> -----Original Message-----
> From: John R. Levine [mailto:[email protected]]
> Sent: Monday, November 02, 2009 10:15 AM
> To: Murray S. Kucherawy
> Cc: [email protected]
> Subject: RE: [ietf-dkim] Interesting Dupe Signatures
> 
> I don't see much benefit for saving the header hash, since it depends
> on
> the order that the headers are listed in the h= header.

Right, but if the "h=" and the "c=" for the header are the same, you can reuse 
them.

> On the other hand, doing both the simple and relaxed body hashes in a
> pass
> over the body would handle about 99.999% of the signatures that anyone
> cares about.  I haven't seen many l= signatures, other than from Cisco,
> and I don't think I've seen any where the l= didn't cover the whole
> message.

That might also be a good optimization, especially if we can assume that SHA is 
cheap.

_______________________________________________
NOTE WELL: This list operates according to 
http://mipassoc.org/dkim/ietf-list-rules.html

Reply via email to