[My apologies for the delay in this as well. Barry, you may scowl at me
when we next meet.]
The leading broilerplate is out of date. Possibly the trailing
broilerplate too.
I would suggest striking the second sentence of section 2 ("Any
comparator used..."), as this document is not the source of that
requirement and the list of exempted comparators is no longer accurate.
In the third paragraph of section 3.1, I think "of the message" should
be changed "from the message".
The last comma in section 3.1 ("...considered true[,] if any pair...")
should be removed
In section 3.2, the phrasing of what the various tests do is a bit odd.
The first paragraph talks about counting entities, but then there's no
further mention of "entities". If that first paragraph is left as is,
then I think the following paragraphs should be phrased more like:
When used with the "address" test, the contained "addr-spec" elements
[RFC2822] are the counted entity. Note that addr-specs contained in
group addresses are counted, but the group itself does not affect the
count.
When used with the "envelope" test, any envelope-part whose value is
non-empty will contribute one (1) to the count. In particular, the
"to" envelope-part will always contribute exactly one.
(We need to leave the door open for other envelope-parts to be defined.)
When used with the "header" test, the instances of the named fields
are the counted entity. The values of the instances has no effect on
the count.
...though that text seems clunky too.
The reference to MATCH-TYPE in the ABNF will be rejected as there's no
previous definition of the MATCH-TYPE non-terminal, not even in the
base-spec, so it can't be added to with "=/". I'm not sure what the
best fix for this is.
Philip Guenther