[Numpy-discussion] [SciPy-dev] NumPy 1.0.3.x and SciPy 0.5.2.x
Wed Aug 29 16:53:12 CDT 2007
Ivan Pan wrote:
> On 8/19/07, Ondrej Certik <email@example.com> wrote:
>> I don't know what the native way of installing packages on Mac OS
>> X is
Boy, I wish this weren't such a mess. Quite some time ago, a bunch of us
on the pythonmac list tried to establish the idea of a "one 'standard'
python for OS-X", and a set of pre-built packages for it. It is the one
you find here:
That Python is the same as the one you find at python.org too. It is the
closest one comes to a "native" set of packages for OS-X.
It would be really nice if the scipy/numpy projects would provide
binaries (or at least have setup.py ready to go) for that repository. I
really like being able to tell folks ONE place to go to get python
packages. There are a number of Mac folks that help build the packages
For a while, SciPy had a key problem -- no one knew how to build
Universal(Intel and PPC) packages from Fortran code, and that repository
really should have Universal binaries, so that folks don't have to think
about what hardware they are running, and can bundle up apps with Py2App
that will work on any Mac (with a new enough OS).
I understand that the Universal problem has been solved now. I hope that
if the SciPy project "officially" releases binaries for OS-X, they will
be Universal binaries compatible with that Python.
About fink/macport. They are fine systems that have some real use.
However, they really should be thought of as different platforms (or at
least different distributions), much like CygWin, or Ubuntu vs. Fedora.
If you make a fink package, you are making a fink package, NOT an OS-X one.
Anyway, sorry to be so pedantic when I don't think I will get the time
to do the building myself, but I wanted to lay out a goal anyway. I do
offer to do some testing, etc if needed.
Christopher Barker, Ph.D.
Emergency Response Division
NOAA/NOS/OR&R (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
More information about the Numpy-discussion