The mapping of the <geometry-box> for clip-path is implemented differently than in the spec: https://drafts.fxtf.org/css-masking-1/#the-clip-path Spec changed after requests of the CSS WG to match the mapping in Fill And Stroke and CSS Transform spec. https://github.com/w3c/fxtf-drafts/commit/2c7d4824101be55bcc0b4604fb3ac86b86176227
Created attachment 379318 [details] Patch
Hm. Seems like we don’t have tests in place for this change. All bots are green. Will upload a test.
Comment on attachment 379318 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=379318&action=review > Source/WebCore/ChangeLog:14 > + Covered by existing tests. If this is covered by existing tests, then why wasn’t there a failing test that changes to succeeding when we fix the bug?
Oh, I see your comment now, sorry!
Created attachment 379375 [details] Patch
Comment on attachment 379375 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=379375&action=review LGTM. > Source/WebCore/rendering/shapes/BoxShape.cpp:86 > + // stroke, view-box compute to border-box for HTML elements. s/stroke/stroke-box/?
Comment on attachment 379375 [details] Patch This looks like it conflicts with your other patch.
Created attachment 380286 [details] Patch for landing
Similar to https://bugs.webkit.org/show_bug.cgi?id=201892 Not sure if I don't have committer rights anymore (last path last year) or if there is a mistake in the contributions.json file. Simon, could you cq+ please?
Comment on attachment 380286 [details] Patch for landing Clearing flags on attachment: 380286 Committed r250778: <https://trac.webkit.org/changeset/250778>
All reviewed patches have been landed. Closing bug.
<rdar://problem/56044543>