Scott Kitterman wrote:
Hi Scott, your tool still hates mx::%%%_/.claranet.de/27 ;-)
It will parse that now.
LOL, maybe we need two SPF validators, "expert mode" for the
weird constructs, and "simple mode" for the real SPF records.
BTW, it does accept mx:invalid but that's _really_ invalid:
domain-spec = macro-string domain-end
domain-end = ( "." toplabel ) / macro-expand
There's no dot in invalid, and it's no <macro-expand>. For a
minimally "valid" invalid I'd need mx:.invalid with a dot.
It also says "semantically valid" instead of "syntactically":
For a semantical test I'd expect a warning for mx:test.invalid
(or similar), because "host" test.invalid does not have any MX.
Bye, Frank