1366422 - stylo: test_transitions_and_reframes.html has some failures with ::before/::after and transitions

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, defect, P2)

Description

[Cameron McCormack (:heycam)]

These failures I marked as expected fails in bug 1364845, but need to be investigated.

TEST-UNEXPECTED-FAIL | layout/style/test/test_transitions_and_reframes.html | margin of child of element b2 (classes: t m o) ::before pseudo-element - got "100px", expected "25px"
TEST-UNEXPECTED-FAIL | layout/style/test/test_transitions_and_reframes.html | margin of child of element a2 (classes: t m o) ::after pseudo-element - got "100px", expected "25px"
TEST-UNEXPECTED-FAIL | layout/style/test/test_transitions_and_reframes.html | margin of child of element b2 (classes: m o t) ::before pseudo-element - got "100px", expected "25px"
TEST-UNEXPECTED-FAIL | layout/style/test/test_transitions_and_reframes.html | margin of child of element a2 (classes: m o t) ::after pseudo-element - got "100px", expected "25px"

Comment 1

[Hiroyuki Ikezoe (:hiro)]

Attachment: A simplified test case

Initially I thought the transition in this test case does not start on Gecko either. I was trying to make the transition work on Gecko for a half of day because I thought I make some mistakes in the test case.  Finally I noticed *the Gecko* is not Gecko, it's a stylo build. In these days I use a stylo build for daily use!

Comment 2

[Hiroyuki Ikezoe (:hiro)]

As per a try in bug 1368240 comment 11, this will be fixed by bug 1368240.
I believed Emilio fix this bug too. :)

Comment 3

[Emilio Cobos Álvarez (:emilio)]

Right, so from that test case it means that the descendant restyle hints weren't accounting for the pseudo-element combinator depth, so we never restyled them proper. This is fixed by bug 1368240 then.

Comment 4

[Emilio Cobos Álvarez (:emilio)]

Err, sorry, bugzilla apparently leaves the form state across refreshes.

Comment 5

[Hiroyuki Ikezoe (:hiro)]

Now bug 1368240 is closed!  Thank you Emilio!