HTML parser: fix a regression with foster parenting in foreign content

[zcorpan]

The regression was introduced in f690ad9
(PR #6399)

For this case:

<table><math><p>foo

the <p> token would not have foster parenting enabled, thus inserting it
into the table.

Fixes #6439.

• At least two implementers are interested (and none opposed):
  • …
  • …
• Tests are written and can be reviewed and commented upon at:
  • Already covered by existing tests:
    • https://github.com/html5lib/html5lib-tests/blob/082a4be911f6bea27215038beb36ef7de822305b/tree-construction/tests9.dat#L270
    • https://github.com/html5lib/html5lib-tests/blob/082a4be911f6bea27215038beb36ef7de822305b/tree-construction/tests10.dat#L236
• Implementation bugs are filed:
  • Chrome: N/A
  • Firefox: https://bugzilla.mozilla.org/show_bug.cgi?id=1205631
  • Safari: N/A

(See WHATWG Working Mode: Changes for more details.)

---

/parsing.html ( diff )

[domenic]

@zcorpan am I correct that this is already implemented in Chrome and Safari? In that case we have implementer interest and should merge this.

[zcorpan]

Yes. It's implemented in Gecko as well, in that it will also foster parent the <p>.