...... Original Message .......
On Mon, 18 Jul 2005 23:27:27 +0200 Frank Ellermann
<nobody(_at_)xyzzy(_dot_)claranet(_dot_)de> wrote:
Scott Kitterman wrote:
always check for type99 and check to make sure that SPF and
TXT are the same when processing in strict mode.
Great, I guess your tool is the only that checks this detail.
I'd appreciate it if anyone who's published type 99/SPF records would let
me know so I can test this.
5 minutes of staring at the code doesn't tell me for sure
do I have a <macro-expand> problem or not.
You catch mx:invalid now, dito mx:com etc. (no dot), testing
mx:%{d} I get the same error, so that's not yet correct. What
are your defaults for the macros like %{d} ?
As Stuart said, I should've checked for '.' after macro expansion. The
%{d} never gets called with this bug.
The 10/10/10 processing limit is fully implemented under
strict processing in pySPF, so more than 10 MX is a PermError
AFAIK more than 10 names for an MX are ignored by the receiver,
but don't cause a PermError (same logic for PTR). The result
PermError for "v=spf1 a a a a a a a a a a a" is nice :-) Bye
The 10 MX limit is a MUST, so I think PermError is appropriate.
Scott K