Fix regression in pseudo-element handling #117
Merged
+43
−36
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When improving support for
:not()
, I had made a change that accidentally prevented rewriting rules that had pseudo-elements with pseudo-states applied, e.g.::part(foo):active
or::-webkit-scrollbar-thumb:hover
. Though we want to avoid generating selectors that apply classes to these pseudo-elements (it is invalid CSS), we still want to generate the selectors involving the.pseudo-<state>-all
classes. Since we do not append those classes to the pseudo-elements (e.g..pseudo-active-all ::part(foo)
), they are valid CSS.I have updated the tests to ensure we generate the
.pseudo-<state>-all
-based selectors, but not the.pseudo-<state>
-based ones.Even after this fix, I noticed that the ShadowRootWithPart story was not consistently being rendered properly. It turned out that the stylesheet rewrite was happening before the
React.useEffect
hook was inserting the styles into the element. By switching to including the styles by importing, the timing issue is resolved.