At 03:43 PM 2002-08-21 +0200, Norbert Bollow wrote: >Why not put both into the same header (possibly folded across multiple >lines)?
Because they might be added in different stages. I'll admit that I can't come up with a good reason for doing it now, but I do not see the problem with allowing for it as long as a program can parse any of the individual headers and come up with the same answer. >List-Policy-Archive: Ask (Compilation Copyright © 2002 Example Inc. > All Rights Reserved. For policy details refer to > <http://example.com/policy/will-sue-unauthorised-archivers.html>) > > > They should be respected by a program no matter where they appear - in a > > plain text section, in the mime headers, in the unlabled area that follows > > the RFC822 headers and precedes the first mime headers, or in the top > level > > RFC822 headers. If they are put into a html section, or a section that is > > encoded in a scheme other than seven bit ascii (uuencode, base64, > > quoted-printable), they must be repeated in one of the above places for > > programmatic access. > >If the policy file is HTML, it could be required that the >specifications are enclosed in <PRE> .. </PRE>. The problem is the potential for encoding. If someone decides to encode the html section in base64 or quoted printable (and this could be done by the MTA if the section is 8-Bit, even if it is not sent encoded), then the body headers may not be machine parsable until they are decoded, whereas the places I mentioned are not, as far as I know, ever encoded. You can be reasonably sure that the top level headers will never be encoded. I would be loath to require that the programmer writing a parser for this decode sections or interpret html. I am not sure how you stop your section from being encoded, except for not using 8 bit characters. -- "A fear of weapons is a sign of retarded sexual and emotional maturity." -- Sigmund Freud, General Introduction to Psychoanalysis (1952) Nick Simicich - [EMAIL PROTECTED] - http://scifi.squawk.com/njs.html
