Closed
Bug 589393
Opened 15 years ago
Closed 15 years ago
Thunderbird MozMill regressions following latest tracemonkey merge
Categories
(Thunderbird :: OS Integration, defect)
Thunderbird
OS Integration
Tracking
(blocking2.0 -)
RESOLVED
FIXED
Thunderbird 3.3a1
| Tracking | Status | |
|---|---|---|
| blocking2.0 | --- | - |
People
(Reporter: standard8, Assigned: standard8)
Details
(Keywords: regression)
We're seeing various MozMill issues as a result of the latest tracemonkey merge.
The most obvious errors are:
TEST-UNEXPECTED-FAIL | e:\buildbot\win32-comm-central-check\build\mail\test\mozmill\folder-tree-modes\test-custom-smart-folder.js | setupModule
EXCEPTION: no message!
at: nonesuch line 0
TEST-UNEXPECTED-FAIL | e:\buildbot\win32-comm-central-check\build\mail\test\mozmill\folder-tree-modes\test-smart-folders.js | setupModule
EXCEPTION: no message!
at: nonesuch line 0
TEST-UNEXPECTED-FAIL | e:\buildbot\win32-comm-central-check\build\mail\test\mozmill\folder-tree-modes\test-unread-folders.js | setupModule
EXCEPTION: no message!
at: nonesuch line 0
Which if I recall correctly means we either haven't been able to create a message in a folder, or that message is not selectable.
There's also consistent unexpected disconnects:
Disconnect Error: Application unexpectedly closed
TEST-UNEXPECTED-FAIL | (runtestlist.py) | Exited with code 1 during directory run
These are normally crashes, but not showing up as crashes, so I'm not sure if there's something else in the mozmill/jsbridge/Thunderbird set-up causing these.
Example log:
http://tinderbox.mozilla.org/showlog.cgi?tree=Thunderbird&errorparser=unittest&logfile=1282341148.1282346895.11034.gz&buildtime=1282341148&buildname=WINNT%205.2%20comm-central%20check&fulltext=1
On Mac the situation looks different, but I'm not convinced at what is going on.
http://tinderbox.mozilla.org/showlog.cgi?log=Thunderbird/1282341616.1282347068.11758.gz
I'm going to clobber all these builders just in case, however I think the Windows ones are definitely showing errors as the first build after tracemonkey was a clobber.
| Assignee | ||
Comment 1•15 years ago
|
||
Request blocking on the basis that I think a tracemonkey merge causing regressions at least needs investigation as to what the issue is.
Note, I'm afk until Monday but I've cc'ed other folks who may be able to help.
blocking2.0: --- → ?
Keywords: regression
Comment 2•15 years ago
|
||
We should really get a thunderbird column on the tracemonkey tree to avoid breaking you guys over and over. Want to file a bug for that?
| Assignee | ||
Comment 3•15 years ago
|
||
(In reply to comment #2)
> We should really get a thunderbird column on the tracemonkey tree to avoid
> breaking you guys over and over. Want to file a bug for that?
Yes, I can file that on Monday.
| Assignee | ||
Comment 4•15 years ago
|
||
I have a possible fix for this:
- set firstRunDone()
+ set firstRunDone(aAlwaysTrue)
in mail/components/search/content/searchCommon.js
The complaint was that a setter must have at least one argument.
I'm not sure if it will fix all the mozmill issues, but I've pushed try to see if what happens.
Comment 5•15 years ago
|
||
rs+ on that if you need it.
Updated•15 years ago
|
blocking2.0: ? → -
| Assignee | ||
Comment 6•15 years ago
|
||
Sorry for not following up earlier - this was fixed by the change in comment 4, checked in to comm-central here:
http://hg.mozilla.org/comm-central/rev/d80a5cbdaa41
Assignee: general → bugzilla
Status: NEW → RESOLVED
Closed: 15 years ago
Component: JavaScript Engine → OS Integration
Product: Core → Thunderbird
QA Contact: general → os-integration
Resolution: --- → FIXED
Target Milestone: --- → Thunderbird 3.2a1
You need to log in
before you can comment on or make changes to this bug.
Description
•