scrollbars lost since Bug 1644271 and Bug 1645520 didn't resolve it
Categories
(Core :: Panning and Zooming, defect, P2)
Tracking
()
| Tracking | Status | |
|---|---|---|
| firefox77 | --- | unaffected |
| firefox78 | --- | unaffected |
| firefox79 | --- | verified |
| firefox80 | --- | verified |
People
(Reporter: euthanasia_waltz, Assigned: kats)
References
(Regressed 1 open bug, Regression)
Details
(Keywords: regression)
Attachments
(2 files)
On
LinuxMint(19.3 mate),
XGA(1024x768) display,
unmaximized window or showing sidebar,
some sites(e.g. https://www.yahoo.co.jp/ http://forums.mozillazine.org/ https://www.itmedia.co.jp/ https://distrowatch.com/)
scrollbars are not displayed.
(apz.allow_zooming=false (default), apz.mvm.force-enabled=true (default))
Scrollbars are displayed when apz.mvm.force-enabled=false.
Updated•6 years ago
|
Updated•6 years ago
|
| Assignee | ||
Comment 1•6 years ago
|
||
Hm, weird. looks like the scrollbars are positioned "outside the root content" rather than "inside the visual viewport". So if the content is wider than the window the vertical scrollbar is clipped by the window. Same for horizontal scrollbar.
I can investigate. Possibly one of my other patches for bug 1644271 will end up fixing this.
| Assignee | ||
Comment 2•6 years ago
|
||
Doesn't look like my patch queue for bug 1644271 fixes this, I'll need to investigate.
| Assignee | ||
Comment 3•6 years ago
|
||
Replacing this GetIsViewportOveridden call with false seems to fix the problem. GetIsViewportOverridden just checks for the existence of an MVM, so presumably whatever other code this is meant to be in sync with is checking more than just the existence of an MVM. Probably for the allow_zooming flag or mobile viewport support. I can look more tomorrow, no time left today.
| Assignee | ||
Comment 4•6 years ago
|
||
I think that code was only needed with containerful scrolling which is no longer a thing. In particular it seems related to the line that was removed here, so we should be able to set it to false unconditionally now. Maybe.
| Assignee | ||
Comment 5•6 years ago
|
||
| Assignee | ||
Comment 6•6 years ago
|
||
| Assignee | ||
Comment 7•6 years ago
|
||
Depends on D80054
| Assignee | ||
Updated•6 years ago
|
Comment 9•6 years ago
|
||
Backed outfor gv-junit failures on VerticalClippingTest.
Backout link: https://hg.mozilla.org/integration/autoland/rev/4d6e89cfbb2d8d9f9fbfc599e39e5e06401cb254
Failure log: https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=306762989&repo=autoland&lineNumber=13019
| Assignee | ||
Updated•6 years ago
|
| Assignee | ||
Comment 10•6 years ago
|
||
I guess for some reason this breaks the clipping required by the setVerticalClipping API and/or dynamic toolbar. The dynamic toolbar sticky-pos tests are failing too, in addition to the gv-junit tests, on Android.
Obviously I can fix this by restoring the old behaviour on Android, by using UsesMobileViewportSizing instead of GetIsViewportOverridden, and I'll update the patch to do that. I'm not sure though if that condition is conceptually correct - maybe it needs to detect dynamic-toolbar-ness instead.
Updated•6 years ago
|
Comment 11•6 years ago
|
||
kats, there are also these reftest failures:
https://treeherder.mozilla.org/logviewer.html#?job_id=306769813&repo=autoland
https://treeherder.mozilla.org/logviewer.html#?job_id=306777611&repo=autoland
| Assignee | ||
Comment 12•6 years ago
|
||
Thanks. The updated patches should take care of those as well.
Comment 13•6 years ago
|
||
Comment 14•6 years ago
|
||
| bugherder | ||
https://hg.mozilla.org/mozilla-central/rev/f444d609e7c3
https://hg.mozilla.org/mozilla-central/rev/f0d2b2411631
Updated•6 years ago
|
Comment 15•6 years ago
|
||
I was able to reproduce the issue on Firefox Nightly (2020-06-15) under Ubuntu 20.04 using the infos on Comment 0.
The issue is no longer reproducible on Firefox 79.0b4 and Firefox Nightly 80.0a1 (2020-07-06) under Ubuntu 20.04.
Updated•5 years ago
|
Description
•