Can't rely on "browsing-context-discarded" for uninstalling SHistoryListeners
Categories
(Firefox :: Session Restore, defect, P2)
Tracking
()
Fission Milestone | M7a |
Tracking | Status | |
---|---|---|
firefox-esr78 | --- | unaffected |
firefox88 | --- | unaffected |
firefox89 | --- | wontfix |
firefox90 | --- | wontfix |
firefox91 | --- | fix-optional |
People
(Reporter: u608768, Assigned: u608768)
References
Details
In bug 1700963, I made it so we listen for "browsing-context-discarded" messages to detect that we can uninstall the context's associated SHistoryListener.
But it appears that the embedder element is always cleared before we fire "browsing-context-discarded", so we're not actually uninstalling listeners using this codepath. We might need to add a "browsing-context-will-clear-embedder" or something to know that the embedder element is being cleared, and use that instead.
(In reply to :kashav from comment #0)
We might need to add a "browsing-context-will-clear-embedder" or something to know that the embedder element is being cleared, and use that instead.
This doesn't work because we still end up clearing the embedder before SessionStore receives the notification.
Updated•4 years ago
|
Comment 2•4 years ago
|
||
Set release status flags based on info from the regressing bug 1700963
Updated•4 years ago
|
Comment 3•4 years ago
|
||
That's a P2/S3 and we are in our last week of betas, so I am marking this bug as wontfix for 89.
Comment 4•3 years ago
|
||
Is there a user-facing impact here, that we need to track for 90?
No, there really shouldn't be any impact at all, since we're still removing listeners. I think calling it a regression was a little misleading.
Updated•3 years ago
|
Bug 1706445 makes it so we unregister these listeners more reliably, so this shouldn't be a problem anymore. If this does become a problem, we might be to just unregister the listener after destroying the context's session history object.
Description
•