- Adjust dependencies, 'numpy/core/multiarray.so' is also linked to
math/openblas (reported by QA script)
- Add patches in order to detect 'sys/endian.h' header
Based on the work done by Eijiro Shibusawa <phd_kimberlite@yahoo.co.jp>,
OpenBLAS port maintainer.
Hide py-nose dependency under new TESTS option.
Convert to USES=blaslapack.
Differential Revision: D2060
Reviewed by: koobs, thierry
versions of FreeBSD, we not support anymore.
- remove LATEST_LINK. There is PYTHON_PKGNAMEPREFIX set and in future
we may have this package built against different python versions.
- clarify LICENSE (BSD3C)
- suppor staging
while here:
- utilize PYTHON_PY3K_PLIST_HACK
- install docs uncoditionally into stage area
- convert LIB_DEPENDS to the new notation
- sort packing list
PR: 186677
Submitted by: Johannes Jost Meixner <xmj@chaot.net>
correctly detect and depend on it in math/py-numpy
* Replace the old numpy documentation with the new numpy book
* Explicitly tell the distutil to use gnu95 fortran compiler
* Use ${PYTHONPREFIX_SITELIBDIR} rather than ${PYTHON_SITELIBDIR} when
generating the packing list for science/py-scipy
* Bump PORTREVISIONs, for packing lists are changed
PR: ports/136058
Submitted by: Li-Lun Wang (Leland Wang) <llwang AT infor.org> (maintainer)
on amd64.
A better fix would be to set pic_flags instead, and have scipy build a
shared object for dfftpack instead a static archive, but I didn't figure
out how. [1]
- Pass maintainership to llwang@infor.org
PR: ports/112429 [1]
Submitted by: Li-Lun "Leland" Wang <llwang@infor.org> (new maintainer)
- Add options for ATLAS
- Fix ATLAS detection
- Use feclearexcept(3) rather than fpsetsticky(3) on 5.3+
- Add CONFLICTS for f2py. Both ports install bin/f2py
PR: ports/103827
Submitted by: Li-Lun Wang
Approved by: maintainer timeout (6 weeks)
called NumPy. This package contains:
* a powerful N-dimensional array object
* sophisticated (broadcasting) functions
* basic linear algebra functions
* basic Fourier transforms
* sophisticated random number capabilities
* tools for integrating Fortran code.
NumPy derives from the old Numeric code base and can be used as a
replacement for Numeric. It also adds the features introduced by numarray
and can also be used to replace numarray.
Note: Development for Numeric has ceased, and users should transisition to
NumPy as quickly as possible.
WWW: http://numpy.scipy.org/
PR: ports/102458
Submitted by: Tony Maher <anthony.maher@uts.edu.au>
