[cairo] [PATCH 3/3] image: _cairo_image_surface_fixup_unbounded_boxes() fix for single box

[Taekyun Kim]

2011/6/21 Chris Wilson <chris at chris-wilson.co.uk>

> On Tue, 21 Jun 2011 16:23:50 +0900, Taekyun Kim <podain77 at gmail.com>
> wrote:
> >
> > Because _cairo_image_surface_fixup_unbounded() just clears area
> > between bounded and unbounded, we should give tight bounded extents
> > to it when boxes contains single box and there's no clip region.
>
> And here extents->bounded should already equal the box. So where is that
> discrepancy coming from?
> -Chris
>
> --
> Chris Wilson, Intel Open Source Technology Centre
>


extents->bounded is calculated using fill extents function.
If we have two overlapping rectangle paths, then it will be reduced to a
single
box by tessellation. But fill extents is calculated before tessellation, so
extents->bounded might be larger than actual filling geometry. Thus,
extents->bounded might not be the same even with a single box.
Please see question no. 4 in following thread.
http://lists.freedesktop.org/archives/cairo/2011-June/022027.html

-- 
Best Regards,
Taekyun Kim
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cairographics.org/archives/cairo/attachments/20110621/695161ae/attachment.htm>