Mailing List Archive

> I bet it depends on several factors, but is still there
> a reasonable method to always handle it intelligently,
> or is this is a hopelessly heuristic problem?

The rule is basically this ( from RFC's )

If the email is multipart ( ie, has MIME headers ), then the first text body ( textfile0 ) will contain the normal "This is a MIME encoded email, if you are reading this...", thus textfile1 will be the body.

If the email is strictly plain-text, then textfile0 will be the body.

I -could- put in a facility which will allow the body to be written out to a specifically named file (ie, --body_name=some_file_name ), because obviously the MIME decoding module will know if the email is multipart or not.


Regards.

--
Paul L Daniels http://www.pldaniels.com
Linux/Unix systems Internet Development
ICQ#103642862,AOL:cinflex,IRC:inflex
A.B.N. 19 500 721 806

> If the email is multipart ( ie, has MIME headers ), then the first text body ( textfile0 ) will contain the normal "This is a MIME
encoded email, if you are reading this...", thus textfile1 will be the body.
>
> If the email is strictly plain-text, then textfile0 will be the body.
>
> I -could- put in a facility which will allow the body to be written out to a specifically named file
(ie, --body_name=some_file_name ), because obviously the MIME decoding module will know if the email is multipart or not.


That would be a nice thing, but maybe also an overkill.
As to my understanding, a much simpler, solution is readily
available, by changing 1 single (wel,l OK, probably a
few more... ;) ) char:

It is already a very nice thing in Ripmime to call the
almost never used root part of mime messages "textfile0".

Now, simply declaring that non-multipart messages go to
"textfile1" (instead of textfile0), would magically align
unipart messages with the most frequent case of multipart
messages, where also "textfile1" is the actual mail message
text.

This would break compatibility somewhat (which is better
to so earlier then later), but would save some extra
work for both Ripmime and its clients once and for all.

Cheers,
Sab

> It is already a very nice thing in Ripmime to call the
> almost never used root part of mime messages "textfile0".
>
> Now, simply declaring that non-multipart messages go to
> "textfile1" (instead of textfile0), would magically align
> unipart messages with the most frequent case of multipart
> messages, where also "textfile1" is the actual mail message
> text.

Another quite handy property of this, just came up:

Currently, there is no easy way for us, clients, to
recognize if the mail was multipart or not.

Now, the presence or absence of "textfile0" would be
a clear indicator for the difference, so the client
can see the case easily.

(Some natural sugaring to this approch would be
perhaps not even calling it "textfile0", remembering
the changeable prefix and numbering, but e.g. simply
"mime" or anything alike.)

BTW, thanks for this VERY COOL piece of stuff being
made available for us!!

Sab