python-scipy_1_2_1-gnu-hpc-1.2.1-bp152.1.64

- Use python2 dependency names.

- Fix Provides

- Fix rpmlint warnings

- Create python 2 backwards-compatibility version since 1.3.0+
  doesn't support python 2

- Update to 1.2.1
  * SyntaxError: Non-ASCII character 'xe2' in file scipy/stats/_continuous_distns.py on line 3346, but no encoding declared
  * Version 1.2.0 introduces `too many indices for array` error in `optimize.newton()`
  * scipy.stats.gaussian_kde normalizes the weights keyword argument externally.
  * scipy.linalg.qr_update gives NaN result
  * CI: Is scipy.scipy Windows Python36-32bit-full working?

- Use direct number in the Version tag

bsc#1130564: Apply update from the openSUSE package
- Properly create and tear down default version links when the
  HPC master packages are installed/uninstalled.
- Make use of %hpc_modules_init to make modules also known to
  client.
- Module file:
  * remove PATH element. Package has no binary,
  * make cosmetic changes.
- Remove use of %%python_module in dependency.

- Trim filler wording from description.

- Some futher changes:
  * Remove the use of fftw. The code doesn't link against it
    anywhere. For HPC we would have to build things separately
    for different MPI flavors as fftw3 exists only with HPC
    support there.
  * restructure the build process: since the environment for
    the right python version of Numpy needs to be loaded, wrap
    entire build (and install) in %%{python_expand: ..}.

- Add support for HPC builds:
  * Add _multibuild file
  * Add standard and gnu-hpc builds
  * Create initialization for both flavors to set the correct
    target directories in macros and replace install paths
    with these.
  * Restructure the build process.
  * Create 'master' packages for non-HPC builds.
  * Create environment module information,

- Update to version 1.2.0
  * Many changes. Please see changelog at:
    https://github.com/scipy/scipy/blob/v1.2.0/doc/release/1.2.0-notes.rst

- Fix build on SLE

- Update to version 1.1.0
  * Many changes. Please see changelog at:
    https://github.com/scipy/scipy/blob/v1.1.0/doc/release/1.1.0-notes.rst

- update to version 1.0.1:
  * Issues closed for 1.0.1
    + #7493: ndimage.morphology functions are broken with numpy 1.13.0
    + #8118: minimize_cobyla broken if disp=True passed
    + #8142: scipy-v1.0.0 pdist with metric=`minkowski` raises
    `ValueError:...
    + #8173: scipy.stats.ortho_group produces all negative
    determinants...
    + #8207: gaussian_filter seg faults on float16 numpy arrays
    + #8234: scipy.optimize.linprog interior-point presolve bug with
    trivial...
    + #8243: Make csgraph importable again via from scipy.sparse
    import*
    + #8320: scipy.root segfaults with optimizer 'lm'
  * Pull requests for 1.0.1
    + #8068: BUG: fix numpy deprecation test failures
    + #8082: BUG: fix solve_lyapunov import
    + #8144: MRG: Fix for cobyla
    + #8150: MAINT: resolve UPDATEIFCOPY deprecation errors
    + #8156: BUG: missing check on minkowski w kwarg
    + #8187: BUG: Sign of elements in random orthogonal 2D matrices in
    "ortho_group_gen"...
    + #8197: CI: uninstall oclint
    + #8215: Fixes Numpy datatype compatibility issues
    + #8237: BUG: optimize: fix bug when variables fixed by bounds are
    inconsistent...
    + #8248: BUG: declare "gfk" variable before call of terminate() in
    newton-cg
    + #8280: REV: reintroduce csgraph import in scipy.sparse
    + #8322: MAINT: prevent scipy.optimize.root segfault closes #8320
    + #8334: TST: stats: don't use exact equality check for hdmedian
    test
    + #8477: BUG: signal/signaltools: fix wrong refcounting in
    PyArray_OrderFilterND
    + #8530: BUG: linalg: Fixed typo in flapack.pyf.src.
    + #8566: CI: Temporarily pin Cython version to 0.27.3
    + #8573: Backports for 1.0.1
    + #8581: Fix Cython 0.28 build break of qhull.pyx

- Don't use openblas on m68k and riscv64

- Update to version 1.0.0
  * Many changes. Please see changelog at:
    https://github.com/scipy/scipy/blob/v1.0.0/doc/release/1.0.0-notes.rst#why-1-0-now
- Rebase no_implicit_decl.patch

- More rpmlint fixes.

- Update to version 0.19.1
  * #7214: Memory use in integrate.quad in scipy-0.19.0
  * #7258: linalg.matrix_balance gives wrong transformation matrix
  * #7262: Segfault in daily testing
  * #7273: scipy.interpolate._bspl.evaluate_spline gets wrong type
  * #7335: scipy.signal.dlti(A,B,C,D).freqresp() fails
  * #7211: BUG: convolve may yield inconsistent dtypes with method changed
  * #7216: BUG: integrate: fix refcounting bug in quad()
  * #7229: MAINT: special: Rewrite a test of wrightomega
  * #7261: FIX: Corrected the transformation matrix permutation
  * #7265: BUG: Fix broken axis handling in spectral functions
  * #7266: FIX 7262: ckdtree crashes in query_knn.
  * #7279: Upcast half- and single-precision floats to doubles in BSpline...
  * #7336: BUG: Fix signal.dfreqresp for StateSpace systems
  * #7419: Fix several issues in sparse.load_npz, save_npz
  * #7420: BUG: stats: allow integers as kappa4 shape parameters
- Add no_implicit_decl.patch
  Fixes implicit-pointer-decl warnings and implicit-fortify-decl error.
- Fix wrong-script-interpreter rpmlint error.

- Update to version 0.19.0
  + Highlights
  * A unified foreign function interface layer, `scipy.LowLevelCallable`.
  * Cython API for scalar, typed versions of the universal functions from
    the `scipy.special` module, via `cimport scipy.special.cython_special`.
- Removed weave subpackage. It was removed upstream in this release.

- Switch to single-spec version
- update to version 0.18.1:
  * #6357: scipy 0.17.1 piecewise cubic hermite interpolation does not
    return...
  * #6420: circmean() changed behaviour from 0.17 to 0.18
  * #6421: scipy.linalg.solve_banded overwrites input 'b' when the
    inversion...
  * #6425: cKDTree INF bug
  * #6435: scipy.stats.ks_2samp returns different values on different
    computers
  * #6458: Error in scipy.integrate.dblquad when using variable
    integration...
  * #6405: BUG: sparse: fix elementwise divide for CSR/CSC
  * #6431: BUG: result for insufficient neighbours from cKDTree is
    wrong.
  * #6432: BUG Issue #6421: scipy.linalg.solve_banded overwrites input
    'b'...
  * #6455: DOC: add links to release notes
  * #6462: BUG: interpolate: fix .roots method of PchipInterpolator
  * #6492: BUG: Fix regression in dblquad: #6458
  * #6543: fix the regression in circmean
  * #6545: Revert gh-5938, restore ks_2samp
  * #6557: Backports for 0.18.1
- update to version 0.18.0:
  (see http://scipy.github.io/devdocs/release.0.18.0.html for full changelog)
  * Highlights of this release include:
    + A new ODE solver for two-point boundary value problems,
    scipy.optimize.solve_bvp.
    + A new class, CubicSpline, for cubic spline interpolation of
    data.
    + N-dimensional tensor product polynomials,
    scipy.interpolate.NdPPoly.
    + Spherical Voronoi diagrams, scipy.spatial.SphericalVoronoi.
    + Support for discrete-time linear systems, scipy.signal.dlti.
- update to version 0.17.1:
  * #5817: BUG: skew, kurtosis return np.nan instead of "propagate"
  * #5850: Test failed with sgelsy
  * #5898: interpolate.interp1d crashes using float128
  * #5953: Massive performance regression in cKDTree.query with L_inf
    distance...
  * #6062: mannwhitneyu breaks backward compatibility in 0.17.0
  * #6134: T test does not handle nans
  * #5902: BUG: interpolate: make interp1d handle np.float128 again
  * #5957: BUG: slow down with p=np.inf in 0.17 cKDTree.query
  * #5970: Actually propagate nans through stats functions with
    nan_policy="propagate"
  * #5971: BUG: linalg: fix lwork check in *gelsy
  * #6074: BUG: special: fixed violation of strict aliasing rules.
  * #6083: BUG: Fix dtype for sum of linear operators
  * #6100: BUG: Fix mannwhitneyu to be backward compatible
  * #6135: Don't pass null pointers to LAPACK, even during workspace
    queries.
  * #6148: stats: fix handling of nan values in T tests and kendalltau
- specfile:
  * updated source url to files.pythonhosted.org
  * require setuptools
  * Add openBLAS support.
    This can improve performance in many situations.
  * Drop ATLAS support.