How much effort (man months?) did it take to implement a MIME parser?
I integrated the MIME (based on the spec from Atlanta IETF) parser
(read-only) into Sun's Mailtool (User Agent) in about 2 weeks.
Then I spent 2 more weeks to convert Sun's propriertary Mailtool
attachment format to and from MIME format, but kept the UI and behavior
identical.
Finally, I spent another 2 weeks to allow Mailtool to handle the conversion
between Multipart message to and from non-Multipart message.
By the Santa Fe IETF meeting, I had a working version of Mailtool that could
send/receive MIME message and converted back/forth to Sun's propriertary
attachment format.
The parser must produce a data structure within which the body parts
are attached. Must be able to produce a new instance of the message
(or any body part) from the data structure.
My parser produced a generalized structure for MIME and Sun's attachment
format. It used Sun's Classing Engine to map the content-type to the
proper message viewing program.
There must be no policy decisions in the parser on the order of presenting
the body parts; that is decided by other code.
The parser didn't impose any policy on the body parts. But the Mailtool
(UA) did because I tried to make the seamless integration of MIME into the
Mailtool. The Mailtool treated the first text body part differently.
Don't include RichText implementation ..
No, I didn't implement RichText.
-Vincent