|
Hate to say this, but the FWproxy withing Freeweb
is immune to this problem
:)
----- Original Message -----
Sent: Wednesday, May 30, 2001 23:21
Subject: [freenet-devl] Content filter vs
unicode
The attached file (recently posted to bugtraq by "eDvice
Security Services" <[EMAIL PROTECTED]> as an
exploit to another filter proxy) is picked up by
fproxy's filter, but only
the meta tag that sets the character set. This is used
by, for example, the
Freenet China News sites, though they use charset=gb2312
(and have
hyperlinks; they are picked up on the content filter on both counts).
Point
is, any foreign language web page, even without hyperlinks, will trip
the
filter. Therefore foreign language freesite readers will turn off the
content
filter. Isn't there some java support for this stuff? (finding
which chars could
correspond to an "<script" ?).
--
Always
hardwire the explosives
-- Fiona
Dexter quoting Monkey, J. Gregory Keyes, Dark
Genesis
|
