doc shell for out-of-process iframe is not inactive in the case where the corresponding display item for the iframe is not built in the first place
Categories
(Core :: Layout, defect, P2)
Tracking
()
Tracking | Status | |
---|---|---|
firefox76 | --- | fixed |
People
(Reporter: hiro, Assigned: hiro)
References
(Blocks 1 open bug)
Details
Attachments
(2 files)
2.92 KB,
patch
|
Details | Diff | Splinter Review | |
47 bytes,
text/x-phabricator-request
|
Details | Review |
For example;
<div style="height: 300px; overflow: scroll;">
<div style="height: 5000px;"></div>
<iframe fission="true" ...></iframe>
</div>
above case we don't generate the display item for the iframe, so we don't call UpdateEffects() at all.
Attaching file is a mochitest to see the issue easily.
Assignee | ||
Comment 1•5 years ago
|
||
:mattwoodrow, where is the right place to call UpdateEffects with FullyHidden
EffectsInfo?
Comment 2•5 years ago
|
||
Beware of me trying to change the default visibility of docshells in bug 1572798, though there's some orange I haven't had the chance to go back to.
But It may be a better way to handle this...
Assignee | ||
Comment 3•5 years ago
|
||
Oh nice. That's good to know. Setting inactive in the first place seems to work at least for out-of-process shells.
Assignee | ||
Comment 4•5 years ago
|
||
After some more thought, the example in comment 0 is not appropriate. Even if the iframe is far from the display port boundaries, the iframe's doc shell should be active. But if the iframe is under a deck pack which is not selected in the first place, the iframe's doc shell should be inactive, I think. It's the case in bug 1541256.
Assignee | ||
Comment 5•5 years ago
|
||
Maybe I am wrong. It's not yet clear to me what the active means, if it means whether the doc shell is visible or invisible, the example in comment 0 is still applicable.
Comment 6•5 years ago
|
||
Hiro, is this bug still a problem?
If this bug as been fixed since you reported, should we land the new mochitest you posted in comment 0?
Assignee | ||
Comment 7•5 years ago
|
||
Looks like now the test works fine. Yeah, it's time to land it.
Assignee | ||
Comment 8•5 years ago
|
||
Comment 9•5 years ago
|
||
In that case, we'll track this bug for Fission Nightly (M6).
Comment 10•5 years ago
|
||
Comment 11•5 years ago
|
||
bugherder |
Description
•