Hi Oliver,

Oliver Blasnik wrote:

> As you see, IDLE in a non-selected state is valid. Couriers implementation
> is wrong (responding with a error message). Btw, your implementation is
> also wrong in selecting INBOX ;)


Interesting point there. I have been puzzled all along on why outlook 
would possibly want to do an idle without selecting a mailbox. So I 
figured the expected behaviour on the server part was to autoselect 
inbox, which I now realize was a wrong conclusion. The programmer must 
have put this idle call outside the mailbox handling as if the client 
was required to enter idle mode whenever it was idling itself. Funny, I 
guess it takes a MS programmer mind to interpret the rfc like that. It 
sure didn't hit my mind as a reasonable thing to do.

> If going IDLE while not SELECTed, the server MUST NOT report any error
> and also MUST NOT report changes of any mailbox.


Yes, your interpretation makes sense. As a matter of fact, after reading 
your explaination, I'm pretty sure outlook is disregarding any reports 
while it has not explicitly selected a mailbox itself. I think I will 
change the patch to reflect this and test it against an outlook client 
(have to ask someone, I have not access to OL myself).

Thanks, Tomas


_______________________________________________
courier-users mailing list
[EMAIL PROTECTED]
Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users

Reply via email to