Jim Meyering writes:
> I can live with the status quo, but it would make -Wvla more useful
> not to put such constructs in commonly-included header files.
FWIW, I find these definitions from gettext.h are rarely needed, so
unless you really need them, considering doing something like this:
http
Jim Meyering writes:
> Eric Blake wrote:
>> On 12/03/2011 09:00 AM, Simon Josefsson wrote:
>>> What does -funit-at-a-time really do? My gcc 4.4 manual says:
>>>
>>> `-funit-at-a-time'
>>> This option is left for compatibility reasons. `-funit-at-a-time'
>>> has no effect, while `-fno-u
Eric Blake writes:
> On 12/03/2011 09:00 AM, Simon Josefsson wrote:
>> What does -funit-at-a-time really do? My gcc 4.4 manual says:
>>
>> `-funit-at-a-time'
>> This option is left for compatibility reasons. `-funit-at-a-time'
>> has no effect, while `-fno-unit-at-a-time' implies
>>
Simon Josefsson wrote:
> Jim Meyering writes:
>> Eric Blake wrote:
>>> On 12/03/2011 09:00 AM, Simon Josefsson wrote:
What does -funit-at-a-time really do? My gcc 4.4 manual says:
`-funit-at-a-time'
This option is left for compatibility reasons. `-funit-at-a-time'
Simon Josefsson wrote:
> Jim Meyering writes:
>
>> I can live with the status quo, but it would make -Wvla more useful
>> not to put such constructs in commonly-included header files.
>
> FWIW, I find these definitions from gettext.h are rarely needed, so
> unless you really need them, considerin
Hi Jim,
On Minix 3.1.8, I get this spurious output from the test-verify.sh test:
mktemp: not found
PASS: test-verify.sh
This patch fixes it. OK to apply?
2011-12-04 Bruno Haible
tests: Avoid spurious error message on platforms without mktemp program.
* tests/init.sh (mk
Jim Meyering wrote:
> >> Do we have a guarantee that that array dimension is reasonable?
> >
> > Yes. While msgctxt and msgid normally rarely exceed 1 KB (because
>
> That sounds like convention. Is there a guarantee?
No, there is no guarantee. It's the programmer's responsibility to pass
only s
Ben Walton wrote:
> --- a/doc/glibc-functions/sethostname.texi
> +++ b/doc/glibc-functions/sethostname.texi
> @@ -2,18 +2,24 @@
> @subsection @code{sethostname}
> @findex sethostname
>
> -Gnulib module: ---
> +Gnulib module: sethostname
>
> Portability problems fixed by Gnulib:
> @itemize
>
Ben Walton wrote:
> Provide a module that tests the functionality of sethostname().
I have verified that the test 2 is skipped for non-root users and succeeds
for root users on Linux/glibc 2.11, MacOS X 10.5, Solaris 11 2010-11,
Minix 3.1.8, and that the hostname is correctly restored when the tes
Excerpts from Bruno Haible's message of Sat Dec 03 08:09:23 -0500 2011:
Hi Bruno,
> Thanks, I'm applying this for you. I trust that you will follow through
> with the paperwork.
Yes, I'll ensure that gets completed.
> I'm applying a couple of followup tweaks.
Thanks for the tweaks...sorry for
Excerpts from Bruno Haible's message of Sat Dec 03 08:50:37 -0500 2011:
Hi Bruno,
> > +#if !HAVE_GETEUID
> > +# define geteuid() ((uid_t) -1)
> > +#endif
>
> Wow! You have thought at many things. You did it better than if I had
> written this test.
I saw a similar trick somewhere else in gnulib
Ben Walton wrote:
> In one of the subsequent tweaks to this test, you restored a space
> between the function name and the emtpy () at the call site for
> geteuid. I realize I let some foo() slip through so it's
> understandable that you were doing a bunch of corrections. Will this
> not break th
Ben Walton wrote:
> > Oops, that looks like a tab again.
>
> I did install a .dir-locals.el on one system and forgot to copy it to
> other places where I was testing this. Would a .dir-locals.el (and
> vim equivalent) file be something that might be included in the
> project? Something that defi
Bruno Haible wrote:
> On Minix 3.1.8, I get this spurious output from the test-verify.sh test:
>
> mktemp: not found
> PASS: test-verify.sh
>
> This patch fixes it. OK to apply?
>
> 2011-12-04 Bruno Haible
>
> tests: Avoid spurious error message on platforms without mktemp program.
>
On 12/04/11 11:57, Jim Meyering wrote:
d=`unset TMPDIR; { mktemp -d -t -p "$destdir_" "$template_"; } 2>/dev/null` \
That should do the same without the cost of a sub-shell.
Or:
d=`
exec 2>/dev/null
unset TMPDIR
mktemp -d -t -p "$destdir_" "$template_"
`
as we discovered re
Jim Meyering wrote:
> How about using { ...; } instead?
>
> d=`unset TMPDIR; { mktemp -d -t -p "$destdir_" "$template_"; } 2>/dev/null`
> \
>
> That should do the same without the cost of a sub-shell.
Yes, this works too. Applied, with this ChangeLog entry:
2011-12-04 Bruno Haible
Let me add support for sethostname() on Windows platforms.
Unlike on Unix, the hostname change becomes effective (outside the
Control Panel) only after a reboot.
2011-12-04 Bruno Haible
sethostname: Port to Windows platforms.
* lib/sethostname.c: Provide an alternate implement
17 matches
Mail list logo
