Branch: refs/heads/main
Home: https://github.com/WebKit/WebKit
Commit: e25ad075b26cefa4496c0dabbf69e74e62b1edd4
https://github.com/WebKit/WebKit/commit/e25ad075b26cefa4496c0dabbf69e74e62b1edd4
Author: Ryosuke Niwa <[email protected]>
Date: 2026-09-22 (Tue, 22 Sep 2026)
Changed paths:
M
LayoutTests/http/tests/site-isolation/touch-events/content-observation-expected.txt
M
LayoutTests/http/tests/site-isolation/touch-events/content-observation.html
M
LayoutTests/http/tests/site-isolation/touch-events/touch-click-coordinates-expected.txt
M
LayoutTests/http/tests/site-isolation/touch-events/touch-click-coordinates.html
M
LayoutTests/http/tests/site-isolation/touch-events/touch-event-coordinates-expected.txt
M
LayoutTests/http/tests/site-isolation/touch-events/touch-event-coordinates.html
M LayoutTests/platform/ios/TestExpectations
Log Message:
-----------
[iOS] Enable the site isolation touch-events layout tests and rebaseline them
https://bugs.webkit.org/show_bug.cgi?id=324908
Reviewed by Sihui Liu.
LayoutTests/TestExpectations skips http/tests/site-isolation/touch-events for
every platform, on
the same footing as http/tests/site-isolation/ios on the next line, leaving it
to
platform/ios/TestExpectations to opt tests back in. Only two of the six were
ever named there, and
iOS is the only platform that can run them, so the other four have never run
anywhere. Each of the
four was added by an iOS site isolation fix -- 292460, 293712 and 320725 --
whose commit did not
touch TestExpectations.
Running them turns up two unrelated causes. touch-click-coordinates.html,
content-observation.html
and the touchstart/touchmove assertions of touch-event-coordinates.html are off
by 4px in Y and
exact in X, because their iframe sits below the paragraph js-test.js inserts
for description() and
that paragraph is now 4px shorter than when they were baselined. The two
drag-nested tests position
their iframe absolutely and their touchstart is exact, which is what
distinguishes a stale baseline
from a coordinate defect. Rebaseline the first three.
The remaining failure is real: the lift-up of a simulated drag is delivered
21px above its
endpoint, with X exact, in all three tests that drag. That is rdar://156634858,
already bisected to
294532@main, so keep expecting touchend at the correct location and mark the
two drag-nested tests
as failures against it rather than baking 150,79 into them. Their touchstart
and touchmove
assertions still pass, and enabling them puts the defect in front of CI, which
could not see it
from any angle before.
touch-event-coordinates.html keeps its existing Pass Failure Timeout
expectation: its failure is
now just the one touchend line, but the timeout the bug also mentions has not
been observed here.
*
LayoutTests/http/tests/site-isolation/touch-events/content-observation-expected.txt:
* LayoutTests/http/tests/site-isolation/touch-events/content-observation.html:
*
LayoutTests/http/tests/site-isolation/touch-events/touch-click-coordinates-expected.txt:
*
LayoutTests/http/tests/site-isolation/touch-events/touch-click-coordinates.html:
*
LayoutTests/http/tests/site-isolation/touch-events/touch-event-coordinates-expected.txt:
*
LayoutTests/http/tests/site-isolation/touch-events/touch-event-coordinates.html:
* LayoutTests/platform/ios/TestExpectations:
Canonical link: https://commits.webkit.org/321661@main
To unsubscribe from these emails, change your notification settings at
https://github.com/WebKit/WebKit/settings/notifications
