[Ifile-discuss] html tag stripping

[David Bushong]

Hello, all.

I've noticed an increasing number of spam making it through ifile.  Some of 
them I can't think of a good answer for (the ones with nothing but a 
MIME-attached image and link); these are more the domain of something like
SpamAssassin.  However, there's another type that I think ifile could do
a very good job on.  Here's a sample from the body of a recent mail:

Yo<kc34sma21py2>uve rea<khuyowp1wuizl>d about them in the 
P<ks4nj3w258mkq1>apers....

(If you're reading this list in HTML, try turning it off).  Basically, this
completely ruins ifile's effectiveness.  However a simple addition to the
word tokenizer to skip anything between matched <>'s would completely avoid
this problem (as well as stop making "font", "color", etc. my most popular
words, spam or otherwise).

Of course this should be optional, but I think it would be a great addition,
as adding a preprocessor to handle this seems like a horrible duplication of
effort as ifile's already tokenizing...

Thoughts?

--David Bushong