# # spec file for package python-numpy # # Copyright (c) 2015 SUSE LINUX Products GmbH, Nuernberg, Germany. # # All modifications and additions to the file contributed by third parties # remain the property of their copyright owners, unless otherwise agreed # upon. The license for this file, and modifications and additions to the # file, is the same license as for the pristine package itself (unless the # license for the pristine package is not an Open Source License, in which # case the license is the MIT License). An "Open Source License" is a # license that conforms to the Open Source Definition (Version 1.9) # published by the Open Source Initiative. # Please submit bugfixes or comments via http://bugs.opensuse.org/ # %define with_atlas 0 Name: python-numpy Version: 1.9.2 Release: 1.2 Url: http://www.numpy.org/ Summary: NumPy array processing for numbers, strings, records and objects License: BSD-3-Clause Group: Development/Libraries/Python Source: https://pypi.python.org/packages/source/n/numpy/numpy-%{version}.tar.gz # PATCH-FIX-OPENSUSE numpy-buildfix.patch -- openSUSE-specific build fixes Patch0: numpy-buildfix.patch # PATCH-FIX-OPENSUSE numpy-1.9.0-remove-__declspec.patch -- fix for spurious compiler warnings that cause build failure Patch1: numpy-1.9.0-remove-__declspec.patch BuildRoot: %{_tmppath}/%{name}-%{version}-build BuildRequires: blas-devel BuildRequires: lapack-devel BuildRequires: python-devel >= 2.6 Requires: python >= %{py_ver} Provides: numpy = %{version} %if %{with_atlas} == 1 BuildRequires: libatlas3-devel %endif %if 0%{?suse_version} BuildRequires: fdupes BuildRequires: gcc-fortran %if 0%{?suse_version} <= 1110 %{!?python_sitearch: %global python_sitearch %(python -c "from distutils.sysconfig import get_python_lib; print(get_python_lib(1))")} %else %py_requires %endif %else BuildRequires: gcc-gfortran %if ! 0%{?fedora_version} Provides: python-numeric = %{version} Obsoletes: python-numeric < %{version} %endif %endif %description NumPy is a general-purpose array-processing package designed to efficiently manipulate large multi-dimensional arrays of arbitrary records without sacrificing too much speed for small multi-dimensional arrays. NumPy is built on the Numeric code base and adds features introduced by numarray as well as an extended C-API and the ability to create arrays of arbitrary type which also makes NumPy suitable for interfacing with general-purpose data-base applications. There are also basic facilities for discrete fourier transform, basic linear algebra and random number generation. %package devel Summary: Development files for numpy applications Group: Development/Libraries/Python Requires: %{name} = %{version} Requires: blas-devel Requires: lapack-devel Requires: python-devel >= %{py_ver} %if 0%{?suse_version} Requires: gcc-fortran %py_requires -d %else Requires: gcc-gfortran %endif %if %{with_atlas} == 1 Requires: libatlas3-devel %endif %description devel This package contains files for developing applications using numpy. %prep %setup -q -n numpy-%{version} %patch0 -p1 %patch1 -p1 # Fix non-executable scripts sed -i "1d" numpy/{compat/setup,distutils/{conv_template,cpuinfo,exec_command,from_template,setup,system_info},f2py/{__init__,auxfuncs,capi_maps,cb_rules,cfuncs,common_rules,crackfortran,diagnose,f2py2e,f90mod_rules,func2subr,rules,setup,use_rules},ma/setup,matrixlib/setup,setup,testing/{print_coercion_tables,setup}}.py %build CFLAGS="%{optflags} -fno-strict-aliasing" python setup.py build %install python setup.py install --root="%{buildroot}" --prefix="%{_prefix}" rm -rf %{buildroot}%{python_sitearch}/numpy/{,core,distutils,f2py,fft,lib,linalg,ma,matrixlib,oldnumeric,polynomial,random,testing}/tests # Don't package testsuite %if 0%{?suse_version} %fdupes -s %{buildroot}%{_prefix} %endif %files %defattr(-,root,root) %doc *.txt %{_bindir}/f2py %{python_sitearch}/numpy/ %{python_sitearch}/numpy-%{version}-py*.egg-info %exclude %{python_sitearch}/numpy/*/*/*.c %exclude %{python_sitearch}/numpy/*/*.h %exclude %{python_sitearch}/numpy/*/*/*.h %exclude %{python_sitearch}/numpy/*/*/*/*.h %exclude %{python_sitearch}/numpy/core/lib/libnpymath.a %files devel %defattr(-,root,root) %{python_sitearch}/numpy/*/*/*.c %{python_sitearch}/numpy/*/*.h %{python_sitearch}/numpy/*/*/*.h %{python_sitearch}/numpy/*/*/*/*.h %{python_sitearch}/numpy/core/lib/libnpymath.a %changelog * Mon Mar 2 2015 toddrme2178@gmail.com - Update to 1.9.2: Bugfix release * #5316: fix too large dtype alignment of strings and complex types * #5424: fix ma.median when used on ndarrays * #5481: Fix astype for structured array fields of different byte order * #5354: fix segfault when clipping complex arrays * #5524: allow np.argpartition on non ndarrays * #5612: Fixes ndarray.fill to accept full range of uint64 * #5155: Fix loadtxt with comments=None and a string None data * #4476: Masked array view fails if structured dtype has datetime component * #5388: Make RandomState.set_state and RandomState.get_state threadsafe * #5390: make seed, randint and shuffle threadsafe * #5374: Fixed incorrect assert_array_almost_equal_nulp documentation * #5393: Add support for ATLAS > 3.9.33. * #5313: PyArray_AsCArray caused segfault for 3d arrays * #5492: handle out of memory in rfftf * #4181: fix a few bugs in the random.pareto docstring * #5359: minor changes to linspace docstring * #4723: fix a compile issues on AIX * Thu Nov 6 2014 toddrme2178@gmail.com - Update to 1.9.1: Bugfix release * gh-5184: restore linear edge behaviour of gradient to as it was in < 1.9. The second order behaviour is available via the `edge_order` keyword * gh-4007: workaround Accelerate sgemv crash on OSX 10.9 * gh-5100: restore object dtype inference from iterable objects without `len()` * gh-5163: avoid gcc-4.1.2 (red hat 5) miscompilation causing a crash * gh-5138: fix nanmedian on arrays containing inf * gh-5203: copy inherited masks in MaskedArray.__array_finalize__ * gh-2317: genfromtxt did not handle filling_values=0 correctly * gh-5067: restore api of npy_PyFile_DupClose in python2 * gh-5063: cannot convert invalid sequence index to tuple * gh-5082: Segmentation fault with argmin() on unicode arrays * gh-5095: don't propagate subtypes from np.where * gh-5104: np.inner segfaults with SciPy's sparse matrices * gh-5136: Import dummy_threading if importing threading fails * gh-5148: Make numpy import when run with Python flag '-OO' * gh-5147: Einsum double contraction in particular order causes ValueError * gh-479: Make f2py work with intent(in out) * gh-5170: Make python2 .npy files readable in python3 * gh-5027: Use 'll' as the default length specifier for long long * gh-4896: fix build error with MSVC 2013 caused by C99 complex support * gh-4465: Make PyArray_PutTo respect writeable flag * gh-5225: fix crash when using arange on datetime without dtype set * gh-5231: fix build in c99 mode * Mon Sep 8 2014 toddrme2178@gmail.com - Update to 1.9.0 Highlights * Numerous performance improvements in various areas, most notably indexing and operations on small arrays are significantly faster. Indexing operations now also release the GIL. * Addition of nanmedian and nanpercentile rounds out the nanfunction set. - Remove upstreamed numpy-double-double-le.patch - Add numpy-1.9.0-remove-__declspec.patch Fixes spurious warnings during build. These warnings are related to a windows-specific function that is not built, so the patch just removed the function. See: http://lists.opensuse.org/opensuse-packaging/2014-09/msg00004.html - Added patch tag for numpy-buildfix.patch * Mon Aug 11 2014 toddrme2178@gmail.com - Update to 1.8.2: Bugfix release * gh-4836: partition produces wrong results for multiple selections in equal ranges * gh-4656: Make fftpack._raw_fft threadsafe * gh-4628: incorrect argument order to _copyto in in np.nanmax, np.nanmin * gh-4642: Hold GIL for converting dtypes types with fields * gh-4733: fix np.linalg.svd(b, compute_uv=False) * gh-4853: avoid unaligned simd load on reductions on i386 * gh-4722: Fix seg fault converting empty string to object * gh-4613: Fix lack of NULL check in array_richcompare * gh-4774: avoid unaligned access for strided byteswap * gh-650: Prevent division by zero when creating arrays from some buffers * gh-4602: ifort has issues with optimization flag O2, use O1 - Switch to pypi download location - Changed Url to current home page - Minor spec file cleanups * Mon Apr 14 2014 arun@gmx.de - bugfix release 1.8.1 * Thu Dec 5 2013 mls@suse.de - add numpy-double-double-le.patch for ppc64le * Thu Oct 31 2013 toddrme2178@gmail.com - Update to 1.8.0 Highlights: * New, no 2to3, Python 2 and Python 3 are supported by a common code base. * New, gufuncs for linear algebra, enabling operations on stacked arrays. * New, inplace fancy indexing for ufuncs with the ``.at`` method. * New, ``partition`` function, partial sorting via selection for fast median. * New, ``nanmean``, ``nanvar``, and ``nanstd`` functions skipping NaNs. * New, ``full`` and ``full_like`` functions to create value initialized arrays. * New, ``PyUFunc_RegisterLoopForDescr``, better ufunc support for user dtypes. * Numerous performance improvements in many areas. - Add a new flag to easily enable/disable atlas support for if it ever gets fixed in the future - Rebase numpy-buildfix.patch * Fri May 3 2013 toddrme2178@gmail.com - Update to 1.7.1 * Bugfixes * Fri Apr 5 2013 idonmez@suse.com - Add Source URL, see https://en.opensuse.org/SourceUrls * Tue Mar 12 2013 highwaystar.ru@gmail.com - update to 1.7.0 * This release includes several new features as well as numerous bug fixes and refactorings Highlights: - ``where=`` parameter to ufuncs (allows the use of boolean arrays to choose where a computation should be done) - ``vectorize`` improvements (added 'excluded' and 'cache' keyword, general cleanup and bug fixes) - ``numpy.random.choice`` (random sample generating function) New Features: - Reduction UFuncs Generalize axis= Parameter - Reduction UFuncs New keepdims= Parameter - Datetime support - Custom formatter for printing arrays - New function numpy.random.choice - New function isclose - Preliminary multi-dimensional support in the polynomial package - Ability to pad rank-n arrays - New argument to searchsorted - Added experimental support for the AArch64 architecture. * For additional details check release notes at http://sourceforge.net/projects/numpy/files/NumPy/1.7.0/ - numpy-aarch64.diff: removed, now upstream * Tue Mar 5 2013 dmueller@suse.com - add numpy-aarch64.diff: * Fix build for AArch64 * Fri Aug 17 2012 toddrme2178@gmail.com - Disable atlas buildrequires until the problems with atlas are worked out * Mon Jun 4 2012 toddrme2178@gmail.com - Add libatlas2 buildrequires * Fri Jun 1 2012 toddrme2178@gmail.com - Remove blas/lapack tests since these build successfully on all targets now - Add documentation packages These are separate packages because a lot of packages depend on numpy, so building the documentation inside the base spec file would slow down the build process for the entire project * Fri Jun 1 2012 kukuk@suse.de - Old products don't provide python(abi). * Mon May 21 2012 toddrme2178@gmail.com - Update to version 1.6.2: * bugfixes * Fri May 18 2012 toddrme2178@gmail.com - Fix rpmlint warnings * Mon Apr 30 2012 toddrme2178@gmail.com - Removed tests for unsupported openSUSE versions * Mon Apr 23 2012 toddrme2178@gmail.com - Add python 3 package - devel package BuildRequires -> Requires * Sun Apr 15 2012 idonmez@suse.com - Fix SLE_11 build * Fri Apr 13 2012 badshah400@gmail.com - Modify BuildRequires to match recent name changes of lapack and blas in project. * Sat Oct 1 2011 behrisch@users.sf.net - Added static lib to devel package for scipy build * Wed Sep 21 2011 saschpe@suse.de - Update to version 1.6.1: * Please see http://downloads.sourceforge.net/project/numpy/NumPy/1.6.1/Changelog - Spec file cleanup: * Set license to BSD-3-Clause (SPDX style) * Removed testsuite from packages * Removed rpmlintrc file and actually fixed the issues * Sun Jan 23 2011 toddrme2178@gmail.com - update to 1.5.1 Numpy 1.5.1 is a bug-fix release with no new features compared to 1.5.0. Relevant bugs fixed: * linalg: lstsq should always return real residual * lib: fix negative indices in s_ and index_exp * core: fix uint64 -> Python int cast * core: richcompare should return Py_NotImplemented when undefined * lib: close file handles after use in numpy.lib.npyio.* * core: ensure PEP 3118 buffers can be released in exception handler * core: fix clongdouble cast to Python complex() * core: fix detection for ``isfinite`` routine * Mon Sep 6 2010 coolo@novell.com - update to 1.5.0 to fix build against python 2.7 highlights 1.3 -> 1.4: * New datetime dtype support to deal with dates in arrays * Faster import time * Extended array wrapping mechanism for ufuncs * New Neighborhood iterator (C-level only) * C99-like complex functions in npymath highlights 1.4 -> 1.5: * support for python3 and python 2.7 * Mon May 31 2010 bg@novell.com - fix build for hppa * Fri Jan 15 2010 aeszter@gwdg.de - re-enable build on non-SUSE distros - do not provide python-numeric on Fedora, since a package by that name exists * Wed Jan 13 2010 lars@linux-schulserver.de - use a more generic way to define py_sitedir on all targets * Mon Jan 11 2010 coolo@novell.com - remerge d:l:p version with O:F version - split -devel package and add support for !suse targets * Wed Apr 8 2009 matejcik@suse.cz - update to 1.3.0 * official python 2.6 support (although it did (mostly) work before) * new multiarray APIs * Wed Feb 11 2009 ro@suse.de - fix filelist by using "--record-rpm" instead of "--record" (and do not quote lines starting with "%%dir ") * Wed Feb 4 2009 matejcik@suse.cz - package taken directly from devel:languages:python repository, original changelog follows: * Tue Dec 2 2008 James Oakley - 1.2.1-1 - Update to 1.2.1 * Mon Aug 11 2008 James Oakley - 1.1.1-1 - Update to 1.1.1: - Python 2.3.x fixes * Robert Kern -- remove development branch of f2py. * Pierre GM -- masked array, fix dictionary update. * Pierre GM -- masked array, fix itertools groupby dependency. * Pierre GM -- masked array, fix use of iterators. * Neil Muller -- fix test dependency on pretty print (#828). - General improvements * David Cournapeau -- scons build fixes/improvements. * Stefan van der Walt -- Howto document updates. * Fernando Perez -- cython updates. * Robert Kern -- PyPI metadata fixes. - Bug fixes * #791 -- Travis Oliphant, std/var not properly calling array_finalize. * #798 -- Stefan van der Walt, piecewise exposes raw memory. * #810 -- David Huard, Verbose argument not effective for ndarrays in assert_equal. * #828 -- Neil Muller, numpy file format format test failures on Solaris. * #825 -- Neil Muller/Charles Harris, SIGBUS in UNICODE_getattr on Sparc Linux. * #837 -- Pauli Virtanen, Infinite loop in fromfile & fromstring * #841 -- Charles Harris, make numpy complex types print like Python. * #843 -- Michael Abbott, reference count error in PyArray_CanCoerceScalar * #844 -- David Cournapeau, failure with numpy inner. * #848 -- Michael Abbott/Charles Harris/Travis Oliphant, reference leak. * #849 -- Michael Abbott, reference to deallocated object. * #850 -- Michael Abbott, memory leak. * #854 -- Pauli Virtanen, Non-standard branch cuts (arccosh). * #857 -- Pierre GM, dividing masked array with newaxis fails. * ------- David Cournapeau, ctypes load_library fix. * ------- Ryan May, Stefan van der Walt, fix loadtxt. * ------- Robert Kern, make numeric.py functions return booleans. * ------- Robert Kern, correct old numeric typecodes. * ------- Robert Kern, fix raising exception in tests. * ------- Robert Kern, make sure Zipf results are in allowable range. * ------- Robert Kern, fix c++ style comment.