New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 700346 link

Starred by 2 users

Issue metadata

Status: Archived
Owner:
Closed: Sep 19
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows
Pri: 2
Type: Bug-Regression



Sign in to add a comment

[OOPIFs] tooltip and cursor pointer don't always update when moving over iframe

Reported by cool...@gmail.com, Mar 10 2017

Issue description

UserAgent: Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/57.0.2987.98 Safari/537.36

Steps to reproduce the problem:
1. install the unpacked extension from the attached twitch5.zip
2. navigate to chrome-extension://TWITCH5-EXTENSION-ID/player.html?channel=coolcmd
3. hit INSERT key on the keyboard to open the chat
4. move the mouse fast enough from left to right

What is the expected behavior?
while above the chat (third-party iframe), mouse cursor must have default shape (arrow), and tooltip must be hidden.

you can see expected behavior if run command
chrome --force-fieldtrials=SiteIsolationExtensions/Control
which disables OOPIFs (in my version of Chrome)

What went wrong?
while above the chat (third-party iframe), mouse cursor have incorrect shape (hand) and tooltip. see the attached iframe.mp4.

Did this work before? Yes 55

Does this work in other browsers? Yes Firefox

Chrome version: 57.0.2987.98  Channel: stable
OS Version: 6.1 (Windows 7, Windows Server 2008 R2)
Flash Version: 

part of the <div id="размерчата"> is located ABOVE the <iframe>. specific CSS rule:
#размерчата {margin-right: -8px}
this is cause the issue.

 
Cc: ranjitkan@chromium.org
Labels: Needs-Feedback
@ coolcmd: Looks like the  twitch5.zip was missed in the attachment, request you to please attach it again.

Thanks.!

Comment 2 by gov...@chromium.org, Mar 10 2017

Cc: pbomm...@chromium.org
Also was this working on Chrome version 56?

Comment 3 by cool...@gmail.com, Mar 10 2017

here the attachments (your POST form is *evil*)

>Also was this working on Chrome version 56?
56 and 57 working identical in my case
iframe.mp4
591 KB View Download
twitch5.zip
120 KB Download
Project Member

Comment 4 by sheriffbot@chromium.org, Mar 10 2017

Labels: -Needs-Feedback
Thank you for providing more feedback. Adding requester "ranjitkan@chromium.org" to the cc list and removing "Needs-Feedback" label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Comment 5 by creis@chromium.org, Mar 10 2017

Cc: ekaramad@chromium.org kenrb@chromium.org aval...@chromium.org paulmeyer@chromium.org nasko@chromium.org creis@chromium.org lfg@chromium.org
Components: UI>Input Internals>Sandbox>SiteIsolation
Owner: wjmaclean@chromium.org
Thanks for the report!  wjmaclean@, can you help triage and find an owner for this?  (I know paulmeyer@ worked on tooltips in  issue 609932 .)
Cc: wjmaclean@chromium.org
Owner: paulmeyer@chromium.org
Paul, could you triage this? You have (I think) the most experience with OOPIF tool tips.

Comment 7 by nasko@google.com, Mar 10 2017

Given the video, I don't think it is tooltip specific bug. Whenever the tooltip is displayed over the chat window, the mouse cursor is also not the expected one. The bug seems to occur when the cursor is moved faster and doesn't occur when moved slower and both cursor type and tooltip are both always in the same state, so I would suspect just a generic bug related to mouse leave/enter state.
Owner: wjmaclean@chromium.org
Actually, I'll take this back for now in case it's mouse-leave/enter related.

Comment 9 by creis@chromium.org, Mar 10 2017

Summary: [OOPIFs] tooltip and cursor pointer don't always update when moving over iframe (was: [OOPIFs] tooltip and cursor pointer are freeze above iframe)
Comment 8: Thanks!  I'm updating the summary accordingly.

Comment 10 by kenrb@chromium.org, Mar 10 2017

The cursor problem is likely to be  bug 614540 . Race condition between two cursor updates.

I don't know about the tooltip text, but it could be similar.
In the case of mouse enter/leave, it would seem like there's a semantic ordering that could be introduced ... for example, if we receive the leave after the enter, and the renderer ids (for example) don't match between the two we would not use the leave for setting the current cursor? Or perhaps we would queue the 'enter' until the 'leave' has been received and processed?

I'm not quite sure (yet) how the drag-handle regions are defined, but something similar might work there too?

Comment 12 by cool...@gmail.com, Mar 14 2017

BTW, workaround for my case:
#размерчата:hover + #чат {pointer-events: none}

Comment 13 by creis@chromium.org, Apr 28 2017

Status: Assigned (was: Unconfirmed)
Assigning to wjmaclean@, per comment 8.  Maybe we can take a look at it for M60?

Comment 14 by creis@chromium.org, Jul 26 2017

Sounds like the cursor update part was resolved in kenrb@'s r486528 for  issue 614540 .

wjmaclean@: Should this bug stay open for the tooltip issue?
Yes, keeping it open until the tooltips issue is resolved seems reasonable.
Status: Archived (was: Assigned)

Sign in to add a comment