Re: [pygame] BUG: pygame.draw.arc

[Alex Holkner <aholkner@xxxxxxxxxxxxxx>]

Peter Shinners wrote:

> On Wed, 2006-08-02 at 22:03 +1000, Alex Holkner wrote:
>  
>
> > If a start angle of 0 or math.pi*3/2 is given, the returned update 
> > rectangle does not include the right-most (for angle 0) or bottom-most 
> > (for angle pi*3/2) pixel.  Other start angles, with arcs that cover 
> > those same pixels, work correctly, as do end angles at those points.
> >    
>
> Hmm, do you have any guesses why? The arc drawing was all contributed
> code that I'm not framiliar with. If it is something simple like
> rounding I could probably tweak it a little. It would take me some time
> to get into the actual logic of it.
>  
Afraid not, I had a look for a few minutes but could see no reason; it 
seems to apply the same trig operations to each point regardless of 
whether it is start or end.  Internally it calls the line drawing 
function between each calculated pixel (yikes!), so possibly something 
is going wrong there.

The arc code in pygame-ctypes (nearly working) uses a different algorithm.

Alex.