On Wed, 4 Dec 2002, David Harris wrote: >On 4 Dec 2002 at 8:15, Andreas Aardal Hanssen wrote: >> >> If someone does a STATUS on a mailbox that the IMAP server has not seen >> >> before - what is expected output for UIDNEXT and UIDVALIDITY? >> >I don't understand this question. >> Since UIDVALIDITY is an IMAP-feature, and an "unseen" mailbox, that is >> - a mailbox that you have never selected before and never done any >> operation on before with IMAP - has no UIDVALIDITY, what is the >> expected behavior of the IMAP server when the client issues a STATUS >> <mailbox> (UIDNEXT UIDVALIDITY) ? >I think you might be confusing IMAP, which is a general-purpose, >abstracted protocol for accessing a message store, with the specific >implementational details of the particular message store for which it is >acting.
I'm not confusing these two - I'm asking specifically about the user level expectation that greatly affects the implementation details. So far, it seems that people agree that STATUS should either (1) show the current UIDVALIDITY/UIDNEXT, or (2) create a UIDVALIDITY and then parse the mailbox and assign UID values to the messages there, then (1). Andy >>From the IMAP perspective, there's really no such thing as "a mailbox >the server has never seen before": in the case where a server is >offering the contents of a mailbox (assuming that you mean "mailbox" in >the IMAP sense of "a folder") for the first time, then it should simply >create a UIDVALIDITY value appropriate for its internal purposes - the >client can make no assumptions about what the value itself actually >means. Clearly, if the server has never accessed the mailbox before, >then nor has the client, so the client has no basis for expecting anything >in this situation. > >If the message store your server is presenting does not inherently >record message ordering, or if it uses a unique ID scheme other than a >32-bit incrementing value, then it is up to your server to create and >preserve some kind of record that handles the translation between that >scheme and the scheme required by IMAP. The exact details of how >you go about this are up to you, but you have to do it - it's not in any >sense optional. > >> Sorry again - I refer to the process of assigning UIDVALIDITY value to a >> mailbox and UID values to messages. > >Without wanting to stress the obvious here, Mark is absolutely right - the >client can infer nothing from the value of UIDVALIDITY except that it is >the same as or different from the last value the server expressed. > >Cheers! > >-- David -- > >------------------ David Harris -+- Pegasus Mail ---------------------- > Box 5451, Dunedin, New Zealand | e-mail: [EMAIL PROTECTED] > Phone: +64 3 453-6880 | Fax: +64 3 453-6612 > >Thought for the day: > Cleavage (n): something you can approve of and > look down on at the same time. -- W. Garnett. > > > > -- Andreas Aardal Hanssen
