On Thu, Feb 23, 2012 at 3:15 PM, Dave Cridland <[email protected]> wrote: > On Thu Feb 23 15:43:56 2012, Tony Hansen wrote: >> >> Totally off topic: I wish I knew how to get them to fix that. > > > And now you know...
I'm sure there are other corner cases as well, just looking through the code there are all sorts of caveats and such to work around bugs generated by specific clients. To bring this back on-topic, if the goal is simpler, is it simpler for the server to implement parsing or to punt that to the client? This specific case could have been solved by the client implementing a viewer for rfc822, though obviously its heavier weight if the embedded message has many other levels. OTOH, its simpler on clients if they don't have to implement all of the work-arounds for broken mail that we do. Or maybe a much simpler flattened structure for messages as parsed by the server and you can download the whole thing to parse the original if you want higher fidelity. Of course, if we aren't parsing the messages on the server, we have to ask why we're implementing IMAP5 and not POP4 (still not sure if that statement will make Marc laugh or cringe). Brandon _______________________________________________ imap5 mailing list [email protected] https://www.ietf.org/mailman/listinfo/imap5
