[mail-vet-discuss] authentication-results grammar w.r.t. propspec

Tony Hansen tony at att.com
Mon Oct 22 13:01:42 PDT 2007


I've noticed that a number of implementations haven't been putting in a
propspec when the result is a failure. And I think this may be the
correct thing to do in certain cases.

If this is the consensus of the group, then this change to the ABNF
would be in order, to make propspec optional:

old:      header = "Authentication-Results:" [CFWS] authserv-id
old:               [CFWS] *(";" methodspec propspec [CFWS])

new:      header = "Authentication-Results:" [CFWS] authserv-id
new:               [CFWS] *(";" methodspec [propspec CFWS])

	Tony


More information about the mail-vet-discuss mailing list