[cairo] [cairo-announce] cairo snapshot 1.7.6 now available

Carl Worth cworth at cworth.org
Wed Sep 24 12:20:47 PDT 2008

On Sun, 2008-09-21 at 12:44 +0200, Theo Veenker wrote:
> A big thank you to all developers for the continuing improvement of
> cairo!

Hi Theo,

Thank you for your testing and feedback.

> I've tested 1.7.6 with my code and I still notice some problems.
> Cairo_stroke_extents() doesn't yield correct results for rotated
> or upscaled paths.
> Refs:
>    http://lists.cairographics.org/archives/cairo/2008-May/014207.html
> and
>    http://lists.cairographics.org/archives/cairo/2007-February/009702.html
> This is a serious problem for me as I rely on extents calculation when
> updating a partly changed complex drawing (only redrawing damaged areas).

Oops! That's obviously a really old bug---with multiple reports in the
past, and we just haven't fixed it. Sorry about that!

This (and the other issues you mentioned) just aren't going to make it
into 1.8.0, but hopefully we can do 1.8.2 with these soon.

> And the following are less important I think, but since my reports
> about them where ignored I like to annoy you again. About filling
> a path with a surface pattern:
> - the SVG backend does 'repeat' for CAIRO_EXTEND_NONE and CAIRO_EXTEND_PAD
> - PDF and PS backends don't implement CAIRO_EXTEND_PAD

I'm pretty sure Adrian is looking into this for PDF at least.

> - for a rotated shape filled with a surface pattern with CAIRO_EXTEND_NONE
>    the XLIB backend for some rotate angles erases the bounding box of
>    the filled part to black, thus leaving a black rectangle around the
>    filled part within the shape
> Ref:
>    http://lists.cairographics.org/archives/cairo/2008-August/014922.html

This is likely a driver-specific bug. And I happen to have an open bug
report against the Intel driver for this, so I'll be fixing that soon.

Thanks again,


More information about the cairo mailing list