324710 - Consider fetching IMAP message flags in the background

Status: RESOLVED DUPLICATE of bug 1747311

(MailNews Core :: Networking: IMAP, defect)

Description

[Scott MacGregor]

For servers that have a hard time fetching the flags for the messages in a folder, we could look into fetching the flags in the background and fetching them in smaller chunks.

Could be a perceived performance win for users using IMAP servers which don't handle this operation very well.

Comment 1

[Ludovic Hirlimann [:Usul]]

David was this implemented in auto-synch ?

Comment 2

[David :Bienvenu]

No, this bug is about the flag synching we do when we open a folder; autosync is about syncing the message bodies. Right now, when you open a folder that doesn't support condstore (see bug 336151), we do a fetch 1:* FLAGS to see if any of the flags have changed before we fetch any new message headers - this bug is proposing fetching the flags in chunks in the background.

Comment 3

[Wayne Mery (:wsmwk)]

(In reply to David :Bienvenu from comment #2)
> No, this bug is about the flag synching we do when we open a folder;
> autosync is about syncing the message bodies. Right now, when you open a
> folder that doesn't support condstore (see bug 336151), we do a fetch 1:*
> FLAGS to see if any of the flags have changed before we fetch any new
> message headers - this bug is proposing fetching the flags in chunks in the
> background.

this would help gmail, which currently doesn't support condstore.

however, it looks like gmail might support it in the not too distant future http://comments.gmane.org/gmane.mail.imap.general/3199

Comment 4

[Ludovic Hirlimann [:Usul]]

Removing myslef on all the bugs I'm cced on. Please NI me if you need something on MailNews Core bugs from me.

Comment 5

[Wayne Mery (:wsmwk)]

Gene, do you think this still has practical value? But only or mainly for very large folders?

Comment 6

[gene smith]

This is probably covered by condstore (but disable by default) and what I did to improve condstore and add qresync here (but on back burner): bug 1747311.
Maybe close as a duplicate of bug 1747311?