Bruno Haible <[EMAIL PROTECTED]> writes:
> + #error "Please port gnulib fseterr.c to your platform! Look at the
> definitions of ferror and clearerr on your system, then report this to
> bug-gnulib."
Thanks! I forgot another reason to use this approach: it allows us to
discover more easily wh
Bruno Haible <[EMAIL PROTECTED]> writes:
> Simon's http://autobuild.josefsson.org/gnulib/ summary page could be more
> useful (and larger :-)) if more people submit test results. To this
> effect, I'm adding a facility that produces an 'autobuild' script in
> "gnulib-tool --create-megatestdir".
N
Bruno Haible <[EMAIL PROTECTED]> writes:
> I'm applying this fix. Hope it's ok with you, Simon.
Yup, thanks!
/Simon
Bruno Haible <[EMAIL PROTECTED]> writes:
> Hi Simon,
>
> The gc-des module contains the lib/des.h header, which include .
> So it should depend on 'stdbool'. Currently, it doesn't, and the compiler
> on Solaris 8 complains:
>
> cc -O -DHAVE_CONFIG_H -I. -I.. -g -c des.c
> "./des.h", line 28:
Bruno Haible <[EMAIL PROTECTED]> writes:
> Should gnulib compile its own inet_ntop() in this situation?
> If yes, you need to add a '#define inet_ntop rpl_inet_ntop' to gnulib's
> arpa/inet.h. If no, the autoconf test should look whether inet_ntop is
> available in -lnsl, and make this additional
Bruno Haible <[EMAIL PROTECTED]> writes:
> Hi Simon,
>
> On OSF/1 5.1, with cc as compiler, the getaddrinfo module fails to compile:
>
> cc: Warning: getaddrinfo.c, line 334: In this declaration, parameter 2 has a
> different type than specified in an earlier declaration of this function.
> (mis
Bruno Haible <[EMAIL PROTECTED]> writes:
> Hi Simon,
>
> On MacOS X 10.3.9 the 'gc-gnulib' (or 'gc-des') module fails to build:
>
> depbase=`echo gc-gnulib.o | sed 's|[^/]*$|.deps/&|;s|\.o$||'`;\
> gcc -std=gnu99 -DHAVE_CONFIG_H -DEXEEXT=\"\" -DEXEEXT=\"\" -DNO_XMALLOC
> -DEXEEXT=\"\" -I. -I.. -
Bruno Haible <[EMAIL PROTECTED]> writes:
> "gnulib-tool --create-megatestdir --dir=..."
Don't forget to add --with-tests too.
Should --create-megatestdir add this automatically?
/Simon
Bruno Haible <[EMAIL PROTECTED]> wrote:
> Hi Jim,
>
> The cycle-check module uses 'inline', therefore should make sure it's
> supported
> or defined away. Currently it leads to a failure like this, on Solaris 2.5.1
> with SunPRO C 5.0:
>
> cc -O -DHAVE_CONFIG_H -I. -I.. -g -c cycle-check.c
>
Bruno Haible <[EMAIL PROTECTED]> wrote:
> The i-ring.h file includes . Therefore it should depend on
> the 'stdbool' module. Currently it gives an error on Solaris 10 for example:
>
> cc -xarch=v9 -O -DHAVE_CONFIG_H -I. -I.. -g -c i-ring.c
> "/usr/include/stdbool.h", line 42: #error: "Use of
I'm looking at Bruno's data structure modules, in particular
linked-list, and consider using it to hold the information related to
all open fd's handled by select() in GnuTLS's gnutls-serv. One
important feature of our current implementation is that I can register
my custom "free" function, which
Simon Josefsson wrote:
> Should --create-megatestdir add this automatically?
I'd say, better keep it optional. Although currently I use --with-tests always,
there are cases when one cares whether the code compiles in the first place,
ignoring the tests.
Bruno
Jim Meyering wrote:
> > * m4/cycle-check.m4 (gl_CYCLE_CHECK): Require AC_C_INLINE.
>
> That's fine, even though I consider such an
> old version of Solaris to be irrelevant.
The same lack of 'inline' is also observed on Solaris 7 and HP-UX 11. [1]
> You're welcome to apply it.
Done.
Bruno
Simon Josefsson wrote:
> I changed the namespace to gl_des as below.
After the complaints "gnulib is changing too fast" a few weeks ago, I think
it may help the gnulib users if we keep track of the incompatible changes
(removed modules, renamed functions etc.). I'm making a start by adding this
NE
FYI, I've just added these two new files.
* build-aux/bootstrap: New file.
* build-aux/bootstrap.conf: New file, from coreutils.
Variants of this bootstrap script are being used in at least 6 or
7 different projects, and having it here seems like the best way
to keep everyone up t
I haven't been paying attention to the *printf discussions, but
looking at current CVS, shouldn't the vasprintf module use stdio.h for
the prototypes?
/Simon
2007-03-12 Simon Josefsson <[EMAIL PROTECTED]>
* modules/vasprintf: Depend on stdio for prototypes, and remove
vasprintf
Le lundi 12 mars 2007 à 12:18 +0100, Simon Josefsson a écrit :
> Bruno Haible <[EMAIL PROTECTED]> writes:
>
> > Should gnulib compile its own inet_ntop() in this situation?
> > If yes, you need to add a '#define inet_ntop rpl_inet_ntop' to gnulib's
> > arpa/inet.h. If no, the autoconf test should
Bruno Haible <[EMAIL PROTECTED]> writes:
> Simon Josefsson wrote:
>> I changed the namespace to gl_des as below.
>
> After the complaints "gnulib is changing too fast" a few weeks ago, I think
> it may help the gnulib users if we keep track of the incompatible changes
> (removed modules, renamed f
Bruno Haible <[EMAIL PROTECTED]> writes:
> The use of this facility is as follows:
> 1) On a machine with recent automake, autoconf, m4 installed and with a
> gnulib cvs checkout (typically a Linux machine), use
> "gnulib-tool --create-megatestdir --dir=..."
> Note: The created
Simon Josefsson wrote:
> > 2) Transfer this directory to a build machine (HP-UX, Cygwin, or
> > whatever).
>
> Step 2 can be simplified by doing:
>
> ./configure
> make dist
>
> since gnulib-tool already create a dummy configure.ac that
> AC_CONFIG_SUBDIR include every gnulib module. Running
Bruno Haible <[EMAIL PROTECTED]> writes:
>> any future modules that I contribute (I have a few in mind)
>
> Of course I'd like to hear details :-) (when you have time for it).
The one that comes to mind immediately is to finish up and check
in the module for the "trunc" and "round" families of fu
Simon Josefsson wrote:
> I haven't been paying attention to the *printf discussions, but
> looking at current CVS, shouldn't the vasprintf module use stdio.h for
> the prototypes?
Yes, you're right. Since, as Paul said, gnulib not only attempts to mimic
POSIX APIs but also glibc APIs, and glibc de
Simon Josefsson wrote:
> Bruno Haible <[EMAIL PROTECTED]> writes:
>
> > According to the autobuild logs [1], the module 'printf-frexp' fails the
> > tests on a Linux/x86 machine of yours. I cannot reproduce this, on a
> > glibc-2.3.6 machine and on a glibc-2.5 machine.
> > ...
> > [1] http://autob
Bruno Haible <[EMAIL PROTECTED]> writes:
> Changing the function definition to use ANSI C argument list syntax (without
> adding 'restrict' there) fixes the problem. Changing the function definition
> by adding 'restrict' in the first and fourth argument (keeping the K&R syntax)
> also fixes the p
Hello Jim,
> Variants of this bootstrap script are being used in at least 6 or
> 7 different projects, and having it here seems like the best way
> to keep everyone up to date. However there's plenty of room for
> improvement.
Well, I'd suggest two improvements:
First, the name. "Bootstrapping"
Bruno Haible <[EMAIL PROTECTED]> writes:
> 2007-03-11 Bruno Haible <[EMAIL PROTECTED]>
>
> * lib/quotearg.c: Include early, before the definition of
> the iswprint macro. Needed on Solaris 2.5.1.
Thanks; please install. (I just now tried to install it but the CVS
server access was
26 matches
Mail list logo
