The way I read it, "Charset" is little more than a comment. An
implementation may use the Charset header if it wants to, but on the
recipient side, the other implementation is free to ignore it.
The safe assumption if there is no Charset header should be that the
plaintext is UTF8.
I've cc'd the list, so perhaps someone there will comment as well.
please have a look at felix question - its about the UTF8-handling in
messages. PGP8 does it "different" and we would be happy about some
clarification in this topic. what yould you suggest? is it worth to put
it on the list and if, i would appreciate if you could just forward
thanks a lot,
Sent: Montag, 27. Oktober 2003 18:26
Subject: Non UTF-8 Text in Message Body
just to clear things up a bit for me:
Encoding: Is it allowed to encode the message body of a mail in any
other character set than UTF-8 without using the "Charset" armor header
key? If I interpret the standard and the current draft correctly, this
header key MUST be used if text is encoded in any other character set,
but I think there are some implementations that do not obey this rule...
Decoding: Is it allowed to assume that an encoded message body is just
plain ISO-8859-1 if no "Charset" armor header key is present?
Thanks and best regards,
Glueck & Kanja Technology AG