spf-discuss
[Top] [All Lists]

Re: draft-schlitt-spf-01 now available

2004-11-20 12:44:10
In <5705(_at_)rama(_dot_)pamho(_dot_)net> "Roger Moser" 
<Roger(_dot_)Moser(_at_)rama(_dot_)pamho(_dot_)net> writes:

Wayne wrote:

* The ABNF for the Received-SPF: headers needs to use LWSP instead of SP.

The syntax for the Received-SPF: header is still broken. It is ambiguous.

Yep, it always has been, and it isn't just the the additional keywords
that have problems.  There is nothing in the ABNF for the comment
portion of the Received-SPF: header that prevents closing parenthesis.

I think quality SPF implementations won't generate Received-SPF
headers that contain closing parenthesis, semicolons in values, etc.
I think that bad SPF implementations won't generate Received-SPF
headers that conform to any ABNF that we create.  Any parser is going
ot either have to know what SPF implementation the receiving MTA used,
or assume the worst.


I guess I could try and clean up the mess.  But, before I spend the
time, I need to know:

1) is backwards compatiblity with spf-draft-200406 important?  (That
   is basically what I've done.)

2) Should we just make certain characters illegal?

3) Should we try to create an escape mechanism?


-wayne