On Fri, May 06, 2005 at 06:35:34AM +0200, Frank Ellermann wrote:
Alex van den Bogaerdt wrote:
RFC2821 does not define "missing dot".
| From: John C Klensin
[...]
|>| Domain = (sub-domain 1*("." sub-domain)) / address-literal
|>
|> Is that intentionally "at least one dot", or is it a typo ?
| Yes, that was intentional.
[...]
| you are welcome to repost as needed.
Good. Section 2.3.5 defines domain to be one or more
dot-separated components (note "one or more"), then refers
to section >>>5<<< (not 4) and [22] (which is rfc1035).
rfc1035 defines "tv" a valid domain:
<domain> ::= <subdomain> | " "
<subdomain> ::= <label> | <subdomain> "." <label>
<label> ::= <letter> [ [ <ldh-str> ] <let-dig> ]
Path taken: domain ::= <letter><let-dig>
So, rfc2821 seems to contradict itself by formalizing a syntax it
didn't describe, pointing to a formal syntax it _does_ describe.
I can't place "you are welcome to repost as needed" but if that
is about the rfc, it seems to me a bug is acknowledged.
Anyway, with multiple references to "tv" being valid, and just
one reference to the opposite, I still say "tv" is alright as
domain, even by rfc2821 standards although I must admit this
error somewhat weakens my argument. However, to end this useless
subtopic: It all starts with a HELO parameter being something random
or not. _ONLY_ a host actually named "tv" (or, as you could see
it: "tv.<empty_label_at_root>") may use this as the HELO parameter.
And by using SPF, the owner of "tv" can actually express his policy
about this.
There's nothing random about helo names.
Alex