On Mon, Mar 7, 2011 at 4:22 PM, Dan Halbert <halb...@halwitz.org> wrote: > On 3/7/2011 6:48 PM, Christopher Barker wrote: >> On 3/7/11 3:36 PM, Dan Halbert wrote: >>> We currently have some straightforward NumPy code that indirectly >>> implements a C API defined by a third party. We built a Cython layer that >>> directly provides the API in a .a library, and then calls Python. The >>> layering looks like this: >>> >>> C main program -> API in Cython -> Python -> NumPy >>> >>> This is difficult to package for distribution, because of the Python and >>> NumPy dependencies. >> >> I'd say learn py2exe, py2app and friends, and be done with it. >> >> Otherwise, I wonder if you could re-factor your Cython code enough to >> remove all python dependencies -- or at least have something you could >> compile and run without a python interpreter running. > > What we ship is the library and associated files. The library gets > linked with someone else's main program. So py2exe and pyinstaller, etc. > can't help much here, since we don't control the executable. I have > thought about doing more in Cython, but I still need a libpython2.x.so, > etc., I believe.
Can't you link libpython and all of the other dependancies into a single .so file? This is the "embedding Python" idea--you can embed it in another library, you don't have to be the top-level executable. - Robert _______________________________________________ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion