Re: IMAP-problems

[Darko Obradovic <dobradovic gmx de>]

On Thu, 17 Apr 2003 23:25:18 -0400
Peter Bloomfield <PeterBloomfield@bellsouth.net> wrote:

> I'm guessing you checked `Always show INBOX' when you set the prefix, 
> or else you wouldn't get INBOX at all, I believe.  In that case, the 
> issue may be at lines 183-188 in libbalsa/folder-scanners.c:
> 
>      if (list_inbox)
>          /* force INBOX into the mailbox list
>           * delim doesn't matter, so we'll give it '/'
>           * and we'll mark it as not scanned, in case it
>           * has any subfolders */
>          mailbox_handler("INBOX", '/', FALSE, cb_data);
> 
> If you change the last FALSE to TRUE, you may get rid of the 
> duplicates.  Let us know!

You're right on all aspects. One problem less with the patched version now. :)
I don't know what the correct behaviour is according to the RFCs, but, forgive me I'm mentioning Sylpheed all the time, it's the only alternative I have and use for the moment, does all that as I would expect it in my situation: including INBOX (this is not an option anywhere, hard-coded default) without any childs. They must have some reasons for that I guess.

But Courier-IMAP, although definitely IMAP4rev1-compliant according to some serious sources, seems to be a little strange, check out this one for example:
http://www.mail-archive.com/lists-bincimap@infeline.org/msg00244.html
In the RFC, I found this one, but not understanding much, as usual...
>>>>>
  The special name INBOX is included in the output from LIST, if
  INBOX is supported by this server for this user and if the
  uppercase string "INBOX" matches the interpreted reference and
  mailbox name arguments with wildcards as described above.  The
  criteria for omitting INBOX is whether SELECT INBOX will return
  failure; it is not relevant whether the user's real INBOX resides
  on this or some other server.
<<<<<

Going to the next point in scanning: When calling the LIST-command, the server returns for example
0x82ac978 s: * LIST (\HasNoChildren) "." "INBOX.lists.balsa"
and although there is a nochildren-flag, balsa is scanning every folder for subfolders again.
Looking into section 7.2.2 of RFC 2060 for that, it isn't officially listed, only \Noinferiors \Noselect \Marked and \Unmarked are defined there. It is defined in an ietf-draft here:
http://www.ietf.org/proceedings/00dec/I-D/draft-ietf-imapext-list-extensions-00.txt
I don't know what your philosophy towards the standards is, but implementing it would save some scanning-time, especially on large boxes or slower connections I guess.

enough for now, although there are some more püoints left, but I'm afraid to scare you away... *g*

bye,

Darko