I've been perusing the spec, and was reminded that we still have the simple body canon. All it does is strip off trailing CR LF pairs, and hash what's left. Bare CRs? Eight bit glop? No problem.
If we do an implementor's guide, I'd have it say that if your goal is simply to shoot a message in whatever form from one MTA to another, use simple canon and it'll work. On the other hand, if your goal is to make your messages and signatures maximally tolerant of unknown relays down the line, be sure it's perfectly 2822 conformant and use a relaxed signature. Those are both reasonable scenarios, we deal with both of them, so I don't see any problems left to be solved. R's, John PS: Sec 3.4.3 says to add a trailing CR LF if none is present, although a message transmitted by SMTP has to have one. But I don't think that's worth fixing. _______________________________________________ NOTE WELL: This list operates according to http://mipassoc.org/dkim/ietf-list-rules.html
