summaryrefslogtreecommitdiff
path: root/numpy/distutils
Commit message (Collapse)AuthorAgeFilesLines
* BUG: adjust gfortran version search regexJonathan Helmus2017-09-222-2/+4
| | | | | | | Adjust the gfortran regex to require a '.' to be present in the version number. This avoids matching a hash or text containing a '-' such as the version string from the compilers produced by crosstool-NG. For example: GNU Fortran (crosstool-NG 8a21ab48) 7.2.0
* BLD: remove -xhost flag from IntelFCompiler. Closes gh-9042.Ralf Gommers2017-08-141-1/+1
| | | | | | | Note that this was discussed extensively in gh-7287, but removing this -xhost flag was missed in the PR that closed that issue. [ci skip]
* BUG: fix regression in 1.13.x in distutils.mingw32ccompiler.Ralf Gommers2017-08-141-1/+1
| | | | | | Issue was introduced in gh-8454. Thanks to @jennirinker for pointing out the issue and fix. Closes gh-9553.
* BUG: Handle MinGW module compilation in Py2.7 venvsRoland Kaufmann2017-06-271-1/+9
| | | | | | | | | | | If we are using the virtualenv module in Python 2.7 (as opposed to venv in Python 3.3+), then the location of the base distribution is stored in the real_prefix attribute instead of the base_prefix attribute. This changeset adds the content of this attribute to the search paths if it exist (i.e. in practice, if we are running Python 2.7), so that the runtime libraries do not have to be duplicated in the virtualenv directory.
* BUG: Make extensions compilable with MinGW on Py2.7Roland Kaufmann2017-06-271-3/+7
| | | | | | | | | | | | The original changeset 1c8ecc7 to make extensions compilable with MinGW used the base_prefix attribute unconditionally. However, this is not available in versions before Python 3.3. This changeset introduces a check for whether the attribute is available so that the code will work without errors also on Python 2.7. However, this may lead to surprising results when compiling in a virtualenv. Hat tip to Diorcet Yann (@diorcety) for pointing out this problem.
* BUG: Fix Intel compilation on Unix.Charles Harris2017-06-263-4/+24
| | | | | | | | | | | Fixes two problems: * c compilers do not have a find_executables method. * get_version return a LooseVersion instance, not string. Closes #9278. [ci skip]
* DOC: BLD: fix lots of Sphinx warnings/errors.Ralf Gommers2017-06-101-13/+15
|
* BUG: Compilation crashes in MSVC when LIB or INCLUDE is not setDavid Hagen2017-05-181-5/+5
|
* TST: fix test_basic failure on WindowsChristoph Gohlke2017-05-101-1/+1
| | | Echo is a command in the Windows command interpreter
* BUG: distutils, add compatiblity python parallelizationJulian Taylor2017-05-041-3/+44
| | | | | | | | | | | | | | | | | | | | Python 3.5 also added build parallelization at the extension level instead of the file leve numpy uses. This causes two problems: - numpy.distutils is not threadsafe with duplicated source files When source files are duplicated in multiple extensions the output objects are overwritten which can truncate in a parallel context. This is fixed by keeping track of the files being worked on and wait for completion if another thread is already using the object name. - The parallelization on two nested levels causes oversubscription. When building multiple extensions with multiple source files the number of jobs running is multiplied. This is fixed by adding a semaphore that limits the number of jobs numpy starts to the defined amount. closes gh-7139
* BUG: distutils, place fortranobject files in subfolderJulian Taylor2017-05-041-6/+6
| | | | | Placing them all under the same name in the top level folder breaks when using the parallel extension compilation option of python 3.5.
* BUG: check compiler flags to determine the need for a rebuildJulian Taylor2017-05-023-5/+34
| | | | | | | Some projects compile an extension multiple times with different flags. As it is not intended to get the extension name from the compiler object, store the last used flags in the dependency file and also check if they match the flags of the current build.
* MAINT: Update Intel compiler options.Charles Harris2017-04-192-4/+16
| | | | | | | | | The '-openmp' option was deprecated in Intel version 15 and removed in version 18. The replacement is '-qopenmp'. Closes #8941. [skip ci]
* ENH: automatically determine compile dependenciesJulian Taylor2017-04-103-4/+77
| | | | | Use these dependencies to avoid unnecessary recompilations of unchanged files.
* BUG: return values of exec_command were swappedEric Wieser2017-04-051-1/+1
| | | | This caused a nonsensical failure on my windows machine
* TST: Respect compiler customizationsMarvin Schmidt2017-03-301-0/+3
| | | | | Without calling `compiler.customize(None)` customizations using environment variables like CC won't be honoured
* MAINT: remove some unused variablesJulian Taylor2017-03-241-6/+1
|
* TST: move exec_command tests out of the file itselfJulian Taylor2017-03-242-165/+127
|
* TST: remove exec_command svn testJulian Taylor2017-03-241-6/+0
| | | | Not sure what it even tests, it fails with svn installed.
* MAINT: restore stderr return value of exec_commandJulian Taylor2017-03-241-7/+1
| | | | | The old os.system code just redirected stderr to stdout so mimic that behaviour. This fixes at least the testsuite.
* MAINT: print the received lengths on template length mismatchJulian Taylor2017-03-241-2/+2
|
* MAINT: remove an extra newline with the Popen based execJulian Taylor2017-03-241-1/+1
|
* Merge pull request #7862 from pitrou/rewrite_exec_commandJulian Taylor2017-03-111-255/+53
|\ | | | | BLD: rewrite np.distutils.exec_command.exec_command()
| * BLD: rewrite np.distutils.exec_command.exec_command()Antoine Pitrou2016-07-251-255/+53
| | | | | | | | | | | | exec_command() is currently a mess of several implementations using outdated Python APIs and various hacks. This rewrites it to use the standard subprocess module. See PR #7614 for previous discussion.
* | MAINT: better warning message when running build_src from sdistRalf Gommers2017-03-091-3/+11
| | | | | | | | Addresses comment of @stefanv on gh-7131.
* | BUG: Skip custom library check if not MSVC runtimeRoland Kaufmann2017-02-231-3/+12
| | | | | | | | | | | | | | | | | | | | If Python has been built using a different runtime, for example using the GNU compiler in MSYS, the version string will of course not contain a reference to the MSVC runtime. Instead of attempting to build an import library for yet another runtime, we assume that the compiler used to build the extension is now the same as was used to build Python, and that it will pass the right runtime to the linker without any further command-line options.
* | BUG: Correct signature of inherited methodRoland Kaufmann2017-02-231-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The msvccompiler module is made in the same mold as the msvc9compiler module; saving the two environment variables around the call to initialize. However, the signaure in the msvccompiler module's version of the MSVCCompiler class, there is no plat_name argument to the method. Since this is not used anyway but just passed along, it can simply be removed. This bug can be exposed even if you are not using the Intel Fortran compiler, because the class is loaded in an attempt to probe for all compilers, if the correct compiler is not found yet.
* | BUG: blas_info should record include_dirsBenda Xu2017-02-151-0/+2
| | | | | | | | | | blas specified in site.cfg could have non-standard include dirs. It should be read and saved in distutils/__config__.py
* | Fix building extensions with MinGW in WinPython 3.4 (#8454)Roland Kaufmann2017-01-221-41/+75
| | | | | | MAINT: Fix building extensions with MinGW in WinPython 3.4
* | MAINT: Remove shebang from test modulesSami Salonen2017-01-151-1/+0
| |
* | MAINT: use env in shebang instead of absolute pathSami Salonen2017-01-152-2/+2
| |
* | DEP: Fix escaped string characters deprecated in Python 3.6.Charles Harris2016-12-1412-25/+26
| | | | | | | | | | | | | | In Python 3.6 a number of escape sequences that were previously accepted -- for instance "\(" that was translated to "\\(" -- are deprecated. To retain the previous behavior either raw strings must be used or the backslash must be properly escaped itself.
* | Merge pull request #8356 from rolk/8356_mingw_virtenvCharles Harris2016-12-121-6/+24
|\ \ | | | | | | Allow extensions to be built with MinGW in a virtualenv
| * | STY: Avoid using string formatting in log callsRoland Kaufmann2016-12-091-6/+5
| | | | | | | | | | | | | | | | | | | | | The log class (derived from distutils.log) does string formatting with extra arguments passed, if (and only if) the log level is above the set threshold. That way, the cost of string formatting only incurs if the message is actually going to be displayed.
| * | BUG: Search for Python DLL in base distr. with MinGWRoland Kaufmann2016-12-081-4/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If an extension is built in a virtualenv, the compiler adapter for the MinGW distribution will search for the libs/ directory containing the import libraries for the Python runtime, in the virtualenv. However, these files are not copied when the virtualenv is created (using the venv module), so the build fails. This patch searches the base distribution directory for the libraries in addition to the virtualenv, allowing extensions to be built without manually copying the libs/ directory first.
* | | Merge pull request #8355 from rolk/8355_mingw_py35Charles Harris2016-12-122-20/+44
|\ \ \ | | | | | | | | Fix building extensions with MinGW for Python 3.5
| * | | MAINT: Remove assumptions about RTL name, for MinGWRoland Kaufmann2016-12-091-9/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The MinGW compiler adapter assumes that the C runtime library that Python was built with has the name msvcrXXX. This is incorrect for Python 3.5 and later. (The name of the runtime is now vcruntimeXXX). This patch removes parsing of the hardcoded pattern and replaces it with calls to new functions in misc_util that returns the desired values directly, isolating the hardcoding to that module.
| * | | ENH: Monkey-patch CygwinCCompiler for Py3.5 on Wind.Roland Kaufmann2016-12-091-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | CygwinCCompiler has its own test for the C runtime library which raises an error if the version extracted from the version string does not match a pre-defined list. The runtime for Visual Studio 14 (aka 2015) which is used to build Python 3.5 is not on this list. As this call is made in the constructor of the base class, it must be monkey-patched to get around.
| * | | ENH: Recognize C RTL used by Py3.5 on WinRoland Kaufmann2016-12-091-1/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Python 3.5 is built against vcruntime140 instead of msvcrXXX, and the internal version number reported by the version string is 1900. This cause building extensions to fail, complaining about a missing compiler. This patch adds the latest runtime library to the list.
| * | | MAINT: Isolate hardcoded RTL name and vers. on Win.Roland Kaufmann2016-12-091-11/+22
| |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are three concepts in the C runtime library used by Python on Windows (from the Microsoft Visual C compiler); the internal build number (e.g. 0x1600), the major version of the runtime (e.g. 10.0), and the name of the link library of the runtime (e.g. msvcr100). This patch separates getting these three values into separate functions so that other code can be relieved of parsing the string return value of the msvc_runtime_library function.
* | | MAINT: Other cleanup Python < 2.7 and Python3 < 3.4Alexandr Shadchin2016-12-091-18/+1
|/ /
* | BUG: do not link to Accelerate if OpenBLAS, MKL or BLIS are found.Ralf Gommers2016-11-241-2/+4
| | | | | | | | Closes gh-8293.
* | MAINT: change path to env in distutils.system_info. Closes gh-8195.Ralf Gommers2016-10-221-7/+5
| | | | | | | | Also clean up some unused variables.
* | MAINT: Fix ResourceWarning new in Python 3.6.Charles Harris2016-10-191-0/+1
| | | | | | | | | | | | | | | | | | In Python 2.6 the process destructor emits a ResourceWarning if the process has not terminated when destruction occurs. Fix by calling the wait method before return. The Popen class may be used as a context manager in Python 3.2+, unfortunately that does not include 2.7.
* | BUG: Fixed build breakOleksandr Pavlyk2016-10-061-0/+2
| | | | | | | | Added missing import of platform_bits
* | BUG: distutils: fix issue with rpath in fcompiler/gnu.pyRalf Gommers2016-10-021-1/+3
| | | | | | | | | | | | ``-Wl,rpath=`` is Linux-specific, OS X needs a comma. This addresses gh-6486, which was already closed but not actually addressed.
* | Fix test code on cpuinfo's main functionGustavo Serra Scalet2016-09-301-4/+4
| | | | | | | | | | Due to future print import (py3k style), the commented code should also have the strings enclosed with parentheses or else it'll break.
* | BUG: Fix fid.close() to use os.close(fid)Sebastian Berg2016-09-041-1/+1
| | | | | | | | | | | | | | The error snuk in with some other file closing fixes and found by Warren Weckesser. Closes gh-8013
* | TST: Add tests for stacklevel in warnings and "ignore" filters.Sebastian Berg2016-09-021-1/+1
| | | | | | | | | | Enforces that stacklevel is used in warnings.warn and "ignore" is not used in filterwarnings or simplefilter for most of numpy.
* | ENH: Add stacklevel to all (or almost all) our function callsSebastian Berg2016-09-027-23/+28
| |
View Lorry Controller Status