Is this actually a problem for mail? Typically there is one user per
account. It's not like I'm going to be manipulating my mail from my
phone and from nmh simultaneously. I don't have that many hands. Maybe
I'm not being sufficiently imaginative.
My computers manipulate my mail for me (inc, sort, filter) while
I'm using my phone. This has been a problem for me roughly forever.
The imap spec has a concept of uidvalidity, so in the event of anything
that mucks with the underlying mailstore (such as what you might get
after running sortm), the imap server just needs to be smart enough
to notice and tell the client that there is new uidvalidity, thus
triggering a reconnect and rescan. From rfc3501:
The unique identifier of a message MUST NOT change during the
session, and SHOULD NOT change between sessions. Any change of
unique identifiers between sessions MUST be detectable using the
UIDVALIDITY mechanism discussed below.
presumably user doing something makes "should not" situations ok.
This burned me with uw-imap; anytime an mh folder changed, it would
mark a folder with a new uidvalidity number and I'd need a rescan (such
as mail filters inserting a new message). The imap server needs to be
smart enough to tell the client, and the client to reconnect.
I have actually beaten dovecot into submission to successfully navigate
mh folders, although have NOT added functionality in to deal with this
nor have I handled the case of new mail showing up in a folder, or tie-ins
with sequences (so no awareness of unseen, yet). Time and stuff.
It should be possible to have an imap server that can muck about with an
mh backend, although there may be annoyances like this that the user has
to contend with.
As far as using the mh tools to talk to imap, but that seems much
harder. :)
-Todd
_______________________________________________
Nmh-workers mailing list
Nmh-workers(_at_)nongnu(_dot_)org
https://lists.nongnu.org/mailman/listinfo/nmh-workers