Google Docs fonts menu has a huge scroll-position-change (jumping upwards in the font list), if you scroll down and then move cursor around to choose a font
Categories
(Core :: Layout: Scrolling and Overflow, defect)
Tracking
()
People
(Reporter: abenson, Unassigned)
Details
Attachments
(1 file)
1.78 MB,
video/mp4
|
Details |
Observed on macOS
STR:
- Open a google doc (new or existing)
- Type some text in the doc, or select existing text
- Click the fonts menu and then scroll to the bottom of the list
- Move mouse cursor to make a font selection
At this point, the font list appears to "jump" back to the top, making it impossible for me to select a font lower down on the list. It is also difficult to navigate the flyout menus in the font list because as the mouse cursor traverses over the scrollbar area, the flyout disappears.
Comment 1•9 months ago
|
||
Video from Aaron Benson that shows the issue on macOS. I have been able t reproduce this on Windows.
Comment 2•9 months ago
|
||
Moving to Layout for first triage
Comment 3•9 months ago
|
||
It is also difficult to navigate the flyout menus in the font list because as the mouse cursor traverses over the scrollbar area, the flyout disappears.
This part is bug 1836872.
I don't think the "Jump back to the top" part is covered by that bug, though.
Comment 4•9 months ago
|
||
I can reproduce the jump - it seems to happen if I move my cursor to/from a font name that triggers a flyout menu.
Comment 5•9 months ago
|
||
Have been able to reproduce in FF x124 and Win 11
Comment 6•9 months ago
|
||
Here's a performance profile where the "jump" happens: https://share.firefox.dev/3OUHh8w
Comment 7•9 months ago
|
||
I haven't been able to reproduce in Chrome or Safari, so this seems to be Firefox-specific. (Minor caveat: I can reproduce a very-small single-menu-item-height "jump" if I hover the very top of the font list, after having scrolled to the end; because, hovering that top list-entry causes the list to scroll to show that entry. That seems to be by-design and much less severe than what's being discussed here.)
In Firefox Nightlies: I can reproduce the large "jump" at least as far back as Nightly 2022-01-01 (v97.0a1), so this is not a recent regression on our side. If this issue only just started happening, then it wasn't due to a change on our side. (Still might be arguably our bug, but a longstanding bug that [maybe] Google Docs just started tripping over in a recent deployment.)
I can reproduce the jump even if I spoof a Chrome UA string, too, so there's no UA-sniffing involved in triggering the bad condition.
Updated•9 months ago
|
Comment 8•9 months ago
|
||
I just did a bit of debugging. I thought this might have been similar or made worse by our recent changes to make Element.focus()
center an element. After further investigation, this does not appear to be the case. We do not seem to trigger the scroll do to the logic in ComputeNeedToScroll
. The ScrollOperationParams.mTriggeredByScript flag is set to ScrollTriggeredByScript::Yes
in recent debugging sessions, which AFAIK indicates there is an explicit scroll operation that is triggering it. This doesn't necessarily mean there is no bug in our code, but I think this indicates this isn't a bug due to our recent changes to Element.focus()
.
I'll continue to attempt to debug this a bit more.
Updated•9 months ago
|
Reporter | ||
Updated•9 months ago
|
Comment 9•8 months ago
|
||
I think this is a duplicate of Bug 1589505
Comment 10•8 months ago
|
||
The severity field is not set for this bug.
:hiro, could you have a look please?
For more information, please visit BugBot documentation.
Comment 11•8 months ago
|
||
I would say this is a duplicate of bug 1836872.
Comment 12•8 months ago
|
||
Wrong bug number. :/
Updated•8 months ago
|
Description
•