From: Charles Lindsey (chl@clw.cs.man.ac.uk)
Date: Thu Jan 09 2003 - 05:13:52 CST
In <200301071852.h07IqaA16299@astro.cs.utk.edu> Keith Moore <moore@cs.utk.edu> writes:
>I don't dispute the observation. However, I do dispute the reason.
>In my experience the problem with RFC 2047 implementations is usually
>that implementations try to "decode" RFC 2047 for the purpose of
>internal representation and then to "encode" again for retransmission,
>rather than to "decode" RFC 2047 only for the purpose of display.
Yes, but the situation in which that occurs is when a Usenet article
arrives and the user generates a "followup". What happens is that the
user's agent generates a pro-forma of the new article, preloaded with
"Subject: Re: old subject" and "Joe Bloggs said" and the like. This is
presented to the user, in his favourite editor, and he hacks it arnound to
his liking and then sends it out.
Now it is inevitable that strings that were encoded in RFC 2047 will have
been decoded into the local charset, for how else can the user edit them
when composing his followup? And they may get re-encoded in RFC 2047 when
they are sent out again. A few cycles of that, and all resemblance to the
original RFC 2047 will have been lost.
-- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Tel: +44 161 436 6131 Fax: +44 161 436 6133 Web: http://www.cs.man.ac.uk/~chl Email: chl@clw.cs.man.ac.uk Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K. PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5