[exim-dev] wrong decoding algorithm in $mime_filename

Top Page
Delete this message
Reply to this message
Author: Kjetil Torgrim Homme
Date:  
To: exim-dev
Subject: [exim-dev] wrong decoding algorithm in $mime_filename
I noticed that spec.txt says that $mime_filename contains the RFC2047
decoded Content-Disposition content. this is very wrong, a
Content-Disposition can't even contain a correct RFC2047 encoded-word
due to whitespace rules. to encode/decode MIME parameters, you have to
use RFC 2231, "MIME Parameter Value and Encoded Word Extensions:
Character Sets, Languages, and Continuations."

this ties slightly into my previous message: those who wish to break the
standards (since Mozilla, Outlook etc. do so), can use rfc2047_decode
themselves. but this won't be too useful unless there is a
rfc2047_decode which is as broken as those implementations, ie. ignoring
the requirement for whitespace surrounding the encoded words. not sure
how to handle that best.
--
Kjetil T.