RESOLVED FIXED 248927
length values in gradients should take effective zoom into account 
https://bugs.webkit.org/show_bug.cgi?id=248927
Summary length values in gradients should take effective zoom into account
Fujii Hironori
Reported 2022-12-07 20:12:39 PST
Created attachment 463932 [details] linear-gradient-and-zoom.html absolute length values in gradients should take effective zoom into account
Attachments
linear-gradient-and-zoom.html (379 bytes, text/html)
2022-12-07 20:12 PST, Fujii Hironori 		no flags
Details
radial-gradient-and-zoom.html (420 bytes, text/html)
2022-12-07 20:12 PST, Fujii Hironori 		no flags
Details
WIP patch (6.48 KB, patch)
2022-12-07 22:30 PST, Fujii Hironori 		no flags
DetailsFormatted DiffDiff
Patch (9.25 KB, patch)
2022-12-08 13:26 PST, Fujii Hironori 		no flags
DetailsFormatted DiffDiff
Patch (4.03 KB, patch)
2022-12-09 12:03 PST, Fujii Hironori 		no flags
DetailsFormatted DiffDiff
Show Obsolete (2) View All Add attachment proposed patch, testcase, etc.
Fujii Hironori
Comment 1 2022-12-07 20:12:54 PST
Created attachment 463933 [details] radial-gradient-and-zoom.html
Fujii Hironori
Comment 2 2022-12-07 22:30:01 PST
Created attachment 463935 [details] WIP patch
Fujii Hironori
Comment 3 2022-12-08 13:26:09 PST
Created attachment 463948 [details] Patch
Simon Fraser (smfr)
Comment 4 2022-12-09 08:25:18 PST
Comment on attachment 463948 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=463948&action=review > Source/WebCore/rendering/style/StyleGradientImage.cpp:671 > +static float positionFromValue(const CSSPrimitiveValue& initialValue, const CSSToLengthConversionData& conversionData, const FloatSize& size, bool isHorizontal, const RenderStyle& style) Instead of passing the style in more places, should we put the effectiveZoom value into CSSToLengthConversionData? In fact, CSSToLengthConversionData seems to have the style already.
Fujii Hironori
Comment 5 2022-12-09 12:01:14 PST
Comment on attachment 463948 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=463948&action=review >> Source/WebCore/rendering/style/StyleGradientImage.cpp:671 >> +static float positionFromValue(const CSSPrimitiveValue& initialValue, const CSSToLengthConversionData& conversionData, const FloatSize& size, bool isHorizontal, const RenderStyle& style) > > Instead of passing the style in more places, should we put the effectiveZoom value into CSSToLengthConversionData? In fact, CSSToLengthConversionData seems to have the style already. Will fix. Thank you.
Fujii Hironori
Comment 6 2022-12-09 12:03:01 PST
Created attachment 463969 [details] Patch
Simon Fraser (smfr)
Comment 7 2022-12-09 12:18:49 PST
Comment on attachment 463969 [details] Patch So much cleaner!
EWS
Comment 8 2022-12-09 18:23:02 PST
Committed 257659@main (18a30951bbec): <https://commits.webkit.org/257659@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 463969 [details].
Radar WebKit Bug Importer
Comment 9 2022-12-09 18:24:16 PST
<rdar://problem/103202297>
Fujii Hironori
Comment 10 2022-12-11 13:31:34 PST
*** Bug 248906 has been marked as a duplicate of this bug. ***
Note You need to log in before you can comment on or make changes to this bug.