Commit Graph

708 Commits

Author SHA1 Message Date
Ronald Oussoren 92397ce93f Explicitly use /usr/bin/arch on OSX, fixes issue 7715 2010-01-17 19:32:00 +00:00
Ronald Oussoren 3c0a1265c1 This patch ensures that the configure-script mentions checking for --enable-universalsdk
and that it doesn't default to the 10.4u SDK when that SDK does not exist.

(This affects OSX)

This patch should fix most of issue 4834, although I haven't gotten enough
information from the user to be sure.
2010-01-17 19:27:57 +00:00
Ronald Oussoren a55af9a9db - Issue #7658: Ensure that the new pythonw executable works on OSX 10.4
- Issue #7714: Use ``gcc -dumpversion`` to detect the version of GCC on
  MacOSX.

- Make configure look for util.h as well as libutil.h. The former
  is the header file that on OSX contains the defition of openpty.

  (Needed to compile for OSX 10.4 on OSX 10.6)

- Use the correct definition of CC to compile the pythonw executable
2010-01-17 16:25:57 +00:00
Benjamin Peterson 1c335e63c9 use pkg-config to find the libffi headers when --with-system-ffi is used #6943 2010-01-01 15:16:29 +00:00
Mark Dickinson 0ef0b910ce More configure fixes: avoid sh 'integer argument expected' error when 'long long' type doesn't exist. 2009-12-31 21:11:48 +00:00
Mark Dickinson 5ce8474cc7 Add missing quotes. 2009-12-31 20:48:04 +00:00
Benjamin Peterson 2c19674b51 add a --with-system-expat option to build pyexpat against the system's lib #7609 2009-12-31 03:17:18 +00:00
Ronald Oussoren 92919a66d2 Issue #6834: replace the implementation for the 'python' and 'pythonw' executables on OSX.
The previous implementation used execv(2) to run the real interpreter, which means that
you cannot use the arch(1) tool to select the architecture you want to use for a
universal build because that only affects the python/pythonw wrapper and not the actual
interpreter.

The new version uses posix_spawnv with a number of OSX-specific options that ensure that
the real interpreter is started using the same CPU architecture as the wrapper, and that
means that 'arch -ppc python' now actually works.

I've also changed the way that the wrapper looks for the framework: it is now linked to
the framework rather than hardcoding the framework path. This should make it easier to
provide pythonw support in tools like virtualenv.
2009-12-24 13:30:58 +00:00
Mark Dickinson 59dc89e87d Issue #7492: Autoconf tests were leaving semaphore files behind. Add
sem_unlink calls to delete those semaphore files.
2009-12-13 21:06:06 +00:00
Benjamin Peterson 91c12ebc3d disable pymalloc tricks with the --with-valgrind option #2422
Patch from James Henstridge.
2009-12-03 02:52:39 +00:00
Antoine Pitrou 30b3b35cba Issue #7333: The `posix` module gains an `initgroups()` function providing
access to the initgroups(3) C library call on Unix systems which implement
it.  Patch by Jean-Paul Calderone.
2009-12-02 20:37:54 +00:00
Mark Dickinson c63392c152 Multiprocessing configure checks don't need LIBM 2009-11-28 13:13:13 +00:00
Mark Dickinson 5afa6d4dcf Issue #7272, continued: don't re-use existing HAVE_BROKEN_POSIX_SEMAPHORES
to indicate that semaphores aren't available;  define a new variable
POSIX_SEMAPHORES_NOT_ENABLED instead.
2009-11-28 10:44:20 +00:00
Martin v. Löwis 50ea4565bd Issue #6508: Add posix.{getresuid,getresgid,setresuid,setresgid}. 2009-11-27 13:56:01 +00:00
Mark Dickinson c4920e86ef Issue #7272: Add configure test to detect whether sem_open works
properly, and use this to skip test_multiprocessing on platforms
where sem_open raises a signal.  This should fix some FreeBSD buildbot
failures for test_multiprocessing.
2009-11-20 19:30:22 +00:00
Ronald Oussoren 315cd0c739 Fix for issue #7085
On MacOSX 10.6 the CoreFoundation framework must be initialized on the main
thread, the constructor function in that framework will cause an SIGABRT when
it is called on any other thread.

Because a number of extension link (indirectly) to CoreFoundation and the
Python core itself didn't the interpreter crashed when importing some
extensions, such as _locale, on a secondary thread.

This fix ensures that Python is linked to CoreFoundation on OSX, which results
in the CoreFoundation constructor being called when Python is loaded. This
does not require code changes.
2009-11-19 16:25:21 +00:00
Mark Dickinson 82864d1ab1 Issue #7228: Add '%lld' and '%llu' support to PyFormat_FromString,
PyFormat_FromStringV and PyErr_Format.
2009-11-15 16:18:58 +00:00
Mark Dickinson a548deeb29 Issue #5792: Extend short float repr support to x86 platforms using suncc or icc.
Many thanks Stefan Krah for help and OpenSolaris testing.
2009-11-15 13:12:43 +00:00
Gregory P. Smith d8cb2d9c59 see issue1006238, this merges in the following patch to ease cross
compiling the printf %zd check.

 http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-lang/python/files/python-2.5-cross-printf.patch?rev=1.1&view=markup
2009-11-02 02:02:38 +00:00
Gregory P. Smith bb21389202 see http://bugs.python.org/issue1006238 this merges in the following
patch to make cross compilation of the chflags check easier:

 http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-lang/python/files/python-2.6-chflags-cross.patch?rev=1.1
2009-11-02 01:37:37 +00:00
Gregory P. Smith ec3b8bd3a9 issue1115: convert some AC_TRY_RUNs into AC_TRY_COMPILEs. 2009-11-01 21:02:52 +00:00
Antoine Pitrou d9ff74e51d Some platforms have rl_completion_append_character but not rl_completion_suppress_append.
Reported by Mark D.
2009-10-26 19:16:46 +00:00
Mark Dickinson 99abd145dc Configure check for double rounding should take BASECFLAGS into account 2009-10-24 13:44:16 +00:00
Mark Dickinson 1d6e2e1833 Issue #7117 (backport py3k float repr) continued:
- add double endianness detection to configure script
 - add configure-time check to see whether we can use inline
   assembly to get and set x87 control word in configure script
 - add functions to get and set x87 control word in Python/pymath.c
 - add pyport.h logic to determine whether it's safe to use the
   short float repr or not
2009-10-24 13:28:38 +00:00
Ronald Oussoren 9ebd242fbb * Update the Mac/README file. Add the list of OSX-specific configure options
to that README file with some explanation.

* Be more strict in the configure script: complain loudly when the user has
  specified invalid combinations of OSX-specific configure arguments.

  The error message refers to the Mac/README file for more information.
2009-09-29 13:00:44 +00:00
Ronald Oussoren a9cfbded4b Fix typo in error message 2009-09-20 20:05:44 +00:00
Ronald Oussoren 84ddd72312 This is an update to r74701. How hard can it be to get a configure test right.
This patch has already been backported as part of the backport of r74701,
which is how I found this problem.
2009-09-08 07:17:10 +00:00
Ronald Oussoren 23d925311d Fix typo in configure.in 2009-09-07 06:12:00 +00:00
Ronald Oussoren 2596758cb4 Fix build issues on OSX 10.6 (issue 6802) 2009-09-06 10:00:26 +00:00
Mark Dickinson 65898e08a7 Add configure-time checks for gamma and error functions. 2009-09-05 10:27:00 +00:00
Georg Brandl fe18a11858 #5047: remove Monterey support from configure. 2009-09-04 07:55:14 +00:00
Alexandre Vassalotti 130ae15183 Add a check to ensure the correct autoconf version is used
for generating the configure script.

Original idea by Martin von Löwis.
2009-07-18 00:31:06 +00:00
Alexandre Vassalotti e7cf1183fe Double-quote the test case for %zd printf() format support to avoid
mangling the array declarations in it.
2009-07-17 06:33:51 +00:00
Alexandre Vassalotti 66da2635ff Clean up the test case for broken poll(). 2009-07-17 06:17:33 +00:00
Alexandre Vassalotti 30267d6516 Double-quote the test cases for chflags() and lchflags()
to ensure they don't get mangled.
2009-07-17 06:10:06 +00:00
Alexandre Vassalotti 0090089f8f Cache the results of all runtime checks.
This will be helpful to people who want to compile Python with a
cross-compiler. Now you can upload the configure script on your host
machine, run it with caching enabled, and download the cached results
on your build machine.
2009-07-17 05:26:39 +00:00
Alexandre Vassalotti 856782e45e Rename the useless AC_INCLUDES_DEFAULT and protect the includes.
This is mostly an aesthetic change.
2009-07-17 04:59:05 +00:00
Alexandre Vassalotti 786c29e39e Use AC_CHECK_SIZEOF to find the size of off_t, pthread_t and time_t. 2009-07-17 04:24:45 +00:00
Ronald Oussoren 450d561028 This is a fix for Issue5809: you shouldn't specify both --enable-framework and
--enable-shared
2009-06-08 21:12:41 +00:00
Antoine Pitrou f2caeed9c6 Issue #3585: Add pkg-config support.
It creates a python-2.7.pc file and a python.pc symlink in the
$(LIBDIR)/pkgconfig directory. Patch by Clinton Roy.
2009-05-24 20:23:57 +00:00
Benjamin Peterson e5afa3b24b support building with subversion 1.7 #6094 2009-05-23 19:24:37 +00:00
Georg Brandl 464432d181 Update bug tracker URL. 2009-05-20 18:24:08 +00:00
Tarek Ziadé e2be83def8 Fixed Issue 5900: distutils.command.build_ext - Ensure RUNPATH is added to extension modules with RPATH if GNU ld is used 2009-05-09 08:28:53 +00:00
Tarek Ziadé 99f660af3f Fixed #5941: added ARFLAGS for the archiver command. 2009-05-07 21:20:34 +00:00
Matthias Klose 61dbdb9677 - configure.in: Don't error, when no --with-dbmliborder option is present 2009-04-29 20:09:50 +00:00
Matthias Klose 51c614e3bf - Issue #4587: Add configure option --with-dbmliborder=db1:db2:... to specify
the order that backends for the dbm extension are checked.
2009-04-29 19:52:49 +00:00
Matthias Klose 10cbe4886e - Issue #4587: Add configure option --with-dbmliborder=db1:db2:... to specify
the order that backends for the dbm extension are checked.
2009-04-29 17:18:19 +00:00
Mark Dickinson 8e5446f902 Backport r71704 (add configure check for C99 round function) to trunk. 2009-04-18 14:41:37 +00:00
Jesse Noller 355b1264b8 issue5545: Switch to Autoconf for multiprocessing; special thanks to Martin Lowis for help 2009-04-02 00:03:28 +00:00
Georg Brandl bcd64a3fc7 #1676135: remove trailing slashes from --prefix argument. 2009-03-31 21:45:18 +00:00
Ronald Oussoren bb8cb0e192 Fix issue where 'make altinstall' or 'make install' would install everything
that needs to be installed on OSX (depending on the configure flags)
2009-03-30 17:53:39 +00:00
Mark Dickinson efc82f7e8e Issue #4258: Use 30-bit digits for Python longs, on 64-bit platforms.
Backport of r70459.
2009-03-20 15:51:55 +00:00
Mark Dickinson 1de3327015 It's wrong to use AC_REPLACE_FUNCS for hypot, since there's no longer any
Python/hypot.c replacement file.  Use AC_CHECK_FUNCS instead.  This change
should be backported to 2.6 and 3.0.
2009-01-04 20:19:41 +00:00
Mark Dickinson d181e3a246 isinf and isnan are macros, not functions; fix configure script
to use AC_CHECK_DECLS instead of AC_CHECK_FUNCS for these.
(See discussion in issue #4506)
2009-01-04 13:57:26 +00:00
Mark Dickinson 04b272336d Add autoconf test to detect x87-style double rounding, as described in
issue #2937.  This information can be helpful for diagnosing platform-
specific problems in math and cmath.  The result of the test also
serves as a fairly reliable indicator of whether the x87 floating-point
instructions (as opposed to SSE2) are in use on Intel x86/x86_64 systems.
2009-01-04 12:29:36 +00:00
Skip Montanaro 516144fef7 If user configures --without-gcc give preference to $CC instead of blindly
assuming the compiler will be "cc".
2009-01-04 10:36:58 +00:00
Ronald Oussoren 79f90498bc Fix for issue4472: "configure --enable-shared doesn't work on OSX" 2009-01-02 10:44:46 +00:00
Skip Montanaro 4d756afd68 typo in comment 2008-12-01 01:55:22 +00:00
Benjamin Peterson 98b1b02fbf note the version that works 2008-11-30 14:43:23 +00:00
Georg Brandl 4b9bcfc98d #4316: fix configure.in markup problem. 2008-11-16 08:00:17 +00:00
Martin v. Löwis bb86d83134 Issue #4204: Fixed module build errors on FreeBSD 4. 2008-11-04 20:40:09 +00:00
Martin v. Löwis 174440b5ec Bump version to 2.7. Regenerate. 2008-10-03 08:59:41 +00:00
Gregory P. Smith 29ec750a86 bugfix to r66283 (see issue #1204). 2008-09-07 19:18:16 +00:00
Gregory P. Smith 980b99b529 - Issue #1204: The configure script now tests for additional libraries
that may be required when linking against readline.  This fixes issues
  with x86_64 builds on some platforms (at least a few Linux flavors as
  well as OpenBSD/amd64).
2008-09-07 05:15:18 +00:00
Gregory P. Smith ff7b2d5ac6 Fix issue 3645: OpenBSD required -lcurses when linking with readline
to get the correct completion_matches function to avoid crashes on
x86_64 (amd64).

I don't have OpenBSD to test myself.  I tested that it does not break
anything on linux.  It is simple.
2008-09-03 05:57:48 +00:00
Ronald Oussoren a4cb472fb0 Fix typo in the `arch` commandline 2008-08-12 12:29:13 +00:00
Ronald Oussoren 5a849d4911 Fix buglet in fix for issue3381 2008-07-22 07:06:00 +00:00
Benjamin Peterson 4347c44f18 #3381 fix framework builds on 10.4 2008-07-17 15:59:24 +00:00
Benjamin Peterson 0e6ea5d715 fix framework install on Mac 10.4 2008-07-16 20:17:04 +00:00
Mark Dickinson dc1688adbf Fix typo in configure.in, and propagate configure.in changes from r64002
to configure and pyconfig.h.in.
2008-06-27 22:20:14 +00:00
Travis E. Oliphant 711c9e954e Add long double check support to configure test. 2008-06-06 22:33:21 +00:00
Ronald Oussoren 0d236eb05a Fix build issue on OSX 10.4 2008-06-06 21:31:33 +00:00
Neal Norwitz 2a3e3d0446 Aldo Cortesi confirmed this is still needed for OpenBSD 4.2 and 4.3.
(I didn't regen configure, since I don't have a working autoconf.)
2008-06-06 04:47:01 +00:00
Ronald Oussoren 5640ce2f1e MacOS X: Enable 4-way universal builds
This patch adds a new configure argument on OSX:
        --with-universal-archs=[32-bit|64-bit|all]

When used with the --enable-universalsdk option this controls which
CPU architectures are includes in the framework. The default is 32-bit,
meaning i386 and ppc. The most useful alternative is 'all', which includes
all 4 CPU architectures supported by MacOS X (i386, ppc, x86_64 and ppc64).

This includes limited support for the Carbon bindings in 64-bit mode as well,
limited because (a) I haven't done extensive testing and (b) a large portion
of the Carbon API's aren't available in 64-bit mode anyway.

I've also duplicated a feature of Apple's build of python: setting the
environment variable 'ARCHFLAGS' controls the '-arch' flags used for building
extensions using distutils.
2008-06-05 12:58:24 +00:00
Benjamin Peterson 8b12ee1072 fix typo (thank Georg) 2008-05-26 14:51:54 +00:00
Benjamin Peterson d7c778477e put a big note on configure.in asking people to run autoconf 2008-05-26 13:01:25 +00:00
Mark Dickinson 0b334a64f2 Revert temporary checkin in revision 63545. 2008-05-23 18:04:06 +00:00
Martin v. Löwis 8c255e4173 Patch #1722225: Support QNX 6. 2008-05-23 15:06:50 +00:00
Mark Dickinson 8f66a4a3db Temporary checkin to get configure to report more
information about the processor on the Debian/alpha
buildbot.  (I'm still trying to track down the cause
of the test_math failures for this machine.)  This
checkin will be reverted within the next 48 hours.
2008-05-23 04:22:50 +00:00
Ronald Oussoren 580c7fec67 Fix for issue #2573: Can't change the framework name on OS X builds
This introduces a new configure option: --with-framework-name=NAME
(defaulting to 'Python'). This allows you to install several copies
of the Python framework with different names (such as a normal build
and a debug build).
2008-05-02 19:45:11 +00:00
Mark Dickinson 6513466270 Issue #1496032. Add -mieee to BASECFLAGS on alpha, when gcc is
the compiler.  This should(?) help to fix failures in test_math
and test_cmath on Linux/alpha.

Also add configure message reporting the result of uname -m, as
a debugging aid.
2008-04-25 16:11:04 +00:00
Mark Dickinson 7a3d8641a9 Make configure test for tanh(-0.) == -0. committed in r62447 actually
work.  (The test wasn't properly linked with libm.  Sigh.)
2008-04-22 00:54:27 +00:00
Mark Dickinson 265d7384b9 test_math and test_cmath are failing on the FreeBSD 6.2 trunk buildbot,
apparently because tanh(-0.) loses the sign of zero on that platform.
If true, this is a bug in FreeBSD.

Added a configure test to verify this.  I still need to figure out
how best to deal with this failure.
2008-04-21 22:32:24 +00:00
Thomas Heller dc96a77c3a Issue #2544: On HP-UX systems, use 'gcc -shared' for linking when gcc
is used as compiler.
2008-04-04 10:07:55 +00:00
Neal Norwitz 9fdfaaf9af Fix compiler warning about finite() missing on Solaris. 2008-03-28 05:34:59 +00:00
Martin v. Löwis aef18b1c67 Patch #2240: Implement signal.setitimer and signal.getitimer. 2008-03-24 13:31:16 +00:00
Christian Heimes 0e9ab5f2f0 Applied patch #1657 epoll and kqueue wrappers for the select module
The patch adds wrappers for the Linux epoll syscalls and the BSD kqueue syscalls. Thanks to Thomas Herve and the Twisted people for their support and help.
TODO: Finish documentation documentation
2008-03-21 23:49:44 +00:00
Brett Cannon aa5778d1b8 Remove our implementation of memmove() and strerror(); both are in the C89
standard library.
2008-03-18 04:09:00 +00:00
Jeffrey Yasskin 1b4e45bab9 Allow Gnu gcc's to build python on OSX by removing -Wno-long-double,
-no-cpp-precomp, and -mno-fused-madd from configure.
 * r22183 added -no-cpp-precomp, which
   http://gcc.gnu.org/ml/gcc/2005-12/msg00368.html claims hasn't been
   needed since gcc-3.1.
 * r25607 added -Wno-long-double to avoid a warning in
   Include/objimpl.h (issue 525481). The long double is still there,
   but OSX 10.4's gcc no longer warns about it.
 * r33666 fixed issue 775892 on OSX 10.3 by adding -mno-fused-madd,
   which changed the sign of some float 0s. Tim Peters said it wasn't
   a real issue anyway, and it no longer causes test failures.
Fixes issue #1779871.
2008-03-17 14:40:53 +00:00
Thomas Heller 8bdf81d2df Merged changes from libffi3-branch.
The bundled libffi copy is now in sync with the recently released
libffi3.0.4 version, apart from some small changes to
Modules/_ctypes/libffi/configure.ac.

I gave up on using libffi3 files on os x.
Instead, static configuration with files from pyobjc is used.
2008-03-04 20:09:11 +00:00
Brett Cannon 91a1dec492 Fix --enable-universalsdk and its comment line so that zsh's flag completion
works.

Thanks to Jeroen Ruigrok van der Werven for the fix.
2008-02-13 19:15:44 +00:00
Brett Cannon 9a8bb0e316 Fix a minor typo. 2008-02-03 02:07:55 +00:00
Christian Heimes 951cc0f474 Fixed bug #1983: Return from fork() is pid_t, not int 2008-01-31 23:08:23 +00:00
Christian Heimes cba36bbe65 Bug #1234: Fixed semaphore errors on AIX 5.2 2008-01-30 22:54:18 +00:00
Christian Heimes 487235109b Fixed a wrong assumption in configure.in and Include/pyport.h. The is finite function is not called isfinite() but finite(). Sorry, my fault. :) 2008-01-20 14:28:28 +00:00
Christian Heimes fb2d25a154 Issue #1646: Make socket support TIPC. The socket module now has support
for TIPC under Linux, see http://tipc.sf.net/ for more information.
Thanks to Alberto Bertogli for the patch
2008-01-07 16:12:44 +00:00
Martin v. Löwis cb78de6d25 Bug #1699: Define _BSD_SOURCE only on OpenBSD. 2007-12-29 18:49:21 +00:00
Christian Heimes 0a8143f646 Applied patch #1635: Float patch for inf and nan on Windows (and other platforms).
The patch unifies float("inf") and repr(float("inf")) on all platforms.
2007-12-18 23:22:54 +00:00
Christian Heimes db3d6cbce0 Fixed #1638: %zd configure test fails on Linux 2007-12-16 21:39:43 +00:00
Guido van Rossum 7c862f8077 Patch #1608. Someone with access to autoconf 2.61 or higher needs to
run it and check in the resulting configure file.
2007-12-13 20:50:10 +00:00