[call for testing] clients: add cliptest program

[Pekka Paalanen]

On Thu, 06 Sep 2012 12:41:15 +0200
Birin Sanchez <birintxo at gmail.com> wrote:

> Hi Pekka,
> 
> I gave this clipping algorithm a try and I found a weird behaviour that 
> I don't know if it is expected of not. See attachment 1.
> 
> And this is how to reproduce it:
> 
>   * Move the clipping area to get it side by side with the surface, 
> you'll see the a green line between the two rectangles.
> * Change surface orientation pressing 'n' until you get to the initial 
> position
> 
> Apart from this the green line showed the intersection on all my tests.

Thank you, you really did find some problematic cases there. The first
green line you got should not appear, I think, since the rects are only
touching, not overlapping. And the half-height green line in your
screenshot could be result from rounding errors, but I'll check that,
too. I accidentally found yet another problem, when reproducing yours.


Good work!
- pq