Bug 1520708 Comment 1 Edit History

Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.

The test is annotated as failing right now, but apparently the "failure" is simply that we show a scrollbar on the testcase, and we *don't* show a scrollbar on the reference case.

This is in fact correct -- the testcase has some tall stuff (5 spacers of height 20vh, which [combined with the text at the top of the testcase] makes its (invisible) content taller than the viewport.

Testcase:
https://hg.mozilla.org/mozilla-central/raw-file/tip/testing/web-platform/tests/css/css-writing-modes/available-size-006.html

Reference case:
https://hg.mozilla.org/mozilla-central/raw-file/tip/testing/web-platform/tests/css/css-writing-modes/reference/available-size-002-ref.html


Anyway -- it started "unexpectedly passing" as of this push:
https://treeherder.mozilla.org/#/jobs?repo=mozilla-central&revision=e3cb5a5ef667973b60314b67e13184f5d9cedb18&searchStr=android%2C7.0%2Cx86%2C%2Cwr6&selectedJob=222383394

...which includes bug 1504659, whose commits have some notes about sometimes making scrollbars stop showing up on Android. So, given that the scrollbar was the only thing causing "failure" here, it makes sense that bug 1504659 fixed the "failure".

For now, we should perhaps update the annotation in the test's .ini file to indicate that it's random on Android or something (since it looks like it's only passing on this one Android platform but not on others (?)...)  And, long-term, we should fix the test so that it's not throwing content off the bottom of its viewport and triggering a scrollbar, so that it actually correctly passes.
The test is annotated as failing right now, but apparently the "failure" is simply that we show a scrollbar on the testcase, and we *don't* show a scrollbar on the reference case.

This is in fact correct behavior -- the testcase has some tall stuff (5 spacers of height 20vh, which [combined with the text at the top of the testcase] makes its (invisible) content taller than the viewport.  So, the testcase needs a scrollbar, and the reference case (which doesn't have these spacers) does not need a scrollbar. So we are correct in rendering them differently.

Testcase:
https://hg.mozilla.org/mozilla-central/raw-file/tip/testing/web-platform/tests/css/css-writing-modes/available-size-006.html

Reference case:
https://hg.mozilla.org/mozilla-central/raw-file/tip/testing/web-platform/tests/css/css-writing-modes/reference/available-size-002-ref.html


Anyway -- it started "unexpectedly passing" as of this push:
https://treeherder.mozilla.org/#/jobs?repo=mozilla-central&revision=e3cb5a5ef667973b60314b67e13184f5d9cedb18&searchStr=android%2C7.0%2Cx86%2C%2Cwr6&selectedJob=222383394

...which includes bug 1504659, whose commits have some notes about sometimes making scrollbars stop showing up on Android. So, given that the scrollbar was the only thing causing "failure" here, it makes sense that bug 1504659 fixed the "failure".

For now, we should perhaps update the annotation in the test's .ini file to indicate that it's random on Android or something (since it looks like it's only passing on this one Android platform but not on others (?)...)  And, long-term, we should fix the test so that it's not throwing content off the bottom of its viewport and triggering a scrollbar, so that it actually correctly passes.

Back to Bug 1520708 Comment 1