[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: [pygame] [Announce] Python bindings for the bitmask collision detection library

It's in pygame now.

There's still a few things I need to do.

- make constructor which takes a surface.
- a 'make_surface' method in C.  should take a threshold for alpha.
- complete the documentation (now in src/mask.doc)
- complete the unittests (test/test/mask_test.py)
- make set_at and get_at raise IndexError when wrong index is given.
- double check the reference counting.
- integrate Mask into examples/testsprite.py


JKrukoff has an idea about how to change the sprite classes so they
can be extended to use different types of collision detection.  He's
going to write a proposal to the mailing list about it later.




On 5/9/07, René Dudfield <renesd@xxxxxxxxx> wrote: 
hi,

I'm in irc now putting bitmask into pygame.

illume is my nick.

Cheers,



On 5/6/07, Ulf Ekström <uekstrom@xxxxxxxxx> wrote:
> Hi.
>
> > Bitmask as a separate module is a good idea.  I think your pygame.mask
> > is a good name.  Then we can provide some pygame.sprite mixins, or
> > methods to help there.
> >
> > - pygame.mask module.
> > - move docs from the source to .doc file.
> > - some unittests in the test/ directory.
>
> Ok, I can do this.
>
> > - move your pygame example to examples/bitmask.py or maybe...
> > bouncy_bitmask.py ?
>
> Sure, no problem. I have not even looked at the pygame source tree, so
> I don't know how you organize stuff. I will have a look.
>
> > I'm not sure about a surface method.  It might simplify things for
> > some people though.  However then it makes surface depend on the
> > bitmask module (not necessarily bad).
>
> It is more complicated, but I can imagine it makes it easier to use
> for beginners. Anyway, I cannot really help with this part so I leave
> it up to you guys.
>
> > Has it been tested on big endian machines?  eg ppc macs?  It looks to
> > have checks in there for 64bit machines - has it been tested on 64bit
> > machines too?
>
> Yes. The library is used in my game 'Airstrike', which has been tested
> on a lot of platforms: sparc, ppc, i386 and even some old amiga
> platform. So I think it works. I develop on ppc and i386.
>
> > 'maskFromSurface' might be a constructor argument?  This would be so
> > common, that we should do our own implementation.
>
> Yes. It is also much faster in C than in Python. I can't really think
> of a good set of arguments to the constructor that allows for either
> (width,height) or a surface, though.
>
>  Do you think it
> > would be very different for different games?  eg, using different
> > alpha levels as 'empty'.  5% alpha might be considered to be gone by
> > some games/sprites.  If you think that's useful, an alpha_threshold
> > argument might be appropriate.
>
> I think it's a good idea. It can default to 50%.
>
> > As you suggest, there's lots of other uses for bitmasks, other than collision.
> >
> > Blitting surfaces masked with a bitmask might be a future application.
> >  Another future application might be to reduce over draw - by using
> > the bitmask to show which pixels we have updated.
> >
> > A mask.setrange() method might be useful.  So you can set a range of
> > bits at once, quickly.
>
> Perhaps this can wait until the next release, I think collision
> detection is the main feature, and it may seem overly complicated if
> we add a lot of extra functions. But again, I think you can decide
> this best.
>
> I will start working on this and try to be on irc next wednesday.
>
> Regards,
> UIf
>