On Wed, 20 Jul 2005, Frank Ellermann wrote:
Stuart D. Gathman wrote:
testspf2 IN TXT "v=spf1 A -all"
testspf2 IN TYPE99 \# 14 0d763d737066312061202d616c6c
[...]
Notice that for testspf2.bmsi.com, the records differ only
in the case of 'A'. Should that be an error? :-)
Yesterday they sent a "proposed standard" to its "last call"
as "draft standard", see <http://purl.net/net/rfc/3597>
| As a result, when a new RR type contains one or more embedded
| domain names, it is possible to have multiple RRs owned by
| the same name that differ only in the character case of the
| embedded domain name(s). This is similar to the existing
| possibility of multiple TXT records differing only in
| character case, and not expected to cause any problems in
| practice.
If you think that this statement in RfC 3597 is wrong, now is
the time to say so... ;-)
The difficulty is that checking for a match then requires fully parsing
both records. When the records must be identical, checking for a
match is trivial. In production, a checker could fetch either
record, and assume they are the same without actually checking.
However, SPF lint tools will need to check.
I'm thinking that pyspf will need an 'anal' option that goes
beyond the current 'strict' option for use in validation.
--
Stuart D. Gathman <stuart(_at_)bmsi(_dot_)com>
Business Management Systems Inc. Phone: 703 591-0911 Fax: 703 591-6154
"Confutatis maledictis, flamis acribus addictis" - background song for
a Microsoft sponsored "Where do you want to go from here?" commercial.