- From: Hyowon via GitHub <noreply@w3.org>
- Date: Tue, 21 Apr 2026 02:10:43 +0000
- To: public-css-archive@w3.org
Hi @romainmenke @cdoublev, the main question was how to handle two independent <radial-extent> keywords in ellipse(), like:
[clip-path: ellipse(farthest-corner closest-corner at 175px 100px);](https://source.chromium.org/chromium/chromium/src/+/main:third_party/blink/web_tests/external/wpt/css/css-masking/clip-path/clip-path-ellipse-closest-farthest-corner.html;l=19)
Chromium and WebKit parsers accept <shape-radius>{2}, where each radius resolves independently to the euclidean corner distance - no aspect ratio preservation. The above WPT test also uses this form.
My [CL](https://chromium-review.googlesource.com/c/chromium/src/+/7754379) implemented this based on the current parser behavior to make the WPT test pass, so I hadn't thought deeply about the spec's intent. But looking at the resulting ellipse, to me the current behavior doesn't seem to have a clear geometric meaning - the ellipse doesn't pass through any corner.
If we want an ellipse that passes through the specified corner and fits the box shape naturally, I feel updating the parser and the WPT test to match the spec could be a reasonable direction.
--
GitHub Notification of comment by hyowon-igalia
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/10812#issuecomment-4285488054 using your GitHub account
--
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Tuesday, 21 April 2026 02:10:43 UTC