[Numpy-discussion] Getting rid of annoying weave nag (original) (raw)
Fernando Perez fperez.net at gmail.com
Fri Jun 9 18:21:00 EDT 2006
- Previous message (by thread): [Numpy-discussion] Getting rid of annoying weave nag
- Next message (by thread): [Numpy-discussion] Building shared libraries with numpy.distutils
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On 6/9/06, David M. Cooke <cookedm at physics.mcmaster.ca> wrote:
> This difference is so dramatic that I think a message is justified > (absent a proper logging framework). It's helpful to know that the > time is going into c++ compilation, and not your code hanging for 30 > seconds.
Ok, I'll give you that one :-) It's the other 1000 uses of print that I'm concerned about. inlinetools.compilefunction takes a verbose flag, though, which eventually gets passed to buildtools.buildextension (which I believe does all the compiling for weave). It's probably more reasonable to have inlinetools.compilefunction default to verbose=1 instead of 0, then buildextension will print 'Compiling code...' (that should be changed to mention weave).
I failed to mention that I agree with you: the proper solution is to use logging for this. For now I'll commit the strict-prototypes fix, and if I find myself with a lot of spare time, I'll try to clean things up a little bit to use logging (there's already a logger instance running in there).
Cheers,
f
- Previous message (by thread): [Numpy-discussion] Getting rid of annoying weave nag
- Next message (by thread): [Numpy-discussion] Building shared libraries with numpy.distutils
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]