Re: [Bug: 21.5-b33] assert failed in rep_bytes_by_first_byte_1 -- bogus unicode?
Henry S. Thompson writes:
Steve Youngs <steve(a)sxemacs.org> writes:
> I stopped seeing this sort of crash in Gnus about 10 years ago
when I
> added the following rule to my ~/.procmailrc...
Thanks, but I have legitimate Chinese correspondents, not all known to
me in advance, so I really can't do that . . .
Actually you probably can if they're using a conformant MUA (this is
*not* something you can bet on with Japanese or Koreans, I admit, but
the Chinese think they're smarter, right?) Spammers tend to violate
the RFCs on purpose because Microsoft MUAs generally try to do
*something* with bogus input, and that something is often exploitable
(or at least historical examples are many :-). Unfortunately it looks
like XEmacs is explodable with the same technique.
However, I'm already pre-processing some header lines -- is it
obvious
what the bogosity was in that string?
Yes. It was an illegal internal code. The fact that it's internal
means it got past the coding system, and somehow into a buffer or
string. (I'd say I have no idea how, but with Gnus, nobody knows how
it works anyway.)
If you have the header as it came off the wire I could say more, perhaps.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://lists.xemacs.org/mailman/listinfo/xemacs-beta