Steven M. Schweda" <[EMAIL PROTECTED]> writes:
>I suppose that if your goal is to reveal fairly exotic compiler
> quirks rather than to test for a usable stdbool.h,
Well, one person's "compiler quirks" are another person's "features"
> cc: Error: ../lib/stdbool.h, line 104: Invalid enum
Hello Paul,
> - if (n == 3)
> + /* On most hosts an ACL is trivial if n == 3, and it cannot be
> + less than 3. On IRIX 6.5 it is also trivial if n == -1.
> + For simplicity and safety, assume the ACL is trivial if n <= 3.
> + Also see file_has_acl.c for som
Hello Ralf,
> 2007-10-22 Ralf Wildenhues <[EMAIL PROTECTED]>
>
> * modules/fstrcmp-tests (test_fstrcmp_LDADD): New, add
> @LIBINTL@ for FreeBSD 6.2, $(LIBTHREAD) for AIX 4.3.3.
This is all right. Please apply.
> I must confess I don't know what I'm doing here, but this patch helpe
Thanks, I installed the following patch to try to clear things up.
Index: gnulib.html
===
RCS file: /web/gnulib/gnulib/gnulib.html,v
retrieving revision 1.11
retrieving revision 1.13
diff -p -u -r1.11 -r1.13
--- gnulib.html 11 Apr 200
Hello Ralf,
> This fixes link failures a couple of systems (AIX 4.3.3, IRIX 6.5),
> compare lib/uniconv/u??-conv-from-enc.c.
>
> 2007-10-22 Ralf Wildenhues <[EMAIL PROTECTED]>
>
> * lib/uniconv/u16-conv-to-enc.c (U_MBLEN): Define.
> * lib/uniconv/u32-conv-to-enc.c (U_MBLEN): Likewi
Bruno Haible <[EMAIL PROTECTED]> wrote:
> Eric Blake wrote:
>> > Any objection to making this module also provide xvprintf and xvfprintf?
>> > M4
>> > has a use of vfprintf, which would be nice to have the wrapper for.
>>
>> As in the following? (Also, it fixes a C89 compiler issue).
>
> Fine w
Hello Jim, all,
on Tru64 4.0D I got this:
| cc -DHAVE_CONFIG_H -DEXEEXT=\"\" -DEXEEXT=\"\" -DNO_XMALLOC -DEXEEXT=\"\" -I.
-I.. -I../../dummy-0/gllib -I../intl -ieee -g -c -o realloc.o
../../dummy-0/gllib/realloc.c
| cc: Error: ../../dummy-0/gllib/realloc.c, line 56: In this statement, a
comm
Eric Blake wrote:
> > Any objection to making this module also provide xvprintf and xvfprintf?
> > M4
> > has a use of vfprintf, which would be nice to have the wrapper for.
>
> As in the following? (Also, it fixes a C89 compiler issue).
Fine with me. Just nitpicking:
- I would call the lo
Jim Meyering <[EMAIL PROTECTED]> writes:
> However, a long-term (forward-looking :-) solution should
> also take into consideration the state of NFSv4 and ZFS.
Yes, I'll see if I can pry free some time to look into that.
> Either is fine with me and will work for that system,
> and you're welcom
On FreeBSD 6.2 I get this kind of error for the test-u*-vs(|n)printf1
tests: (beware, creative mixing of regex and glob syntax above ;-)
gcc -std=gnu99 -g -O2 -Wl,--as-needed -o test-u16-vsprintf1
test-u16-vsprintf1.o ../gllib/libgnu.a /usr/local/lib/libiconv.so -Wl,-rpath
-Wl,/usr/local/lib
* Jim Meyering wrote on Mon, Oct 22, 2007 at 10:55:25PM CEST:
> Ralf Wildenhues <[EMAIL PROTECTED]> wrote:
> > * lib/getloadavg.c
> > [defined (__osf__) && (defined (__alpha) || defined (__alpha__)]:
> > Undef `sys' after including sys/table.h, for Tru64 4.0D.
>
> Looks fine. You're w
Ralf Wildenhues <[EMAIL PROTECTED]> wrote:
> Hello Jim, all,
>
> On Tru64 4.0D (and 5.1), sys/table.h defines sys:
>
> /usr/include/sys/table.h:#define sys si_sys
>
> which make me wonder what the system developers were thinking there:
My! How antisocial of them. :)
> 2007-10-22 Ralf Wildenhu
Eric Blake <[EMAIL PROTECTED]> wrote:
>> Jim Meyering meyering.net> writes:
>> > I've just added a new module:
>> >
>> > http://git.sv.gnu.org/gitweb/?p=gnulib.git;a=commitdiff;h=78da34d40e6a2
>>
>> Any objection to making this module also provide xvprintf and xvfprintf? M4
>> has a use of vf
* Jim Meyering wrote on Mon, Oct 22, 2007 at 10:36:03PM CEST:
> Ralf Wildenhues <[EMAIL PROTECTED]> wrote:
> >
> > OK to apply?
>
> > 2007-10-22 Ralf Wildenhues <[EMAIL PROTECTED]>
> >
> > * tests/test-i-ring.c: Work for C89.
>
> Sure. Thanks.
Done.
> Are you actually building some packa
Hello Bruno, all,
(BTW, this testing is all done with `gnulib --with-tests --test'.)
On FreeBSD 6.2, I get in gltests/:
gcc -std=gnu99 -g -O2 -Wl,--as-needed -o test-fstrcmp test-fstrcmp.o
../gllib/libgnu.a -lm
../gllib/libgnu.a(xalloc-die.o)(.text+0xc): In function `xalloc_die':
../../dummy-
Ralf Wildenhues <[EMAIL PROTECTED]> wrote:
> Hello Jim, all,
>
> OK to apply?
> 2007-10-22 Ralf Wildenhues <[EMAIL PROTECTED]>
>
> * tests/test-i-ring.c: Work for C89.
Sure. Thanks.
Are you actually building some package with a C89-only compiler?
Hello Bruno, all,
On FreeBSD 6.2, test-u16-conv-from-enc fails like this:
$ make test-u16-conv-from-enc
gcc -std=gnu99 -g -O2 -Wl,--as-needed -o test-u16-conv-from-enc
test-u16-conv-from-enc.o /usr/local/lib/libiconv.so -Wl,-rpath
-Wl,/usr/local/lib ../gllib/libgnu.a -lm
$ ./test-u16-conv-fro
Hello Jim, all,
On Tru64 4.0D (and 5.1), sys/table.h defines sys:
/usr/include/sys/table.h:#define sys si_sys
which make me wonder what the system developers were thinking there:
| cc -DHAVE_CONFIG_H -DEXEEXT=\"\" -DEXEEXT=\"\" -DNO_XMALLOC -DEXEEXT=\"\" -I.
-I.. -I../../dummy-0/gllib -I../
Hello Bruno, all,
This fixes link failures a couple of systems (AIX 4.3.3, IRIX 6.5),
compare lib/uniconv/u??-conv-from-enc.c.
OK to apply?
Cheers,
Ralf
2007-10-22 Ralf Wildenhues <[EMAIL PROTECTED]>
* lib/uniconv/u16-conv-to-enc.c (U_MBLEN): Define.
* lib/uniconv/u32-conv-to
Hello Paul,
* Paul Eggert wrote on Wed, Oct 17, 2007 at 01:41:21AM CEST:
>
> That bug isn't in the libc version. At this point it's probably better
> to sync the gnulib version from libc, so I installed this:
>
> 2007-10-16 Paul Eggert <[EMAIL PROTECTED]>
>
> Merge glibc changes into l
Hello Jim, all,
OK to apply?
Cheers,
Ralf
2007-10-22 Ralf Wildenhues <[EMAIL PROTECTED]>
* tests/test-i-ring.c: Work for C89.
diff --git a/tests/test-i-ring.c b/tests/test-i-ring.c
index 0e0691a..a6aeb2b 100644
--- a/tests/test-i-ring.c
+++ b/tests/test-i-ring.c
@@ -37,9 +37,10 @@
i
Eric Blake byu.net> writes:
>
> Jim Meyering meyering.net> writes:
>
> >
> > I've just added a new module:
> >
> > http://git.sv.gnu.org/gitweb/?p=gnulib.git;a=commitdiff;h=78da34d40e6a2
>
> Any objection to making this module also provide xvprintf and xvfprintf? M4
> has a use of vfp
Jim Meyering meyering.net> writes:
>
> I've just added a new module:
>
> http://git.sv.gnu.org/gitweb/?p=gnulib.git;a=commitdiff;h=78da34d40e6a2
Any objection to making this module also provide xvprintf and xvfprintf? M4
has a use of vfprintf, which would be nice to have the wrapper for.
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Sam Steingold on 10/22/2007 8:29 AM:
> I put some gnulib code into my project (in directory gllib) and now I
> _sometimes_ (usually it works) get this error:
>
> /space/sbcl-arch/autobench/+clisp/src/build-aux/missing --run
> automake-1.9
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
Simon Josefsson wrote:
> Micah Cowan <[EMAIL PROTECTED]> writes:
>
>>> The second set of patches is against gnulib to remove most of the rules
>>> in coreutils, to avoid duplication. This removes some features from
>>> gnulib's maint.mk, but has an
Micah Cowan <[EMAIL PROTECTED]> writes:
> Simon Josefsson wrote:
>>> Sadly, old versions of gzip send the help/usage report to stderr, not
>>> stdout, so ">/dev/null" alone does not discard it. (And grep wouldn't
>>> see an "rsyncable", even if it were there.) Folding stderr into stdout
>>> look
Micah Cowan <[EMAIL PROTECTED]> writes:
>> The second set of patches is against gnulib to remove most of the rules
>> in coreutils, to avoid duplication. This removes some features from
>> gnulib's maint.mk, but has anyone been using them heavily? If so, we
>> can add them again by making small
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
Simon Josefsson wrote:
>> Sadly, old versions of gzip send the help/usage report to stderr, not
>> stdout, so ">/dev/null" alone does not discard it. (And grep wouldn't
>> see an "rsyncable", even if it were there.) Folding stderr into stdout
>> lo
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
Simon Josefsson wrote:
> Simon Josefsson <[EMAIL PROTECTED]> writes:
>
>> If we make maint.mk minimal, it would be excellent to make coreutils'
>> Makefile.maint do 'include maint.mk'. This was the idea all along.
>> I'll suggest patches for gnulib
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
I could've sworn the gnulib project page on Savannah had git cloning
information just a couple days ago
(https://savannah.gnu.org/projects/gnulib). At the moment, though, it
appears not to. Or maybe I was just thinking of the "Use git" page.
Interes
Bruno Haible <[EMAIL PROTECTED]> writes:
> 2007-10-22 Bruno Haible <[EMAIL PROTECTED]>
>
> * m4/ulonglong.m4 (AC_TYPE_UNSIGNED_LONG_LONG_INT): Use -1ull, not
> -1u, in preprocessor expression.
Thanks for checking that. I installed that fix into gnulib, and a
similar patch into auto
Bruno Haible clisp.org> writes:
> Thanks for this testing! x86_64 platforms are important nowadays.
Agreed.
> > ./test-yesno: error closing file: Bad file descriptor
> > PASS: test-yesno.sh
>
> Eric?
Fixed as follows:
From: Eric Blake <[EMAIL PROTECTED]>
Date: Mon, 22 Oct 2007 09:39:27 -0600
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
I put some gnulib code into my project (in directory gllib) and now I
_sometimes_ (usually it works) get this error:
make -C gllib
make[1]: Entering directory `/space/sbcl-arch/autobench/+clisp/src/gllib'
cd .. && gmake am--refresh
gmake[2]: Entering
Simon Josefsson <[EMAIL PROTECTED]> writes:
> If we make maint.mk minimal, it would be excellent to make coreutils'
> Makefile.maint do 'include maint.mk'. This was the idea all along.
> I'll suggest patches for gnulib/coreutils to achieve this.
Below are two set of patches. The first one is ag
Paul Eggert wrote:
> On your host, 64-bit signed preprocessor numbers work,
> but the unsigned ones don't.
Yes, it looks like this, assuming that the autoconf tests are correct.
> Can you please try this code:
>
> #if ! (18446744073709551615ULL <= -1u)
> error in preprocessor;
> #endif
>
> an
Jim Meyering <[EMAIL PROTECTED]> writes:
> Bruno Haible <[EMAIL PROTECTED]> wrote:
>> Simon Josefsson wrote:
>>> Fixed in gnulib's maint.mk, thanks!
>>
>> Still, we have code duplication between gnulib and coreutils here, which
>> leads to the same bugs being reported twice, on different mailing l
Bruno Haible <[EMAIL PROTECTED]> wrote:
> Simon Josefsson wrote:
>> Fixed in gnulib's maint.mk, thanks!
>
> Still, we have code duplication between gnulib and coreutils here, which
> leads to the same bugs being reported twice, on different mailing lists.
> Since you created maint.mk from coreutils
Simon Josefsson wrote:
> Fixed in gnulib's maint.mk, thanks!
Still, we have code duplication between gnulib and coreutils here, which
leads to the same bugs being reported twice, on different mailing lists.
Since you created maint.mk from coreutils' Makefile.maint, can you can Jim
look how to redu
Libgcrypt 1.3.0 supports the Camellia cipher, that GnuTLS can use, so I
installed a gnulib module for gc-* to use it.
http://git.savannah.gnu.org/gitweb/?p=gnulib.git;a=commitdiff;h=2d6501c921265a1536ca044f45773e761b639706
/Simon
Paul Eggert <[EMAIL PROTECTED]> wrote:
> Bruno Haible <[EMAIL PROTECTED]> writes:
>
>> - if (n == 3)
>> + /* If the ACL is trivial, there is no point in complaining that
>> + we cannot set it on DEST.
>> + An ACL is trivial if it has only the user/group/other entries,
>> +
51 matches
Mail list logo
