Closed Bug 589618 Opened 10 years ago Closed 10 years ago

[SeaMonkey 2.1, mochitest-browser-chrome] browser_522545.js | sessionstore got correct userTypedValue - Got example.org, expected mozilla.org

Categories

(SeaMonkey :: Session Restore, defect)

defect
Not set

Tracking

(Not tracked)

RESOLVED FIXED
seamonkey2.1b1

People

(Reporter: kairo, Assigned: neil)

References

(Blocks 1 open bug)

Details

(Keywords: intermittent-failure, Whiteboard: [sm-perma][cc-orange])

Attachments

(1 file, 1 obsolete file)

http://tinderbox.mozilla.org/showlog.cgi?log=SeaMonkey/1282499502.1282502435.6363.gz

TEST-UNEXPECTED-FAIL | chrome://mochikit/content/browser/suite/common/tests/browser/browser_522545.js | sessionstore got correct userTypedValue - Got example.org, expected mozilla.org

Something's wrong in our own court in this test. :(
I'm pretty sure I saw this before my userTypedValue patch landed.
The bug was opened before Your patch landed ;)
I don't know why Firefox didn't hit this, but problem is that _collectTabData returns cached data of tab yet to be loaded. Suppose sessionstore opened a tab and cached it's data until load event indicates it's loaded. So if you call getbrowserstate ss actually returns a cached state of a tab. But if you do setbrowserstate before tab loaded and do getbrowserstate before new one loaded, ss will return cache of previous state. Maybe it's somehow related with SM specific "load" and "pageshow" event processing, or empty browser state which this test trying to set triggering this. Anyhow, this patch explicitly clears the cache of tab being reused.
Assignee: nobody → misak.bugzilla
Status: NEW → ASSIGNED
Attachment #470280 - Flags: superreview?(neil)
Attachment #470280 - Flags: review?(neil)
I don't think we're allowing the load event to fire. This should fix that.
Assignee: misak.bugzilla → neil
Attachment #470280 - Attachment is obsolete: true
Attachment #470309 - Flags: review?(misak.bugzilla)
Attachment #470280 - Flags: superreview?(neil)
Attachment #470280 - Flags: review?(neil)
Comment on attachment 470309 [details] [diff] [review]
Actually allow the load event to fire

Oh, how simple ...
This rises another question - why firefox pass this test ?
Attachment #470309 - Flags: review?(misak.bugzilla) → review+
Pushed changeset dca910b78975 to comm-central.
Status: ASSIGNED → RESOLVED
Closed: 10 years ago
Resolution: --- → FIXED
Flags: in-testsuite+
Target Milestone: --- → seamonkey2.1b1
Whiteboard: [sm-perma][orange] → [sm-perma]
Whiteboard: [sm-perma] → [sm-perma][cc-orange]
You need to log in before you can comment on or make changes to this bug.