Summary: | More adjustments to css width calculations of cue boxes | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Roger Fong <roger_fong> | ||||
Component: | Media | Assignee: | Nobody <webkit-unassigned> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Normal | CC: | bfulgham, calvaris, commit-queue, eric.carlson, esprehn+autocc, gyuyoung.kim, philipj, roger_fong | ||||
Priority: | P2 | ||||||
Version: | 528+ (Nightly build) | ||||||
Hardware: | Unspecified | ||||||
OS: | Unspecified | ||||||
Attachments: |
|
Description
Roger Fong
2014-08-11 16:29:51 PDT
Created attachment 236413 [details]
patch
Comment on attachment 236413 [details] patch View in context: https://bugs.webkit.org/attachment.cgi?id=236413&action=review r=me, but please address the possible zero-base-font-size issue I mentioned. > Source/WebCore/html/track/TextTrackCueGeneric.cpp:82 > + double authorFontSize = videoSize.height() * cue->baseFontSizeRelativeToVideoHeight() / 100.0; Can baseFontSizeRelativeToVideoHeight ever be zero (I think it can be)? If the author could arbitrarily set this to zero, I think we should check and fall back to that minimum (10 pt) value from the CSS. committed with review commented applied: http://trac.webkit.org/changeset/172421 |