bogofilter.htmlin

一个C语言写的快速贝叶斯垃圾邮件过滤工具

      robs and robx values.  With three values, min-dev, robs, and
      robx are all set.  If fewer values are given, parameters are set
      as described in the note below.</p><p>The <code class="option">-o
      <em class="replaceable"><code>value[<span class="optional">,value</span>]</code></em></code>
      option allows setting the spam-cutoff ham-cutoff values.  With
      two values, both spam-cutoff and ham-cutoff are set.  If only
      one value is given, parameters are set as described in the note
      below.</p><p>Note: All of these options allow fewer values to be provided.
     Values can be skipped by using just the comma delimiter, in which
     case the corresponding parameter(s) won't be changed.  If only
     the first value is provided, then only the first parameter is
     set.  Trailing values can be skipped, in which case the
     corresponding parameters won't be changed.  Within the parameter
     list, spaces are not allowed after commas.</p><p>INFO OPTIONS</p><p>The <code class="option">-v</code> option produces a report to standard
output on <span class="application">bogofilter</span>'s analysis of the
input.  Each additional <code class="option">v</code> will increase the verbosity of
the output, up to a maximum of 4.  With <code class="option">-vv</code>, the report
lists the tokens with highest deviation from a mean of 0.5 association
with spam.</p><p>Option <code class="option">-y date</code> can be used to override the
current date when timestamping tokens.  A value of zero (0) turns off
timestamping.</p><p>The <code class="option">-D</code> option redirects debug output to
stdout.</p><p>The <code class="option">-x <em class="replaceable"><code>flags</code></em></code> option
allows setting of debug flags for printing debug information.  See
header file debug.h for the list of usable flags.</p><p>CONFIG FILE OPTIONS</p><p>Using GNU longopt <code class="option">--</code> syntax, a config file's
<code class="option"><em class="replaceable"><code>name=value</code></em></code> statement
becomes a command line's
<code class="option">--<em class="replaceable"><code>option=value</code></em></code>.  Use
command <span><strong class="command">bogofilter --help</strong></span> for a list of options and
see <code class="filename">bogofilter.cf.example</code> for more info on
them. For example to change the X-Bogosity header to "X-Spam-Header", use:</p><p>
<code class="option"><em class="replaceable"><code>--spam-header-name=X-Spam-Header</code></em></code>
</p></div><div class="refsect1" lang="en"><a name="environment"></a><h2>ENVIRONMENT</h2><p><span class="application">Bogofilter</span> uses a database directory,
which can be set in the config file.  If not set there,
<span class="application">bogofilter</span> will use the value of
<code class="envar">BOGOFILTER_DIR</code>.  Both can be overridden by the
<code class="option">-d <em class="replaceable"><code>dir</code></em></code> option.  If none of
that is available, <span class="application">bogofilter</span> will use directory
<code class="filename">$HOME/.bogofilter</code>.</p></div><div class="refsect1" lang="en"><a name="configuration"></a><h2>CONFIGURATION</h2><p>The <span class="application">bogofilter</span> command line allows
setting of many options that determine how
<span class="application">bogofilter</span> operates.  File
<code class="filename">@sysconfdir@/bogofilter.cf</code> can be used to set
additional parameters that affect its operation.  File
<code class="filename">@sysconfdir@/bogofilter.cf.example</code> has samples of
all of the parameters.  Status and logging messages can be customized
for each site.</p></div><div class="refsect1" lang="en"><a name="returns"></a><h2>RETURN VALUES</h2><p>0 for spam; 1 for non-spam; 2 for unsure ; 3 for I/O or other errors.</p><p>If both <code class="option">-p</code> and <code class="option">-e</code> are used, the
    return values are: 0 for spam or non-spam; 3 for I/O or other
    errors.</p><p>Error 3 usually means that the wordlist file
<span class="application">bogofilter</span> wants to read at startup
is missing or the hard disk has filled up in <code class="option">-p</code> mode.</p></div><div class="refsect1" lang="en"><a name="integration"></a><h2>INTEGRATION WITH OTHER TOOLS</h2><p>Use with procmail</p><p>The following recipe (a) spam-bins anything that
<span class="application">bogofilter</span> rates as spam, (b) registers the
words in messages rated as spam as such, and (c) registers the
words in messages rated as non-spam as such.  With
this in place, it will normally only be necessary for the user to
intervene (with <code class="option">-Ns</code> or <code class="option">-Sn</code>) when
<span class="application">bogofilter</span> miscategorizes something.</p><pre class="programlisting">
<em class="lineannotation"><span class="lineannotation">
# filter mail through bogofilter, tagging it as Ham, Spam, or Unsure,
# and updating the wordlist
</span></em>
:0fw
| bogofilter -u -e -p

<em class="lineannotation"><span class="lineannotation">
# if bogofilter failed, return the mail to the queue;
# the MTA will retry to deliver it later
# 75 is the value for EX_TEMPFAIL in /usr/include/sysexits.h
</span></em>
:0e
{ EXITCODE=75 HOST }

<em class="lineannotation"><span class="lineannotation">
# file the mail to <code class="filename">spam-bogofilter</code> if it's spam.
</span></em>
:0:
* ^X-Bogosity: Spam, tests=bogofilter
spam-bogofilter
<em class="lineannotation"><span class="lineannotation">
# file the mail to <code class="filename">unsure-bogofilter</code> 
# if it's neither ham nor spam.
</span></em>
:0:
* ^X-Bogosity: Unsure, tests=bogofilter
unsure-bogofilter

# With this recipe, you can train bogofilter starting with an empty
# wordlist.  Be sure to check your unsure-folder regularly, take the
# messages out of it, classify them as ham (or spam), and use them to
# train bogofilter.

</pre><p>The following procmail rule will take mail on stdin and save
it to file <code class="filename">spam</code> if
<span class="application">bogofilter</span> thinks it's spam:
</p><pre class="programlisting">:0HB:
* ? bogofilter
spam
</pre><p>

and this similar rule will also register the tokens in the mail
according to the <span class="application">bogofilter</span> classification:

</p><pre class="programlisting">:0HB:
* ? bogofilter -u
spam
</pre><p>
</p><p>If <span class="application">bogofilter</span> fails (returning 3) the
message will be treated as non-spam.</p><p>This one is for <span class="application">maildrop</span>, it
    automatically defers the mail and retries later when the
    <span class="application">xfilter</span> command fails, use this in your
    <code class="filename">~/.mailfilter</code>:</p><pre class="programlisting">xfilter "bogofilter -u -e -p"
if (/^X-Bogosity: Spam, tests=bogofilter/)
{
  to "spam-bogofilter"
}
</pre><p>The following <code class="filename">.muttrc</code> lines will create
mutt macros for dispatching mail to
<span class="application">bogofilter</span>.</p><pre class="programlisting">macro index d "&lt;enter-command&gt;unset wait_key\n\
&lt;pipe-entry&gt;bogofilter -n\n\
&lt;enter-command&gt;set wait_key\n\
&lt;delete-message&gt;" "delete message as non-spam"
macro index \ed "&lt;enter-command&gt;unset wait_key\n\
&lt;pipe-entry&gt;bogofilter -s\n\
&lt;enter-command&gt;set wait_key\n\
&lt;delete-message&gt;" "delete message as spam"
</pre><p>Integration with Mail Transport Agent (MTA)</p><div class="procedure"><ol type="1"><li><p><span class="application">bogofilter</span> can also be integrated into an MTA
  to filter all incoming mail.  While the specific implementation is MTA dependent, the
  general steps are as follows:</p></li><li><p>Install <span class="application">bogofilter</span> on the mail server</p></li><li><p>Prime the <span class="application">bogofilter</span> databases with a spam
and non-spam corpus.  Since <span class="application">bogofilter</span> will be
serving a larger community, it is important to prime it with a representative
set of messages.</p></li><li><p>Set up the MTA to invoke <span class="application">bogofilter</span> on each
   message.  While this is an MTA specific step, you'll probably need to use the
   <code class="option">-p</code>, <code class="option">-u</code>, and <code class="option">-e</code> options.</p></li><li><p>Set up a mechanism for users to register spam/non-spam messages,
 as well as to correct mis-classifications.
 The most generic solution is to set up alias email addresses to which
 users bounce messages.</p></li><li><p>See the <code class="filename">doc</code> and <code class="filename">contrib</code> directories
   for more information.</p></li></ol></div><p>Use of R to verify <span class="application">bogofilter</span>'s
calculations</p><p>The -R option tells <span class="application">bogofilter</span> to
generate an R data frame.  The data frame contains one row per token
analyzed.  Each such row contains the token, the sum of its database
"good" and "spam" counts, the "good" count divided by the number of
non-spam messages used to create the training database, the "spam"
count divided by the spam message count, Robinson's f(w) for the token,
the natural logs of (1 - f(w)) and f(w), and an indicator character (+
if the token's f(w) value exceeded the minimum deviation from 0.5, - if
it didn't).  There is one additional row at the end of the table that
contains a label in the token field, followed by the number of words
actually used (the ones with + indicators), Robinson's P, Q, S, s and x
values and the minimum deviation.</p><p>The R data frame can be saved to a file and later read
into an R session (see <a href="http://cran.r-project.org/" target="_top">the R
project website</a> for information about the mathematics package
R).  Provided with the <span class="application">bogofilter</span>
distribution is a simple R script (file bogo.R) that can be used to
verify <span class="application">bogofilter</span>'s calculations.
Instructions for its use are included in the script in the form of
comments.</p></div><div class="refsect1" lang="en"><a name="logmessages"></a><h2>Log messages</h2><p><span class="application">Bogofilter</span> writes messages to the system log
    when the <code class="option">-l</code> option is used.  What is written depends on
    which other flags are used.</p><p>A classification run will generate (we are not showing the date
    and host part here):
    </p><pre class="screen">
bogofilter[1412]: X-Bogosity: Ham, spamicity=0.000227
bogofilter[1415]: X-Bogosity: Spam, spamicity=0.998918
</pre><p>Using <code class="option">-u</code> to classify a message and update a wordlist
    will produce (one a single line):
    </p><pre class="screen">
bogofilter[1426]: X-Bogosity: Spam, spamicity=0.998918,
  register -s, 329 words, 1 messages
    </pre><p>Registering words (<code class="option">-l</code> and <code class="option">-s</code>,
    <code class="option">-n</code>, <code class="option">-S</code>, or <code class="option">-N</code>) will produce:
    </p><pre class="screen">
bogofilter[1440]: register-n, 255 words, 1 messages
</pre><p>A registration run (using <code class="option">-s</code>, <code class="option">-n</code>,
    <code class="option">-N</code>, or <code class="option">-S</code>) will generate messages like:
</p><pre class="screen">
bogofilter[17330]: register-n, 574 words, 3 messages
bogofilter[6244]: register-s, 1273 words, 4 messages
</pre></div><div class="refsect1" lang="en"><a name="files"></a><h2>FILES</h2><div class="variablelist"><dl><dt><span class="term"><code class="filename">@sysconfdir@/bogofilter.cf</code></span></dt><dd><p>System configuration file.</p></dd><dt><span class="term"><code class="filename">~/.bogofilter.cf</code></span></dt><dd><p>User configuration file.</p></dd><dt><span class="term"><code class="filename">~/.bogofilter/wordlist.db</code></span></dt><dd><p>Combined list of good and spam tokens.</p></dd></dl></div></div><div class="refsect1" lang="en"><a name="author"></a><h2>AUTHOR</h2><div class="literallayout"><p><br>
Eric&nbsp;S.&nbsp;Raymond&nbsp;<code class="email">&lt;<a href="mailto:[e-mail]">[e-mail]</a>&gt;</code>.<br>
David&nbsp;Relson&nbsp;<code class="email">&lt;<a href="mailto:[e-mail]">[e-mail]</a>&gt;</code>.<br>
Matthias&nbsp;Andree&nbsp;<code class="email">&lt;<a href="mailto:[e-mail]">[e-mail]</a>&gt;</code>.<br>
Greg&nbsp;Louis&nbsp;<code class="email">&lt;<a href="mailto:[e-mail]">[e-mail]</a>&gt;</code>.<br>
</p></div><p>For updates, see the <a href="http://bogofilter.sourceforge.net/" target="_top">
    bogofilter project page</a>.</p></div><div class="refsect1" lang="en"><a name="also"></a><h2>SEE ALSO </h2><p>bogolexer(1), bogotune(1), bogoupgrade(1), bogoutil(1)</p></div></div></body></html>