Bruno Haible <[EMAIL PROTECTED]> writes:
>> One idea would be a gnulib module to implement a replacement
>> getservbyname function that supports all the IANA registered names, or
>> at least also supports https.
>
> Why only https then? I would use a reasonably complete list, and precompile
> it u
Simon Josefsson wrote:
> Thanks. It seems sad that https isn't widely supported. I would guess
> that the reason is that they aren't in /etc/services?
Yes. On the IRIX and Solaris machines, the earlier test failures coincide
with the missing lines in /etc/services. On AIX, http succeeded, althou
Jim Meyering wrote:
> That patch looks fine. I confirmed that
> with coreutils it induces only white-space changes in configure.
That's a good check :-) So I applied the patch. (Eric and Paul didn't comment.)
Bruno
Hello,
Elbert Pol wrote:
> >> diff: -: Invalid seek
> >> FAIL: test-argp-2.sh
> >
> > Your 'diff' program can use some modernization. POSIX
> >http://www.opengroup.org/onlinepubs/009695399/utilities/diff.html
> > goes into some length explaining why the input does not need to be seekable.
> I
Bruno Haible wrote:
You can also try a standalone test program, linked against the system's
sprintf(), like this:
===
#include
int main()
{
char buf[100];
sprintf (buf, "%Lf", -0.0L);
printf ("%s\n", buf);
return 0;
}
===
The resu
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 4/13/2008 1:44 PM:
| Hi Eric,
|
| On HP-UX 11, I got these test failures:
|
| test-ftell.c:85: assertion failed
| ./test-ftell.sh[3]: 9644 Abort
| FAIL: test-ftell.sh
|
| In fact, when reading from a regular file, after th
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Jim Meyering on 4/14/2008 2:37 AM:
| Since "abs" truncates any value in (-1..1) to 0, the tests below would
| succeed for any value in a much larger than intended range: (-0.5, 1.5).
|
| The corrected (stricter) tests still pass on at lea
Bruno Haible <[EMAIL PROTECTED]> writes:
> The getaddrinfo test also fails on AIX 4 and 5:
>
> Finding www.ibm.com service https...
> res 7: Host not found
>
> 7 is EAI_NODATA, meaning "no address associated with hostname". But
> when ("www.ibm.com", "https") fails with this error whereas
> ("www.
Jim Meyering wrote:
> I've pushed this change:
>
> test-strtod: fix typos: s/abs/fabs/
> * tests/test-strtod.c (main): Use fabs, not narrowing-to-int "abs".
Should work fine, because on all platforms where the use of fabs requires
linking with -lm the variable POW_LIBM already contain
Since "abs" truncates any value in (-1..1) to 0, the tests below would
succeed for any value in a much larger than intended range: (-0.5, 1.5).
The corrected (stricter) tests still pass on at least a few targets,
so I've pushed this change:
test-strtod: fix typos: s/abs/fabs/
* te
10 matches
Mail list logo
