[Webkit-unassigned] [Bug 19652] [CAIRO] wrong drawing of border-image

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=19652


alp at nuanti.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |alp at nuanti.com




------- Comment #9 from alp at nuanti.com  2008-09-24 18:54 PDT -------
(In reply to comment #8)
> Created an attachment (id=23291)
 --> (https://bugs.webkit.org/attachment.cgi?id=23291&action=view) [edit]
> Cairo border-image
> 
> cairo_surface's are ref-counted in cairo. Thats why I put auto_ptr<ImageBuffer>
> above the if-clause. Otherwise the image is cleaned-up to early and I get a
> ref-count error (ref-count lower than zero).
> But I measured the call and it takes only 0.000047s extra time as worst-case
> (normaly 0.000009s).
> I don't know (or understand) what you mean with nativeImageForCurrentFrame().
> I need a cairo_suface_t* out of the imageBuffer and
> nativeImageForCurrentFrame() can give it.
> 
> I haven't r? the patch, to discuss the changes again.
> 

Maybe you could do something like this (completely untested) instead of using
ImageBuffer?

    cairo_pattern_t* pattern;
    if (tileRect.size() == size())
        pattern = cairo_pattern_create_for_surface(image);
    else {
        cairo_push_group(cr);
        //IntRect imageSize = enclosingIntRect(tileRect);
        cairo_set_source_surface(cr, image, -tileRect.x(), -tileRect.y());
        cairo_paint(cr);
        pattern = cairo_pop_group(cr);
    }


-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.