Re: Sieve mime loop issues list

[Ned Freed <[e-mail]>]

> Ned Freed <[e-mail]> writes:
> >> #3: We need to add some way to look at parameter lists:
> >>     Content-Type: text/plain; charset="foo"
> >
> >Agreed, but mostly because of RFC 2231.

> It also avoids problems with comments and quoting (dang structured fields!).


> ...
> >but I do want to see some way to do address and exists tests on inner
> >headers. This is actually quite a bit more important than a parameter
> >test IMO.

> Do you mean "the MIME header of inner parts" or do you mean "the
> headers of nested message/rfc822 parts"?   If both, how should they
> be differentiated?

AFAIK we have not defined a field that involves an address for a MIME
part. So unless you want to avoid false matches on bogons in MIME parts,
I don't think this makes a difference.

In any case, you probably could rig up something to only test message/rfc822
parts using variables, assuming the traversal is top-down.

Chris Newman also suggests renaming this as foreach, in order to make it
clearer we're still not Turing-complete.

				Ned