mirror of
https://github.com/mozilla/gecko-dev.git
synced 2024-10-19 08:15:31 +00:00
9882821e73
The test-case in the bug does something interesting, where it causes a transition on the parent by removing a CSS rule, and that causes us to transition text-underline-offset on our ::marker, via the magic of font-size-relative properties. text-underline-offset, while it gets inherited from ::marker, is not a valid CSS property to specify on marker per spec, so we trim it here: https://searchfox.org/mozilla-central/rev/899bbd9e5a0d6de9bb9f068c48b1445c7905d9cf/servo/ports/geckolib/glue.rs#5709-5712 And that causes us to create a transition with an empty effect and everything goes downhill from here. For now, just bail out in a nicer way than we were doing. I still need to look into whether we should handle inherited transitions differently from non-inherited one in this case... I think our behavior after this patch would be correct for the test-case (because text-underline-offset would transition on the parent and ::marker would inherit it). If you specify transition only on the marker we'd refuse to transition (which I guess it is somewhat of a sensible behavior). Differential Revision: https://phabricator.services.mozilla.com/D105124 |
||
|---|---|---|
| .. | ||
| base | ||
| build | ||
| docs | ||
| forms | ||
| generic | ||
| inspector | ||
| ipc | ||
| mathml | ||
| media | ||
| painting | ||
| printing | ||
| reftests | ||
| style | ||
| svg | ||
| tables | ||
| tools | ||
| xul | ||
| moz.build | ||