On 8/12/07, Robert Burrell Donkin <[EMAIL PROTECTED]> wrote:
>
>
>
> IMHO the mailbox design suffers from being a compromise between a good
> IMAP API and a good general API. the interface and implementation is
> over-complicated but there are many features which are likely to be
> IMAP specific in what was intended to be a general API.



Yeah, i see that now. I've tried to understand the current TorqueMailbox
stuff, the lots of listeners, and mappings, but finally i've just
copy-pasted the relevant codes with some minor restructuring. When I will
have more time, I definitely want to look into througly. The good news is
that i've been able to finish my hibernate/spring based implementation, of
course it's not very tested, and more experimental than the experimental
modules of James :) However during my struggle with IMAP i fixed some bugs,
made some improvements here and there (i've created JIRA issues about), so I
think I'm reasonably successful in this case. I hope you can apply the
patches.
 Currently I have just one IMAP related question. I see in the current
implementation that there is a 'Namespace' conception (however most of the
time it's hard coded), which is intended to provide some top-level grouping
possibilities for mailboxes. So for example the administrator can create a
#shared-dev namespace which every user can read/write. However from the
code/and the incuded IMAP spec fragments, it's not clear how the clients
will discover this  'namespace' things. I mean, the LIST command should only
list mailboxes  from the specified 'reference', so for example LIST "" "*"
will list from #mail.username.INBOX ... but not #shared-dev.anything. Am I
misses some hidden commands, which not implemented yet, and neither
Thunderbird, KMail nor Evolution ever call without some  server issued
message?


BR,
 Zsombor

Reply via email to