[21535] in Athena Bugs
linux 9.1.24: evolution
daemon@ATHENA.MIT.EDU (Ike)
Fri Feb 28 15:42:10 2003
Message-Id: <200302282042.PAA11987@meeko.mit.edu>
To: bugs@MIT.EDU
Date: Fri, 28 Feb 2003 15:42:04 -0500
From: Ike <ike@MIT.EDU>
System name: meeko.mit.edu
Type and version: i686 9.1.24
Display type:
Shell: /bin/athena/tcsh
Window manager: sawfish
What were you trying to do?
Set evolution to automatically apply filters to new messages
What's wrong:
Evolution appears to have a definition of 'new' which does not include
"all unread messages", but rather "messages which were not in the
INBOX last time I checked". This might not exactly be a bug (it might
be the correct thing to do), but it can cause some seemingly strange
behavior.
Please describe any relevant documentation reference:
there's a question in the olc queue about this; what follows is
my response to that question, describing the erractic behavior:
The manual for Evolution 1.0:
http://www.ximian.com/support/manuals/evolution_10/usage-mail-organize-filters.html
Says: "If you store your mail on the mail server and check your mail
with IMAP, filters are applied to the INBOX when you open it. In some
cases, you will see a number of unread messages in the INBOX, click
the INBOX, and then have the messages move elsewhere as they are
filtered out."
I've been experimenting with this by sending myself test messages and
adding a filter for them based on their subject line, and setting the
option to filter new messages in INBOX. I've sent myself a fair number
of identical test messages, and the behavior is strange, but they seem
to get filtered when:
1) I'm viewing the INBOX, I see zephyr notification of the mail I sent
myself, then right-click on INBOX in the folder list, and select "Open
In New Window".
2) I'm viewing another folder, I see zephyr notification of the mail I
sent myself, then left-click on INBOX in the folder view.
3) I see zephyr notification, and press the Send/Recieve button.
But not when:
1) I see zephyr notification, and then force pine to acknowledge the
new messages (even without reading them and removing their 'New'
flag). In this case, evolution still shows the messages as unread, but
I guess it doesn't consider them 'new', somehow.
My best guess is that evolution has a strange definition of what a
'new' message is, and it clearly means something other than 'any
message that is unread'. I'm not sure whether or not to call it a bug,
but I'll submit a bug report just the same.
