On Thu, 23 Mar 2006, Frank Ellermann wrote:
william(at)elan.net wrote:
domain-spec = macro-string domain-end
domain-end = ( "." toplabel ) / macro-expand
toplabel = ALPHA / ALPHA *[ alphanum / "-" ] alphanum
; LDH rule (See [RFC3696])
Nothing's wrong here, only "must start with ALPHA" is a bit
_too_ restrictive wrt 3696: The 3696 toplabel rule is "must
contain not only digits". But so far that exists only in a
USEFOR draft as ABNF, 3696 hat it in prose.
The SPF spec. took the simple toplabel ABNF of 2396 (obsolete).
The "official" toplabel rule is even more restrictive than SPF.
No issue here from our POV, and we don't need any trailing dot.
You forget about famous TLD as a host case (currently only "ws") that
John Klensin wants to fix by requiring trailing dot when referring
to it as a host so "email(_at_)tld(_dot_)" would be valid address for RFC2821.
And yes, some administrators regularly add "." to the end of domain
specification to signify its FQDN and not local hostname to be searched
within domains specified in /etc/resolv Resolvers all deal with this nicely.
So I need direction here. Do we want to consider existing SPF records
with redirect (that end with ".") at the end to be syntax error now or do
we want some kind of last pre-RFC publication effort to correct SPF
spec and allow for it?
Please lets have comments from those who have implemented the spec
as to how you handle it (so the spec meets majority deployment when
published).
--
William Leibzon
Elan Networks
william(_at_)elan(_dot_)net
-------
Sender Policy Framework: http://www.openspf.org/
Archives at http://archives.listbox.com/spf-discuss/current/
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?listname=spf-discuss(_at_)v2(_dot_)listbox(_dot_)com