Hi Tony,
Just read your new I-Draft and it seems to be solving some of the same problems
that the draft-daboo-sieve-mime-00.txt draft is trying to solve, so it seems
like the two should probably be worked on together.
The original draft is here:
http://www.ietf.org/internet-drafts/draft-daboo-sieve-mime-00.txt
But I made an alternative proposal in response that seemed to have a lot of
support although was never written up into a new draft. You can read my
response here and follow the replies:
http://www.imc.org/ietf-mta-filters/mail-archive/msg02238.html
My proposal for a HEADER-PART tagged argument to the header does mostly the
same job as your part test with :filename/:type/:subtype approach, and my
:bodyparts optional argument does mostly the same job as your for.every.part
loop test. However the proposal does not include facilities for
replace/delete/enclosing body parts.
I think my proposal allows significant extra message matching capabilities with
less additional syntax, so I think it would be good if we could go with testing
syntax more along the lines of what has already been proposed there.
Perhaps there's really two I-Drafts at play here, a written up version of my
proposal for testing body parts, and then an additional one specifying the
ability to replace/delete/enclose body parts which requires a "select" body
part concept. Maybe a foreach test, that is used instead of "if". Then where
you have:
for.every.part {
if anyof ( part :subtype :is "jpeg", part :filename :matches "*.jpg" {
replace "Attachment removed by user filter";
}
}
We'd have this instead:
foreach anyof :bodyparts (header :value :matches "Content-Type" "*/jpeg" ,
header :parameter "filename"
:matches "Content-Disposition" "*.jpg" ) {
replace "Attachment removed by user filter";
}
Cyrus, I expected you'd re-release the draft-daboo-sieve-mime-00.txt I-Draft
encorporating all the feedback but don't see a version 01.txt. Should I do
this?
Nigel