[cairo] how to interprete extents of rotated path

[Theo Veenker]

Behdad Esfahbod wrote:
> 
> 
> On Mon, 2008-05-26 at 22:07 +0200, Theo Veenker wrote:
>  > Hi,
>  >
>  > I'm trying to get the bounding box of a path using 
> cairo_stroke_extents().
>  > The returned numbers are OK when the path is not rotated, but when it is
>  > rotated the bounding box returned doesn't make sense to me.
>  > See http://www.let.uu.nl/~Theo.Veenker/personal/extents/ for my test
>  > program and two images.
> 
> Ugh.  Didn't know about this.
> 
>  > What am I doing wrong here?
> 
> Nothing.  Wrong is the way we compute things internally.  It's the same
> effect that I explained here, in another context:
> 
>   http://lists.cairographics.org/archives/cairo/2008-May/014152.html
> 
> Just last night I was thinking about writing a stroke/path extents
> implementation that works on the double path directly, without any
> tesselation.  That can easily fix this problem.

That's good to know. I'll be happy to test things if you want.

Regards,
Theo