Michael M Slusarz wrote:
Quoting Per Jessen per@computer.org:
I'm trying to get my head around this - the special_use flag is given out to an IMAP client as a _hint_ of where certain messages belong?
Yes. There's no guarantee that any of the messages currently in that mailbox adhere to that attribute though, so you can't make assumptions about any individual message. But from a configuration perspective, it can be useful to at least provide a better default value(s) than a MUA-specific hard-coded value.
Right, thanks.
If that's correct, what would be the purpose of having two (or more) such hints for e.g. \Sent ?
We allow different "identities" to be configured in our client -- e.g. I send out messages under a number of different e-mail/header/signatures. It is *very* handy to have the sent-mail messages saved to a different mailbox for each identity. The RFC defines the \Sent flag to include any mailbox "used to hold copies of messages that have been sent." By this definition, each mailbox should be flagged with \Sent.
Or you have an existing account that contains pre-existing "Sent", "Sent Mail", and "sentmail" mailboxes. All of those are valid sent mailboxes under the \Sent definition.
Yes, that would be an example - I have at least 6 different names for "Sent", some accounts use three or even four. I gather support of these hints is still quite limited, but I'm still struggling with understanding what good multiple hints of e.g. "Sent" is to a client? Is the hint also used to identify places to look for sent mail?
Interesting stuff - now what about the actual problem of mapping multiple such mailboxes into one? My immediate thought was "symlinks" - would that work?
-- Per Jessen, Zürich (12.9°C) http://www.hostsuisse.com/ - virtual servers, made in Switzerland.