Ned Freed wrote:
>>Ned has published the environment extension (Thanks Ned!). I would like
>>to solicit some reviews.
>>
>>
>Yes please.
>
Some comments after a quick scan through the document:
In section 4.1:
> "place" => Sieve processing is normally performing around or after
> the time of final delivery. This item provides
> additional information about the relationship to final
> delivery. Possible return values are "MTA", meaning the
> Sieve is being evaluated before final delivery, "MDA",
> meaning evaluation is occurring during final delivery",
> and "UA", meaning evaluation is occurring after final
Why not "MUA" for consistency?
Sure, why not?
> delivery.
I would also like to add an environment item for protocol invoking the
Sieve engine. The item value can be one of "SMTP", "LMTP" and "IMAP".
(Or can just reuse the WITH field registry for the Received header field).
It's not clear to me that that this provide sufficiently different
information than the place value does.
I also don't believe IMAP is a legal value for with, plus there are a whole
bunch of registered values for WITH that may or may not make sense here.
Also an item that can convey that there is a single SMTP recipient would
be nice as well.
IMO this isn't an environment test, but rather an envelope extension. And there
are a bunch of other things we need to add to envelope besides this - NOTARY
information being the most obvious. I considered defining an envelope extension
in this document, but decided it was too much to start with. If people feel it
wouldn't be too much bloat I can go ahead and put it in as a separate
extension.
Both can be used with the reject action.
Also the [expired] draft-ietf-lemonade-imap-sieve-00.txt draft defines
the following environment items (as IMAPSieve.<var> variables):
cause
mailbox
changed.flags
changed.annotations.
But I guess their registration can be done in the IMAP Sieve draft itself.
I don't feel strongly about it either way. If people think it woul be best
to have these here I can copy the definitions over.
The IANA registration section should use the updated Sieve extension
registration template from 3028bis.
Yes. I'll update them here and in the data/index document. I note that vacation
is in the RFC Editor queue (waiting for the base spec and variables) and
needs a similar update - I guess I'll do that during AUTH48.
Ned