[cairo] SVG backend bug?

Edward Zimmermann Edward.Zimmermann at cib.de
Wed Jul 17 01:56:52 PDT 2013

Its being produced by rendering PDF. The test PDF has been attached to this mail.
Its interesting to also note that Poppler renders it correctly as PNG but its SVG
completely fails (using pdftocairo -png resp. -svg). This too seems to indicate
problems in the SVG backend as our front-ends are fully and completely different.

Using Cairo I've rendered it through a number of back-ends..
Image surface [OK]
PS surface [OK]
PDF surface [OK]
SVG suface [FAILS]

What is also interesting is that if I do a write PNG from the SVG surface the result is
OK (low quality but placement is correct). This seems to indicate that the SVG writer is
messing up.

-----Ursprüngliche Nachricht-----
Von: Bryce W. Harrington [mailto:b.harrington at samsung.com] 
Gesendet: Mittwoch, 17. Juli 2013 01:24
An: Edward Zimmermann
Cc: cairo at cairographics.org
Betreff: Re: [cairo] SVG backend bug?

On Tue, Jul 16, 2013 at 02:55:21PM +0200, Edward Zimmermann wrote:
> Can't seem to figure out what's wrong... I'm OK your OK.. I'm not OK.. Or perhaps..
> there is a buglet in the SVG backend.
> Rendering to Image is correct. Redering to PDF is correct. Rendering to SVG is wrong.
> Enclosed are the trace, XML dump, PNG (what is should be) and the SVG.
> Looks like the transformations are getting messed up...

Yeah, like it's inverted and rotated 180 degrees or something.

> Or am I doing something WRONG?????

To answer that, it might be instructive to know more step-by-step how
you're producing this composition.  Possibly if the matrix is
constructed differently or avoided in favor of simpler transformations
you may be able to work around it.


Bryce Harrington
Senior Open Source Developer  -  b.harrington at samsung.com
Open Source Group             -  Samsung Research America
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 202pages-1.pdf
Type: application/pdf
Size: 11215 bytes
Desc: 202pages-1.pdf
URL: <http://lists.cairographics.org/archives/cairo/attachments/20130717/1a3479ac/attachment-0001.pdf>

More information about the cairo mailing list