Summary: | Several SVG tests asserting in FontCache::getCachedFontData() | ||
---|---|---|---|
Product: | WebKit | Reporter: | Simon Fraser (smfr) <simon.fraser> |
Component: | Layout and Rendering | Assignee: | Nobody <webkit-unassigned> |
Status: | RESOLVED DUPLICATE | ||
Severity: | Normal | CC: | krit, mitz, msaboff, simon.fraser, thorton, zimmermann |
Priority: | P2 | ||
Version: | 528+ (Nightly build) | ||
Hardware: | Unspecified | ||
OS: | Unspecified |
Description
Simon Fraser (smfr)
2012-03-30 19:14:04 PDT
We need a FontCachePurgePreventer somewhere in the SVG code I guess. But why are we doing work in RenderSVGInlineText::willBeDestroyed() ? (In reply to comment #1) > We need a FontCachePurgePreventer somewhere in the SVG code I guess. > > But why are we doing work in RenderSVGInlineText::willBeDestroyed() ? When eg. a <tspan>foo</tspan> gets dynamically removed from a <text> tree, we have to rebuild all layout attributes of the remaining text. In general this shouldn't happen if the whole document tears down. That's probably buggy. (Maybe we can find an approach, which only marks the remaining tree for 'need to remeasure' instead of doing it from willBeDestroyed). Marking as duplicate of bug 81002, which covers this already. *** This bug has been marked as a duplicate of bug 81002 *** I think there's a documentBeingDestroyed that you could consult. |