Fabian Groffen added the comment:
> Which shell doesn't have $(command) support? It is not a bash-ism, but
> is part of the POSIX shell definition (but wasn't present in older sh
> implementations).
Solaris' /bin/sh.
--
_
Fabian Groffen added the comment:
re: single quotes -> double quotes
I made RUNSHARED consistent (although, as you point out, less broken) with the
other RUNSHARED assignments right above. If suggest to tackle the issue of
whitespace support for all RUNSHARED assignments, not just Dar
New submission from Fabian Groffen:
Python fails to run python.exe on OSX when a non-framework build is requested,
due to wrong quoting in configure.ac.
Like all other RUNSHARED, it shouldn't be quoted because then `pwd` won't be
expanded, resulting in a wrong DYLD_LIBRARY_PATH whe
Fabian Groffen added the comment:
I don't want to go through the paperwork nonsense just for a trivial patch,
hence I didn't supply one, but instead provided all the information for you
guys to make the correct fix.
--
___
Python trac
Fabian Groffen added the comment:
it seems errnomodule.c has no idea of ENOTSUP, and that's not the only missing
one.
OSX 10.7:
$ grep "^#define\sE" /usr/include/sys/errno.h | awk '{print $2}' | while read
line ; do grep -q ${line} Modules/errnomodule.c || ech
Fabian Groffen added the comment:
% echo "test" > /var/tmp/testfile
% python
Python 2.7.3 (default, Apr 26 2012, 19:06:37)
[GCC 4.2.1 (Gentoo 4.2.1_p5666, Apple Inc. build 5666) (dot 3)] on darwin
Type "help", "copyright", "credits" or "
Fabian Groffen added the comment:
> I presume Python 3.2+ have the same behavior?
I cannot compile that or get it working normally, so I can't tell for sure.
Judging from the code, I'd say yes.
--
___
Python tracker
<http:/
New submission from Fabian Groffen :
With current working dir an NFS-mounted ZFS share, and /var/tmp (OSX default)
HFS+:
% echo "test" > /var/tmp/testfile
% python
Python 2.7.3 (default, Apr 24 2012, 19:33:45)
[GCC 4.2.1 (Gentoo 4.2.1_p5666, Apple Inc. build 5666) (dot 3)] o
Fabian Groffen added the comment:
I recompiled openssl with noasm flag now for sparcv9 (like I do on 32-bits
compiles), and the example runs fine for me on the T1, proving Python is not to
blame here.
That said, if your problem only occurs with 0.9.8 isn't it then likely that
that ve
Fabian Groffen added the comment:
Sorry for the noise. My issue seems to be caused by a broken openssl. It is a
specific problem with the code for sparcv9 and the T1 CPU. Python can't do
much about this.
--
versions: -Python 2.7
___
P
Fabian Groffen added the comment:
Perhaps I should have been a bit more clear. Python 2.6 works fine for me on
Solaris 10/Sparc64 (64-bits). Python 2.7.1 also works on Solaris 10/Sparc
(32-bits), but not on Sparc64 (64-bits).
--
___
Python
Fabian Groffen added the comment:
I can reproduce this on Solaris 10/Sparc64 only, using Python 2.7.1. My core
files aren't really useful though, nothing it can point to, everything is
corrupt.
--
nosy: +grobian
versions: +Python 2.7
___
P
Fabian Groffen added the comment:
Is it necessary to file a new bug for the same issue, or can we reopen from
here? With Stuart in the loop it looks to me the raised issues are dealt with?
--
___
Python tracker
<http://bugs.python.org/issue7
Fabian Groffen added the comment:
I've asked Stuart to sign up and comment on this bug to get his opinion. Sorry.
--
___
Python tracker
<http://bugs.python.org/i
Fabian Groffen added the comment:
still applies to 2.7
--
nosy: +grobian
versions: +Python 2.7
___
Python tracker
<http://bugs.python.org/issue6308>
___
___
Pytho
Fabian Groffen added the comment:
I think so. From my experience he's mainly not willing to sign up to each and
every bugtracking system, which I fully understand.
Do you guys need signed contracts for each and every (even single line) patch?
Is it better to report the problem and h
Changes by Fabian Groffen :
Added file: http://bugs.python.org/file15920/python-2.6.2-solaris64-crypt.patch
___
Python tracker
<http://bugs.python.org/issue1471
Fabian Groffen added the comment:
I use this patch, which just always uses crypt_i on Solaris to work around the
problem.
--
keywords: +patch
nosy: +grobian
title: Python libcrypt build problem on Solaris 8 -> Python libcrypt build
problem on Solaris 8, 9, 10 and OpenSolaris
versi
New submission from Fabian Groffen :
NFS on certain platforms (most notably AIX, Solaris) use .nfsX files that
appear and disappear automagically. distutils can get confused by that once a
.nfsX file was earlier seen with listdir and then removed by the OS before
its copied.
Simply
New submission from Fabian Groffen :
Create a libpythonX.Y.so library on IRIX. Patch on behalf of Stuart Shelton.
--
components: Build
files: python-2.6-irix-libpython2.6.patch
keywords: patch
messages: 97897
nosy: grobian
severity: normal
status: open
title: Build shared
Changes by Fabian Groffen :
Added file: http://bugs.python.org/file15916/python-2.6.4-irix.patch
___
Python tracker
<http://bugs.python.org/issue7717>
___
___
Python-bug
New submission from Fabian Groffen :
Patches to fix compilation issues on IRIX, on behalf of Frank Everdij and
Stuart Shelton.
--
components: Build
files: python-2.7-irix.patch
keywords: patch
messages: 97896
nosy: grobian
severity: normal
status: open
title: Compilation fixes for IRIX
New submission from Fabian Groffen :
The solaris case implementation of the ipv6 check assumes /usr/xpg4/bin/grep is
available. This gives error messages on e.g. IRIX where /etc/netconfig is
available, but no /usr/xpg4/bin/grep. This is simply fixed by using $GREP
which is set by configure
New submission from Fabian Groffen :
Configure will die on Darwin with the message
Unexpected output of 'arch' on OSX
when GNU arch is in use. The following patch simply adds the output as given
by GNU arch to allow a successful configure run.
--
components: Build
fil
New submission from Fabian Groffen :
configure.in contains a check for the compiler in use for Darwin like this:
gcc_version=`gcc -v 2>&1 | grep version | cut -d\ -f3`
This yields in a wrong answer if the output of gcc -v has "version" in another
place as well, such as when
New submission from Fabian Groffen :
setup.py adds in various places hardcoded paths to locations in the filesystem
that may contain libraries/includes to compile a given module or feature.
While this behaviour is probably interesting for some users, it is undesirable
for distributions that
26 matches
Mail list logo
