[Imap-protocol] QRESYNC and new arrivals which get deleted immediately through VANISHED

[Alexey Melnikov]

Hi Jan,

On 17/07/2012 18:11, Jan Kundrát wrote:

> On 06/08/12 18:02, Jan Kundrát wrote:

>> Finally, my last question is why does QRESYNC require me to issue an

>> ENABLE QRESYNC at all? I feel like the SELECT ... QRESYNC should be

>> enough to tell the server that I'm really expecting the VANISHED

>> responses.

>

> I think I have found an answer to this -- if the client doesn't have 

> any cached state for a particular mailbox, it cannot really SELECT 

> QRESYNC that.


Right.


> If it tried to use a HIGHESTMODSEQ = 1, it would risk an enormous 

> amount of data being transferred in VANISHED EARLIER as the server 

> would have to inform about each and every expunge which has happened 

> in the mailbox since its creation. This is due to the known-uids ABNF 

> item format which doesn't provide any way of sending "nope, I don't 

> know about any UIDs" -- the list can either be missing, or contain at 

> least one item.


Yes.


> If I understand everything correctly, ENABLE QRESYNC is therefore 

> still required as long as one wants to receive VANISHED instead of 

> EXPUNGE.


Exactly.


> My apologies for assuming that it's useless :).

>

> Also is probably means that a QRESYNC-capable clients have to use 

> SELECT ... CONDSTORE when they open a mailbox for the first time.


This can be an option.


> I'd appreciate if someone could verify these assumptions -- they look 

> plausible to me, but I didn't see any flaw in my previous reasoning, 

> either :).