We have a number of places in the SVG code where we create new surfaces and just assume no problems will have occurred. We should check for gfxFooSurface constructor failure by calling gfxASurface::CairoStatus() on new surfaces before we use them.
Bah. Now that I actually _look_ we do at least check Data(). (Could have sworn I saw code without checks before.) http://mxr.mozilla.org/seamonkey/search?string=new+gfx&find=svg&findi=&filter=&tree=seamonkey Maybe we should be checking CairoStatus() instead.
Created attachment 270299 [details] [diff] [review] patch (In reply to comment #1) > Bah. Now that I actually _look_ we do at least check Data(). (Could have sworn > I saw code without checks before.) > You would have done, patches sorted this out as they went along ;-).
Assignee: nobody → longsonr
Status: NEW → ASSIGNED
Attachment #270299 - Flags: review?(jwatt)
Comment on attachment 270299 [details] [diff] [review] patch Thanks. I'm not sure about the assertion in nsSVGUtils.cpp. I think we should really be looking at handling failure there rather than saying it can never happen. I guess that's a separate bug though. r=jwatt
Attachment #270299 - Flags: review?(jwatt) → review+
Attachment #270299 - Flags: superreview?(tor)
Status: ASSIGNED → RESOLVED
Last Resolved: 11 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.