Arnt Gulbrandsen writes:When OE has edited a message, how can you discover that an edit has taken place?
If by "you" you mean a MUA interoperating with Outlook, this is (I think) easy if you store the UID (and/or Message-ID) as well as the timestamp+author "permanent key", so long as you are willing to assume uniqueness of folder+timestamp+author. If you can't find a message in a folder of a particular UID/Message-ID, then if you can find a message of the associated date+author then that latter message is ipso facto (given the uniqueness assumption) an edited version of the original message.
I suppose the timestamp is INTERNALDATE. Note that the timestamp to APPEND is optional; you can't safely assume that clients will preserve it. Doesn't sound unreasonable, though. Testing is in order. (I don't know about any other clients that have "Edit Message" commands, sorry.)
Once again a hypothesized persistent internal key in OE may obviate the need for OE to make this (arguably questionable) assumption, and OE's "Modified" internal metadata also helps. Of course there may be some way that Exchange exposes the "Modified" metadata to IMAP clients that I haven't been able to discover as yet, it only appears that it is not doing so via "plain IMAP".
It probably doesn't. And even if it does, others IMAP servers generally don't. Consider, for example, Outlook talking to the Kerio server via the Kerio MAPI/IMAP provider, or to Cyrus via Bynari Insight Connector.
--Arnt
