Open Bug 496254 Opened 12 years ago Updated 9 years ago

New mail alert says "N new mails" even if all downloaded N mails is moved to local Junk folder by "Junk filtering for message-filter-moved mails" (POP3)

Categories

(MailNews Core :: Filters, defect)

x86
Windows XP
defect
Not set
normal

Tracking

(Not tracked)

People

(Reporter: World, Unassigned)

References

(Blocks 1 open bug)

Details

When all mails(say N mails) are moved to a folder by message filter, and when all the moved mail is spam, new mail alert says "N new mails" even if all downloaded N mails is moved to Junk folder by "Junk filtering for message-filter-moved mails".

Problem is observed with Tb 2.0.0.21 and next Tb trunk build.
> Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1pre) Gecko/20090601 Shredder/3.0b3pre

[Steps to produce]
(0-1) For a POP3 account create message filter.
      - Match all messages, Move to a local mail folder(say Folder-1)
(0-2) At POP3 Mbox, N spam mails exist.
(1) Get New Messages.
(2) All N mails is moved to Folder-1.
(3) Because no new mail in Inbox any more, Junk filter for Inbox does nothing.
    => New mail alert is probably issued at here, then "N new mails"
(5) Junk filtering for mails in Folder-1 which was moved at step (2)
    => All mails is moved to Junk

I think many bugs are already opened for this issue after Tb 2.0 because (5) is new feature from Tb 2.0. External symptom of bug depends on mail count and expressed external symptom of bug depends on bug opener.
  Total downloaded mails == N == A + B + C + D 
    (a) Inbox:          A, non-spam, non-filtered mails 
    (b) Folder-1 to X : B, non-spam, filtered mail by message filter
    (c) Junk #1:        C, spam, moved from Inbox
    (d) Junk #2:        D, spam, moved from Folder-1 to X
  New mail alert: (A + B + D )  Note: C is already excluded.
  For user      : Number of new mails == (A + B)
  This bug's test case : A=0, B=0, C=0, D=N

Workaround:
 - Add action of "Mark As Non-Junk" to filter rule which has "move to folder".
Summary: New mail alert says "N new mails" even if all downloaded N mails is moved to Junk folder by "Junk filtering for message-filter-moved mails" → New mail alert says "N new mails" even if all downloaded N mails is moved to local Junk folder by "Junk filtering for message-filter-moved mails" (POP3)
This has been covered by bug 233930, bug 460279 and bug 11040. I'll leave this open, though, as this symptom is very specific and very important, and the report is written clearly (unlike the others).

Compare bug 272125, which is about biffed Junk mail that is left in Inbox (different bug).
> This has been covered by (snip) bug 11040.

Bug 11040 is for "user controllable biff". But this bug is for above "D" in standard new mail alert. Really covered by Bug 11040?

By the way, AFAIK, Bug 189289(fixed by Tb 1.5) had fixed problem of "new mail alert includes above C" (for Tb 0.x to Tb 1. phenomenon of Bug 272125 Comment #0).
> Really covered by Bug 11040?

Sorry, I missed the POP3 pref part of this bug.
(Slightly Off-Topic)

> this symptom is very specific (snip), and the report is written clearly ...

It's simply because that "my guess/assumption and test case of this bug" bases
on your problem observation/analysis in Bug 272125. My guess doesn't seem
wrong, although I think it's never absolutely correct.

> Sorry, I missed the POP3 pref part of this bug.

I carefully excluded IMAP case, in order to make problem analysis simple, although I believe this bug is applicable to many(but not all) IMAP cases too :-)
FYI.
Another bug which causes "wrong message count" for user:
Bug 381017 Comment #6 reported "negative unread count" due to "Mark As Read" by filter.
FYI. Bug 378582 is for "no new mail alert" when filter moves mails to other account's folder(Bug 378582 is move to "Local Folders" case).
Bug 487418 is found for IMAP case. Because patch for IMAP has been proposed, that bug has blocking‑thunderbird3 flag.
Blocks: 534029
Duplicate of this bug: 534029
Blocks: biff
You need to log in before you can comment on or make changes to this bug.