Hi, On Thu, Feb 7, 2013 at 12:29 PM, Chris Barker - NOAA Federal <chris.barker@noaa.gov> wrote:
On Thu, Feb 7, 2013 at 11:38 AM, Matthew Brett <matthew.brett@gmail.com> wrote:
a) If we cannot build Scipy now, it may or may not be acceptable to release numpy now. I think it is, you (Ralf) think it isn't, we haven't discussed that. It may not come up.
Is anyone suggesting we hold the whole release for this? I fnot, then why would you say we shouldnt' put out a 64 bit Windows numpy binary because we dont' ahve a matching Scipy one? folks that need Scipy will have to find a way to built it themselves anyway, and folks that use numpy without scipy would have a nice binary to use -- That seems like value-added to me.
Right - me too - but we could hold off that question until Ondrej has had a chance to build both I guess?
d) It may or may not be acceptable to add the DLL directory to the PATH on numpy import. David says not, Christophe disagrees, we haven't really discussed that.
I like the "load them with ctypes" approach better, but I don't know how much effort it would be to implement that.
I'm still not totally clear if there is an issue with buildings other extensions tht rely on numpy/scipy -- can they be built with mingGW if numpy/scipy were built with MSVC (Or any other compiler for that matter).
I had the impression that the problem with Mingw64 was the need to load Mingw DLLs at run time - if that's the same for MSVC / MKL, maybe there's no advantage to using the MS tools? But I don't understand the issues very well. Cheers, Matthew