436606 - "ASSERTION: RemoveCurrentDraftMessage can't get draft id" involving message compose window

Status: RESOLVED FIXED

(MailNews Core :: Composition, defect)

Description

[Gary Kwong [:gkw] [:nth10sd] (NOT official MoCo now)]

Attachment: both normal and full backtraces

###!!! ASSERTION: RemoveCurrentDraftMessage can't get draft id: '(NS_SUCCEEDED(rv) && (!curDraftIdURL.IsEmpty()))', file /Users/skywalker/Desktop/Mozilla/cvs/mozilla/mailnews/compose/src/nsMsgCompose.cpp, line 393

The assertion fires when I open a message compose window (or reply to a message), not typing anything, and shifting focus to do something else. After awhile, the assertion will be caught in Terminal.

Comment 1

[Wayne Mery (:wsmwk)]

Gary, could this be related to multiple drafts being saved?

Comment 2

[Gary Kwong [:gkw] [:nth10sd] (NOT official MoCo now)]

Possibly - though I can't say for sure. Maybe bienvenu knows what this assertion might mean?

Comment 3

[Wayne Mery (:wsmwk)]

bienvenu,
perhaps comment 0 sheds light on these other bugs?

https://bugzilla.mozilla.org/buglist.cgi?type1-0-0=anywordssubstr&short_desc=local&field0-0-0=short_desc&bug_severity=critical&bug_severity=major&bug_severity=normal&bug_severity=minor&type1-0-1=substring&resolution=---&query_format=advanced&value1-0-0=folder&short_desc_type=allwordssubstr&type0-0-0=anywordssubstr&value0-0-0=draft%20&field1-0-0=short_desc&product=MailNews%20Core&product=Thunderbird&field1-0-1=component

Comment 4

[:Irving Reid (No longer working on Firefox)]

I'm also seeing this pretty reliably in Beta (v10) after leaving a compose window up for a few minutes.

Comment 5

[Kent James (:rkent)]

The code where this happens is:

336   rv = m_compFields->GetDraftId(getter_Copies(curDraftIdURL));
337   NS_ASSERTION((NS_SUCCEEDED(rv) && (!curDraftIdURL.IsEmpty())), "RemoveCurrentDraftMessage can't get draft id");
338 
339   // Skip if no draft id (probably a new draft msg).
340   if (NS_SUCCEEDED(rv) && mMsgSend && !curDraftIdURL.IsEmpty())
341   

That is, the comment claims that new drafts will have curDraftIdURL.IsEmpty() and it behaves correctly. This is no place for an assertion.

I can get this simply by composing a new message, and trying to save it.

I say we just delete the assertion. It was added with the code for the entire method in Bug 319818.

Comment 6

[Kent James (:rkent)]

Attachment: Remove the assertion

Comment 7

[:Irving Reid (No longer working on Firefox)]

Comment on attachment 669807 [details] [diff] [review]
Remove the assertion

Review of attachment 669807 [details] [diff] [review]:
-----------------------------------------------------------------

There's lots of silent discarding of nsresults in this method, but I'm happy to leave that to another bug unless you want to take it on.

Comment 8

[Kent James (:rkent)]

(In reply to Irving Reid (:irving) from comment #7)
> 
> There's lots of silent discarding of nsresults in this method, but I'm happy
> to leave that to another bug unless you want to take it on.

I'm starting to figure out that that is a hot issue for you. For this bug though, I simply encountered the Assertion myself in some code work I was doing, analyzed it enough to figure out it was a false positive, then decided I would go ahead and do that narrow fix. I'm reluctant to do much more at the moment.

Comment 9

[Kent James (:rkent)]

Checked in https://hg.mozilla.org/comm-central/rev/82d555dbf346