[mailto:owner-spf-discuss(_at_)v2(_dot_)listbox(_dot_)com]On Behalf Of Frank
Sent: Tuesday, July 19, 2005 7:15 PM
Subject: [spf-discuss] Re: Test tool for type 99
I would say "yes". Section 4.5 doesn't say "equivalent", it
Fix it somehow, it's some RfC 3597 oddity. Maybe add "(case
insensitively)" to "identical". Don't try "equivalent", it's
okay to sort adjacent directives with the same result in any
way, or to move modifiers around (only in v=spf1, not spf2.0).
OK. I went and read RFC 3597. The way I read that and the way I read
Wayne's draft, case matters for being identical, so for Stuart's:
testspf2 IN TXT "v=spf1 A -all"
testspf2 IN TYPE99 \# 14 0d763d737066312061202d616c6c
They are not identical (the TYPE99 converts to "v=spf1 a -all" and so they
should fail the validation test (not that any operational implementation
would know or care). Identical is an easy test to implement, if a != b then
raise an error. "Works the same" would be a substantially more complex test
to write. So for now the validator says if a != b then raise an error and
that's all I'm going to do on that.