1907971 - Inconsistent CSS selector matching with svg:use

Status: RESOLVED INVALID

(Core :: CSS Parsing and Computation, defect)

Description

[Jeff Muizelaar [:jrmuizel]]

Attachment: testcase.html

This is based off the reduced test case on bug 1896589.

Chrome and Safari paint both "61"s red and we probably should too.

Comment 1

[Robert Longson [:longsonr]]

No, we all agreed with the w3c not to do this. The SVG specification is clear on this: https://www.w3.org/TR/SVG2/struct.html#UseStyleInheritance

https://issues.chromium.org/issues/40550039 is the Chrome bug and https://bugs.webkit.org/show_bug.cgi?id=249080 is the webkit bug.

Comment 2

[Jeff Muizelaar [:jrmuizel]]

The spec change was made in 2017 and the other browser don't seem eager to adopt the new behaviour. Should we consider changing the spec back and adopting Chrome and Safari's behaviour?

Comment 3

[Robert Longson [:longsonr]]

I've added comments to the webkit and Chrome bugs maybe that will get some reaction.

Comment 4

[Jeff Muizelaar [:jrmuizel]]

Do we have a list of the other bugs that we've marked as invalid?

Comment 5

[Emilio Cobos Álvarez [:emilio]]

Yeah this is not inconsistent at all. In fact that's exactly how shadow dom works outside of svg.

Comment 6

[Daniel Holbert [:dholbert]]

(In reply to Jeff Muizelaar [:jrmuizel] from comment #4)

Do we have a list of the other bugs that we've marked as invalid?

One of these at least was bug 1415186 (which was filed for a codepen demo; it's not clear if any live sites were affected there).

Not sure if there are others.

Comment 7

[Robert Longson [:longsonr]]

I think this is most of them:

bug 1805048 bug 1760954 bug 1415186 bug 1409720 bug 1424466 bug 1421336 bug 1421336 bug 1395221 bug 1406436 bug 1377435

Note that these are over at least 7 years.