%define hdf5dir %_libdir/hdf5-seq
%define oname tables
%def_with python3
%def_disable check
Name: py%oname
Version: 3.2.2
Release: alt1.dev0.git20150828
Epoch: 1
Summary: Managing hierarchical datasets
License: MIT
Group: Development/Python
Url: http://www.pytables.org/
Packager: Eugeny A. Rostovtsev (REAL) <real at altlinux.org>
# https://github.com/PyTables/PyTables.git
# branch: develop
Source: %name-%version.tar.gz
Requires: python-module-%oname = %{?epoch:%epoch:}%version-%release
BuildRequires(pre): rpm-build-python
BuildPreReq: python-module-numpydoc python-module-numpy-addons
BuildPreReq: python-module-lxml python-module-numpy
BuildPreReq: python-module-docutils python-module-matplotlib
BuildPreReq: python-module-sphinx
BuildPreReq: python-module-numexpr-tests
BuildPreReq: python-devel python-module-Pyrex libnumpy-devel
BuildPreReq: libhdf5-devel liblzo2-devel bzlib-devel
BuildPreReq: xsltproc inkscape fop
BuildPreReq: java-devel-default docbook-tldp-xsl docbook-dtds
BuildPreReq: w3c-markup-validator-libs python-module-Cython
BuildPreReq: python-module-numexpr python-module-setuptools
#BuildPreReq: texlive-latex-recommended libblosc-devel
BuildPreReq: libblosc-devel
%if_with python3
BuildRequires(pre): rpm-build-python3
BuildRequires: python3-devel libnumpy-py3-devel python-tools-2to3
BuildPreReq: python3-module-distribute python3-module-Cython
BuildPreReq: python3-module-numexpr-tests
%endif
%description
PyTables is a package for managing hierarchical datasets and designed
to efficiently and easily cope with extremely large amounts of data.
PyTables is built on top of the HDF5 library, using the Python language
and the NumPy package (it also supports numarray and Numeric right
out-of-the-box). It features an object-oriented interface that, combined
with C extensions for the performance-critical parts of the code
(generated using Pyrex), makes it a fast, yet extremely easy to use tool
for interactively dealing with, processing and searching very large
amounts of data. One important feature of PyTables is that it optimizes
memory and disk resources so that data takes much less space (specially
if on-flight compression is used) than other solutions such as
relational or object oriented databases.
%if_with python3
%package py3
Summary: Managing hierarchical datasets (Python 3)
Group: Development/Python3
Requires: python3-module-%oname = %{?epoch:%epoch:}%version-%release
%description py3
PyTables is a package for managing hierarchical datasets and designed
to efficiently and easily cope with extremely large amounts of data.
PyTables is built on top of the HDF5 library, using the Python language
and the NumPy package (it also supports numarray and Numeric right
out-of-the-box). It features an object-oriented interface that, combined
with C extensions for the performance-critical parts of the code
(generated using Pyrex), makes it a fast, yet extremely easy to use tool
for interactively dealing with, processing and searching very large
amounts of data. One important feature of PyTables is that it optimizes
memory and disk resources so that data takes much less space (specially
if on-flight compression is used) than other solutions such as
relational or object oriented databases.
%package -n python3-module-%oname
Summary: Managing hierarchical datasets (Python 3)
Group: Development/Python3
%add_python3_req_skip numarray Scientific
%description -n python3-module-%oname
PyTables is a package for managing hierarchical datasets and designed
to efficiently and easily cope with extremely large amounts of data.
PyTables is built on top of the HDF5 library, using the Python language
and the NumPy package (it also supports numarray and Numeric right
out-of-the-box). It features an object-oriented interface that, combined
with C extensions for the performance-critical parts of the code
(generated using Pyrex), makes it a fast, yet extremely easy to use tool
for interactively dealing with, processing and searching very large
amounts of data. One important feature of PyTables is that it optimizes
memory and disk resources so that data takes much less space (specially
if on-flight compression is used) than other solutions such as
relational or object oriented databases.
This package contains python module of PyTables.
%package -n python3-module-%oname-tests
Summary: Tests and examples for PyTables (Python 3)
Group: Development/Python3
%add_python3_req_skip numarray
Requires: python3-module-%oname = %EVR
%description -n python3-module-%oname-tests
PyTables is a package for managing hierarchical datasets and designed
to efficiently and easily cope with extremely large amounts of data.
PyTables is built on top of the HDF5 library, using the Python language
and the NumPy package (it also supports numarray and Numeric right
out-of-the-box). It features an object-oriented interface that, combined
with C extensions for the performance-critical parts of the code
(generated using Pyrex), makes it a fast, yet extremely easy to use tool
for interactively dealing with, processing and searching very large
amounts of data. One important feature of PyTables is that it optimizes
memory and disk resources so that data takes much less space (specially
if on-flight compression is used) than other solutions such as
relational or object oriented databases.
This package contains tests and examples for PyTables.
%package -n python3-module-%oname-bench
Summary: Benchmarks for PyTables (Python 3)
Group: Development/Python3
%add_python3_req_skip numarray chararray recarray recarray2 Numeric psyco
Requires: python3-module-%oname = %EVR
%description -n python3-module-%oname-bench
PyTables is a package for managing hierarchical datasets and designed
to efficiently and easily cope with extremely large amounts of data.
PyTables is built on top of the HDF5 library, using the Python language
and the NumPy package (it also supports numarray and Numeric right
out-of-the-box). It features an object-oriented interface that, combined
with C extensions for the performance-critical parts of the code
(generated using Pyrex), makes it a fast, yet extremely easy to use tool
for interactively dealing with, processing and searching very large
amounts of data. One important feature of PyTables is that it optimizes
memory and disk resources so that data takes much less space (specially
if on-flight compression is used) than other solutions such as
relational or object oriented databases.
This package contains benchmarks for PyTables.
%endif
%package doc
Summary: Documentation for PyTables
Group: Development/Documentation
BuildArch: noarch
%description doc
PyTables is a package for managing hierarchical datasets and designed
to efficiently and easily cope with extremely large amounts of data.
PyTables is built on top of the HDF5 library, using the Python language
and the NumPy package (it also supports numarray and Numeric right
out-of-the-box). It features an object-oriented interface that, combined
with C extensions for the performance-critical parts of the code
(generated using Pyrex), makes it a fast, yet extremely easy to use tool
for interactively dealing with, processing and searching very large
amounts of data. One important feature of PyTables is that it optimizes
memory and disk resources so that data takes much less space (specially
if on-flight compression is used) than other solutions such as
relational or object oriented databases.
This package contains documentation for PyTables.
%package -n python-module-%oname
Summary: Managing hierarchical datasets
Group: Development/Python
%setup_python_module %oname
%add_python_req_skip numarray
%description -n python-module-%oname
PyTables is a package for managing hierarchical datasets and designed
to efficiently and easily cope with extremely large amounts of data.
PyTables is built on top of the HDF5 library, using the Python language
and the NumPy package (it also supports numarray and Numeric right
out-of-the-box). It features an object-oriented interface that, combined
with C extensions for the performance-critical parts of the code
(generated using Pyrex), makes it a fast, yet extremely easy to use tool
for interactively dealing with, processing and searching very large
amounts of data. One important feature of PyTables is that it optimizes
memory and disk resources so that data takes much less space (specially
if on-flight compression is used) than other solutions such as
relational or object oriented databases.
This package contains python module of PyTables.
%package -n python-module-%oname-tests
Summary: Tests and examples for PyTables
Group: Development/Python
%add_python_req_skip numarray queue
Requires: python-module-%oname = %EVR
%description -n python-module-%oname-tests
PyTables is a package for managing hierarchical datasets and designed
to efficiently and easily cope with extremely large amounts of data.
PyTables is built on top of the HDF5 library, using the Python language
and the NumPy package (it also supports numarray and Numeric right
out-of-the-box). It features an object-oriented interface that, combined
with C extensions for the performance-critical parts of the code
(generated using Pyrex), makes it a fast, yet extremely easy to use tool
for interactively dealing with, processing and searching very large
amounts of data. One important feature of PyTables is that it optimizes
memory and disk resources so that data takes much less space (specially
if on-flight compression is used) than other solutions such as
relational or object oriented databases.
This package contains tests and examples for PyTables.
%package -n python-module-%oname-bench
Summary: Benchmarks for PyTables
Group: Development/Python
%add_python_req_skip numarray chararray recarray recarray2 Numeric psyco
Requires: python-module-%oname = %EVR
%description -n python-module-%oname-bench
PyTables is a package for managing hierarchical datasets and designed
to efficiently and easily cope with extremely large amounts of data.
PyTables is built on top of the HDF5 library, using the Python language
and the NumPy package (it also supports numarray and Numeric right
out-of-the-box). It features an object-oriented interface that, combined
with C extensions for the performance-critical parts of the code
(generated using Pyrex), makes it a fast, yet extremely easy to use tool
for interactively dealing with, processing and searching very large
amounts of data. One important feature of PyTables is that it optimizes
memory and disk resources so that data takes much less space (specially
if on-flight compression is used) than other solutions such as
relational or object oriented databases.
This package contains benchmarks for PyTables.
%prep
%setup
%if_with python3
rm -rf ../python3
cp -a . ../python3
find ../python3 -type f -name '*.py' -exec \
sed -i 's|#!/usr/bin/env python|#!/usr/bin/env python3|' '{}' +
find ../python3 -type f -name '*.py' -exec \
sed -i 's|#!/usr/bin/python|#!/usr/bin/python3|' '{}' +
find ../python3 -type f -name '*.py' -exec 2to3 -w -n '{}' +
%endif
%build
%add_optflags -fno-strict-aliasing
export NPY_NUM_BUILD_JOBS=%__nprocs
%python_build_debug --hdf5=%hdf5dir
%if_with python3
pushd ../python3
%python3_build_debug --hdf5=%hdf5dir
popd
%endif
%install
%if_with python3
pushd ../python3
%python3_install --hdf5=%hdf5dir --root=%buildroot
cp -fR examples %buildroot%python3_sitelibdir/%oname/
cp -fR bench contrib %buildroot%python3_sitelibdir/%oname/
popd
pushd %buildroot%_bindir
for i in $(ls); do
mv $i $i.py3
done
popd
%endif
%python_install --hdf5=%hdf5dir --root=%buildroot
export PYTHONPATH=%buildroot%python_sitelibdir
%make_build -C doc pickle
%make_build -C doc html
install -d %buildroot%_docdir/%name/pdf
install -p -m644 LICENSE.txt README.rst RELEASE_NOTES.txt THANKS \
%buildroot%_docdir/%name
cp -fR LICENSES %buildroot%_docdir/%name
#install -p -m644 doc/build/latex/*.pdf %buildroot%_docdir/%name/pdf
cp -fR doc/build/html %buildroot%_docdir/%name/
cp -fR examples %buildroot%python_sitelibdir/%oname/
cp -fR bench contrib %buildroot%python_sitelibdir/%oname/
%check
%make check
%if_with python3
pushd ../python3
%make check PYTHON=python3
popd
%endif
%files
%_bindir/*
%if_with python3
%exclude %_bindir/*.py3
%endif
%files -n python-module-%oname
%python_sitelibdir/*
%exclude %python_sitelibdir/%oname/examples
%exclude %python_sitelibdir/%oname/tests
%exclude %python_sitelibdir/%oname/*/tests
%exclude %python_sitelibdir/%oname/bench
%files -n python-module-%oname-tests
%python_sitelibdir/%oname/examples
%python_sitelibdir/%oname/tests
%python_sitelibdir/%oname/*/tests
%files -n python-module-%oname-bench
%python_sitelibdir/%oname/bench
%if_with python3
%files py3
%_bindir/*.py3
%files -n python3-module-%oname
%python3_sitelibdir/*
%exclude %python3_sitelibdir/%oname/examples
%exclude %python3_sitelibdir/%oname/tests
%exclude %python3_sitelibdir/%oname/*/tests
%exclude %python3_sitelibdir/%oname/bench
%files -n python3-module-%oname-tests
%python3_sitelibdir/%oname/examples
%python3_sitelibdir/%oname/tests
%python3_sitelibdir/%oname/*/tests
#files -n python3-module-%oname-bench
#python3_sitelibdir/%oname/bench
%endif
%files doc
%_docdir/%name
%changelog
* Sun Aug 30 2015 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:3.2.2-alt1.dev0.git20150828
- New snapshot
* Thu Aug 20 2015 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:3.2.2-alt1.dev0.git20150817
- Version 3.2.2.dev0
* Tue Apr 28 2015 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:3.1.2-alt1.dev.git20150418
- New snapshot
* Mon Nov 03 2014 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:3.1.2-alt1.dev.git20141012
- Version 3.1.2dev
* Sat Aug 02 2014 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:3.1.1-alt3.git20140325
- Added module for Python 3
* Thu Jul 10 2014 Igor Vlasenko <viy@altlinux.ru> 1:3.1.1-alt2.git20140325
- NMU: corrected java deps
* Wed May 07 2014 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:3.1.1-alt1.git20140325
- Version 3.1.1
* Thu Oct 24 2013 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:3.0.0-alt3.git20130601
- Rebuilt with updated NumPy
* Tue Jul 02 2013 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:3.0.0-alt2.git20130601
- Rebuilt with new libhdf5
* Sun Jun 16 2013 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:3.0.0-alt1.git20130601
- Version 3.0.0
* Tue Oct 16 2012 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:2.4.0-alt2.git20120720
- Rebuilt with updated NumPy
* Thu Sep 20 2012 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:2.4.0-alt1.git20120720
- Version 2.4.0
* Tue Jul 10 2012 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:2.3.1-alt1.git20120613
- New snapshot
- Applied repocop patch
* Sun May 13 2012 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:2.3.1-alt1.git20120318
- New snapshot
* Sat Dec 17 2011 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 1:2.3.1-alt1.git20111114
- Version 2.3.1
* Sat Oct 22 2011 Vitaly Kuznetsov <vitty@altlinux.ru> 2.3b1-alt1.svn20110408.1.1
- Rebuild with Python-2.7
* Thu Sep 08 2011 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.3b1-alt1.svn20110408.1
- Rebuilt with libhdf5-7
* Sat Apr 16 2011 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.3b1-alt1.svn20110408
- New snapshot
* Mon Mar 21 2011 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.3b1-alt1.svn20101118.1
- Rebuilt for debuginfo
* Sun Nov 21 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.3b1-alt1.svn20101118
- New snapshot
* Thu Oct 28 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.3b1-alt1.svn20100707.3
- Rebuilt for soname set-versions
* Tue Oct 12 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.3b1-alt1.svn20100707.2
- Fixed underlinking
* Wed Aug 04 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.3b1-alt1.svn20100707.1
- Avoid requirement on RandomArray
* Mon Jul 12 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.3b1-alt1.svn20100707
- Version 2.3b1
- Rebuilt with docbook-style-xsl 1.75.2
* Thu Feb 04 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.2b3-alt1.svn20100203
- Version 2.2b3
* Thu Feb 04 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.2b2-alt1.svn20090911.4
- Fixed creating some extensions
- Added docs, benchmarks and tests/examples
* Wed Feb 03 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.2b2-alt1.svn20090911.3
- Rebuilt with reformed NumPy
* Sat Jan 02 2010 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.2b2-alt1.svn20090911.2
- Rebuilt without python-module-Numeric
* Wed Nov 18 2009 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.2b2-alt1.svn20090911.1
- Rebuilt with python 2.6
* Fri Sep 11 2009 Eugeny A. Rostovtsev (REAL) <real at altlinux.org> 2.2b2-alt1.svn20090911
- Initial build for Sisyphus