summaryrefslogtreecommitdiff
path: root/numpy/distutils
Commit message (Collapse)AuthorAgeFilesLines
* Fix conv_template to correctly handle nested loops with surrounding code.Charles Harris2008-04-241-13/+23
| | | | Add some template debugging aids.
* Add true nesting of loops to the template processing. The previous attemptCharles Harris2008-04-231-123/+176
| | | | | | wasn't very useful. Loops can now be nested within each other using /**begin repeat1 and /**end repeat1**/ and cousins, starting with the current tags for the outermost loops.
* Does not replace g++ by c++ compiler name.David Cournapeau2008-04-231-5/+1
|
* Add nested loops to the template processor.Charles Harris2008-04-231-45/+69
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The syntax is to separate the nested loops with lines containing strings like #repeat = 1# #repeat = 2# ... The number doesn't really matter, but it helps document the depth. Example: The file containing /**begin repeat * # a = 1,2,3# * # b = 1,2,3# * # repeat = 1# * # c = 1,2# * # d = 1,2# */ @a@@b@@c@@d@ /**end repeat**/ produces #line 1 "template.c.src" /* ***************************************************************************** ** This file was autogenerated from a template DO NOT EDIT!!!! ** ** Changes should be made to the original source (.src) file ** ***************************************************************************** */ #line 8 1111 #line 8 1122 #line 8 2211 #line 8 2222 #line 8 3311 #line 8 3322
* Do not show the whole scons command for silent modes.David Cournapeau2008-04-211-1/+4
|
* ran reindent in preparation for the 1.1 releaseJarrod Millman2008-04-201-2/+2
|
* Reverting r4481, see #723Jarrod Millman2008-04-201-0/+63
|
* Use the default terminal colour to print out INFO messages in distutils. Stefan van der Walt2008-04-162-3/+5
| | | | This prevents visibility problems on backgrounds other than black.
* Add one more level for silent modes in scons command.David Cournapeau2008-04-131-4/+5
|
* Fix issue 715.Pearu Peterson2008-04-101-5/+6
|
* mimic gcc driver detection for nocona cpu detection.David Cournapeau2008-03-231-1/+1
|
* Fix has_sse3 and add has_ssse3 function for cpuinfo on linux.David Cournapeau2008-03-231-1/+4
|
* Include patch from jsbronder to fix ticket #644 related to nocona cpu detection.David Cournapeau2008-03-221-1/+5
|
* Handle typo (sse3 vs ssse3) in the cpu arch flags as found in /proc/cpuinfo ↵David Cournapeau2008-03-221-1/+1
| | | | for some linux kernels (at least 2.6.23).
* Add VisualCompaq compiler to the list of fortran compilers if platform is ↵David Cournapeau2008-03-221-1/+1
| | | | cygwin.
* Comment out NotImplemented error when using setup tools.Travis Oliphant2008-03-221-1/+1
|
* Fix typo (bug #704).David Cournapeau2008-03-211-1/+1
|
* In site.cfg, change DEFAULT section to ALL to accommodate building on Python ↵Stefan van der Walt2008-03-192-6/+6
| | | | 2.6.
* Add pkg name at the beginning of logging scons call.David Cournapeau2008-03-161-1/+1
|
* Fixed setting of distutils.[plat]ccompiler.[plat]ccompiler leading to ↵Matthew Brett2008-03-101-2/+1
| | | | attribute error
* Apply David's patch to fix ticket 653.Charles Harris2008-03-061-20/+10
|
* Fix type. Sorry about that.Pearu Peterson2008-02-121-3/+3
|
* hpux fortran compiler support: patch from an user feedback.Pearu Peterson2008-02-121-2/+5
|
* ran reindentJarrod Millman2008-02-082-15/+15
|
* Return as soon as possible of no sconscripts is detected in scons command, toDavid Cournapeau2008-01-251-0/+3
| | | | avoid importing numscons when not needed
* Make distutils command build run scons commandDavid Cournapeau2008-01-251-0/+6
|
* Use eval instead of dynamic import in scons_generate_config_py, so that we doDavid Cournapeau2008-01-231-4/+2
| | | | | not generate spurious *.pyc files.
* Add setupscons.py for pure python packagesDavid Cournapeau2008-01-071-0/+15
|
* Fix missing import from numsconsDavid Cournapeau2008-01-071-0/+1
|
* Import get_configres* functions from numsconsDavid Cournapeau2008-01-061-0/+1
|
* Add function add_configres for numpy.distutils.ConfigurationDavid Cournapeau2008-01-061-0/+4
|
* Define various scons related functions to call numscons instead of beingDavid Cournapeau2008-01-061-3/+6
| | | | | defined directly in numpy, for bootstrapping issues.
* Add scons_make_config function to numpy distutils Configuration.David Cournapeau2008-01-061-0/+6
|
* Use separate numscons package instead of internal local copyDavid Cournapeau2008-01-061-2/+9
|
* Add facilities to generate config file for future sconsified modules.David Cournapeau2008-01-061-0/+45
|
* add an option to change setup filenames to numpy.distutils.ConfigurationDavid Cournapeau2008-01-061-1/+4
|
* Add the infrastructure to add SConstruct files inDavid Cournapeau2008-01-061-1/+45
| | | | | numpy.distutils.Configuration.
* Add scons command to numpy distutils.David Cournapeau2008-01-062-1/+324
|
* Use a custom NumpyDistribution instead of distutils Distribution, to handleDavid Cournapeau2008-01-062-1/+34
| | | | | scons scripts.
* Merged revisions 4612-4617,4628-4631,4638 via svnmerge from David Cournapeau2008-01-062-2/+17
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | http://svn.scipy.org/svn/numpy/branches/cleanconfig_rtm ................ r4612 | cdavid | 2007-12-18 17:58:13 +0900 (Tue, 18 Dec 2007) | 3 lines Initialized merge tracking via "svnmerge" with revisions "1-4611" from http://svn.scipy.org/svn/numpy/trunk ................ r4613 | cdavid | 2007-12-18 18:26:38 +0900 (Tue, 18 Dec 2007) | 1 line Clean ndarrayobject.h headers: do not use SIZEOF_*, but NPY_SIZEOF... ................ r4614 | cdavid | 2007-12-18 18:33:11 +0900 (Tue, 18 Dec 2007) | 1 line Refactor the NPY_ALLOW_THREAD logic outside config.h generation, and put it inside numpy headers ................ r4615 | cdavid | 2007-12-18 18:46:34 +0900 (Tue, 18 Dec 2007) | 1 line separate config.h (private header) and numpyconfig.h (public header) ................ r4616 | cdavid | 2007-12-18 18:57:21 +0900 (Tue, 18 Dec 2007) | 2 lines Add CHAR_BITS as a public symbol in numpyconfig header ................ r4617 | cdavid | 2007-12-18 19:05:38 +0900 (Tue, 18 Dec 2007) | 1 line Fix escaping problem when generating numpyconfig.h on win32 ................ r4628 | cdavid | 2007-12-22 12:20:08 +0900 (Sat, 22 Dec 2007) | 1 line Add code to test for a declaration in header ................ r4629 | cdavid | 2007-12-22 14:52:38 +0900 (Sat, 22 Dec 2007) | 1 line Add a check_decl function to check for declaration ................ r4630 | cdavid | 2007-12-22 15:28:11 +0900 (Sat, 22 Dec 2007) | 1 line Remove distutils test for C declaration from numpy.core, since it is put in distutils now ................ r4631 | cdavid | 2007-12-22 15:58:15 +0900 (Sat, 22 Dec 2007) | 18 lines Merged revisions 4612-4630 via svnmerge from http://svn.scipy.org/svn/numpy/trunk ........ r4620 | cdavid | 2007-12-18 19:41:22 +0900 (Tue, 18 Dec 2007) | 3 lines Initialized merge tracking via "svnmerge" with revisions "1-4619" from http://svn.scipy.org/svn/numpy/branches/distutils_scons_command ........ r4624 | rc | 2007-12-20 19:30:45 +0900 (Thu, 20 Dec 2007) | 2 lines fixed setmember1d for string arrays ........ r4626 | stefan | 2007-12-21 18:00:24 +0900 (Fri, 21 Dec 2007) | 2 lines Add test for setdiff1d on character arrays. ........ ................ r4638 | cdavid | 2007-12-22 18:21:25 +0900 (Sat, 22 Dec 2007) | 1 line Get MATHLIB from numpyconfig.h, not from config.h anymore ................
| * Get MATHLIB from numpyconfig.h, not from config.h anymoreDavid Cournapeau2007-12-221-2/+2
| |
| * Add a check_decl function to check for declarationDavid Cournapeau2007-12-221-0/+15
| |
* | Replace numpy.distutils.exec_command.splitcmdline with shlex.split instead.cookedm2007-12-263-60/+11
| | | | | | | | | | | | | | | | It has the same problems as our old numpy.distutils.ccompiler.split_quoted. splitcmdline still exists, but uses shlex.split, and issues a DeprecationWarning This has the positive side effect of not having numpy.distutils pulled in when numpy is imported -- there was a use of splitcmdline in numpy.testing.
* | Add an 'if 0''d block of code to numpy.distutils.log to ignore some log ↵cookedm2007-12-261-3/+7
|/ | | | messages. Especially useful to turn on if you're developing by using eggs.
* Also link with Python.Robert Kern2007-12-161-1/+1
|
* Make sure the develop command can handle generated script files.Robert Kern2007-12-122-2/+9
|
* Make the 'develop' command from setuptools run build_src --inplace in ↵Robert Kern2007-12-123-2/+19
| | | | addition to build_ext --inplace. This allows SWIG wrappers to be correctly generated.
* Typo.Robert Kern2007-12-041-1/+1
|
* Use the correct link parameters for shared libraries with Intel Fortran on OS X.Robert Kern2007-12-041-0/+10
|
* The PIC flag changed in the 10 version of the Intel Fortran compiler.Robert Kern2007-12-041-10/+8
|