Re: [SLUG] spamassassin vs pine

From: Eben King (eben1@tampabay.rr.com)
Date: Sun Jan 15 2006 - 22:31:11 EST


On Sun, 15 Jan 2006, Mike Branda wrote:

> I'm going to do some looking into the BL/RBL thing and find out if the
> SuSE default setup comes with that stuff. Your analysis will hopefully
> guide you with any custom scoring if we don't come up with anything
> else...

Update:

I wrote a script (63 lines) to extract the raw counts for each test from two
groups of messages (I suggest "spam" and "not spam", but it's up to the
user.

No --help, no --usage, none of that good stuff. Haven't got around
to it yet. Synopsis:

status -g1 mbox mail/linux -g2 mail/probably-spam mail/almost-certainly-spam

Output looks like this:

2 0 ALL_TRUSTED
205 3 AWL
290 0 BAYES_00
4 0 BAYES_05
1 0 BAYES_20
3 0 BAYES_40
2 11 BAYES_50
0 2 BAYES_60
0 1 BAYES_80
1 11 BAYES_99
etc.

(For a sense of scale, that command took about 5 seconds on my 2 GHz
Athlon.)

The first column is how many times that test was mentioned in Group 1's
"X-Spam-Status" headers, and the second column goes with Group 2. "-g1" =
Group 1, "-g2"=Group 2. Don't know of a CLI spreadsheet, so for now I'm
taking the raw data and plugging it into a GUI one.

If anyone wants it, say so, & I'll post it on my web site.

--
  ebenONE@tampabay.ARE-ARE.com / A well-loved and correctly trained domestic
   canine is generally slobbery, excitable, noisy, scatalogically obsessed,
xenophobic, pathetically unjudgmental, embarrassingly uninhibited, unreasoningly
  devoted, heartbreakingly dependent and wretchedly craven. -- P. B. in AFCA

----------------------------------------------------------------------- This list is provided as an unmoderated internet service by Networked Knowledge Systems (NKS). Views and opinions expressed in messages posted are those of the author and do not necessarily reflect the official policy or position of NKS or any of its employees.



This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 16:33:51 EDT