Re: [Mailman-i18n] Re: Add header/footer to MIME multipart
Am Sonntag, 17 November 2002 schrieb barry@python.org (Barry A. Warsaw) :
Yes it would. Question, should we copy all the outer message headers to the inner message, or just some subset? And if a subset, which ones?
It think finding the minimum subset would be best: all MIME headers must be present (Content-Type, Content-transfer-encoding, perhaps Content-disposition, but not MIME-Version). All other headers belong to the outer chrome, AFAICT. Martin =================================================================== EASY and FREE access to your email anywhere: http://Mailreader.com/ ===================================================================
"MvL" == Martin v Löwis
writes:
>> Yes it would. Question, should we copy all the outer message >> headers to the inner message, or just some subset? And if a >> subset, which ones? MvL> It think finding the minimum subset would be best: all MIME MvL> headers must be present (Content-Type, MvL> Content-transfer-encoding, perhaps Content-disposition, but MvL> not MIME-Version). All other headers belong to the outer MvL> chrome, AFAICT. Yep, that's actually what's implemented in cvs now. I copy any header starting with "Content-". Seems to work okay in my limited testing. Thanks, -Barry
Barry A. Warsaw
MvL> It think finding the minimum subset would be best: all MIME MvL> headers must be present (Content-Type, MvL> Content-transfer-encoding, perhaps Content-disposition, but MvL> not MIME-Version). All other headers belong to the outer MvL> chrome, AFAICT.
Yep, that's actually what's implemented in cvs now. I copy any header starting with "Content-". Seems to work okay in my limited testing.
RFC 2045 http://www.ietf.org/rfc/rfc2045.txt also states: | 2.4. Entity [...] | [...] only those fields whose names begin with "content-" actually | have any MIME-related meaning. So your decision is perfectly right! :) Cheers, Martin
participants (3)
-
barry@python.org
-
Martin Ramsch
-
Martin v. L�wis