Re: [pygame] BUG: required modules not raising error when failing to import was Re: [pygame] Surfarray.array_alpha() bugfix, py2exe/py2app segfault

[Zack Schilling <zack.schilling@xxxxxxxxx>]

I don't think manually including bufferproxy when it might not be  
needed is a good solution, nor do I think getting py2exe to search for  
compiled C modules calling other compiled C modules is likely to happen.

I think some documentation or a quick exception when the package fails  
to load would be sufficient.  Wouldn't it make more sense to change  
"import_pygame_bufferproxy()" in pygame.h to throw an exception with a  
helpful message?

"Required library, pygame.bufferproxy, faliled to import. Check your  
pygame installation or try manually include this library."

-Zack

On Feb 5, 2009, at 3:04 PM, Brian Fisher wrote:

> On Wed, Feb 4, 2009 at 10:25 PM, <mva@xxxxxxxxxxxx> wrote:
> Brian Fisher <brian@xxxxxxxxxxxxxxxxxxx>:
> I thinks it's very important that packaging tools work out of the  
> box with
> pygame and have no mysterious failures. If it takes a hack, it takes  
> a hack.
>
> I think, packaging tools should do their job right and do not behave  
> wrong.
>
> I agree with that completely. Lets help them by making our package  
> dependencies more tractable :)
>
>
>
> If there are alternatives to the hack, they should be used. But not  
> solving
> the problem is not an alternative.
>
> That's not what I wrote, if you'd have read my mail. I wrote that we  
> should
> fix that on the user-kevel with providing a matching script for  
> py2exe.
>
> I'm sorry, I didn't mean to misconstrue what you said. The "problem"  
> I was referring to was py2exe/py2app working "out of the box". I  
> know you were offering a constructive solution, I just want to solve  
> a slightly different problem