Re: [imapext] draft-ietf-appsawg-multimailbox-search

[Bron Gondwana <[e-mail]>]

On Thu, Mar 6, 2014, at 09:42 PM, Barry Leiba wrote:
> IMAP denizens...
> 
> Chris Newman has asked that we consider moving the Multimailbox Search
> extension to Standards Track (it was published as RFC 6237 as
> Experimental).  Appsawg is handling that:
> 
> http://datatracker.ietf.org/doc/draft-ietf-appsawg-multimailbox-search/
> 
> I'd like to ask the IMAP crowd to do two things:
> 
> 1. Review the document and post comments to [e-mail]
> 
> 2. Tell us, in your review or separately here, if you (a) have
> implemented it and/or (b) have interest in implementing it (and
> specify client, server, or both).

We didn't implement it because it wasn't possible to track the results
over time - instead we have a command xconvmultisearch (yes, I know,
ugly name) which:

a) only works within single users, because:
b) we keep a single highestmodseq value across all mailboxes within a single
   user, so we can
c) lock the entire user from modifications while we run the search, and
   hence ensure a consistent read, and
d) return the highestmodseq of that read, so that further requests can
   ask for updates since the previous modseq - and they'll get a list of
   anything added, removed or just touched.

This also handles the folding of related (roughly thread=references, with
some trickyness) messages into a single "exemplar".

So we're strayed quite a long way from IMAP to get what we need.

Anyway - my main problem with it is that users don't just want search - they
want sort.  Running the search across multiple mailboxes means you still need
to fetch all the data for all the matches before you can get a meaningful sort
by any parameter other than mailbox name as the first sort index.  And that
can be a lot of download to the client if you searched a common term.]

I'm not signed up for apps-discuss.  Guess I'd better go do that.

Cheers,

Bron.


-- 
  Bron Gondwana
  [e-mail]