Fast Anti-Spam Filtering by Bayesian Statistical Analysis

Edit Package bogofilter

Bogofilter is a Bayesian spam filter. In normal operation, it takes an
e-mail message or text from standard input, does a statistical check
against lists of "good" and "bad" words, and returns a status code
indicating whether the message is spam. Bogofilter is designed with
fast algorithms (including the Berkeley DB system), coded directly in
C, and tuned for speed, so it can be used for production by sites that
process a lot of mail.

Refresh
Refresh
Source Files
Filename Size Changed
bogofilter-1.2.4.tar.bz2 0000887848 867 KB
bogofilter.changes 0000019370 18.9 KB
bogofilter.spec 0000003233 3.16 KB
Revision 22 (latest revision is 35)
Stephan Kulow's avatar Stephan Kulow (coolo) accepted request 184123 from Lars Müller's avatar Lars Müller (lmuelle) (revision 22)
- Update to version 1.2.4.
  * Fix three crashes in command line and environment variable parsers
    that caused NULL pointer dereferences with long option variants
    of bogofilter --syslog-tag, or bogoutil --timestamp-date, or when
    bogotune -M<file> cannot derive the bogofilter directory.
  * Add getopt_long_chk(), a getopt_long variant that checks if the
    overlapping short and long options agree on whether their argument
    is not required, mandatory, or optional.  If they disagree, the
    program aborts.
  * Fix a crash in command line parser that causes a NULL pointer
    dereference when --db-cachesize is used without argument.
    Found with getopt_long_chk().
  * Change lexer API/ABI a bit so as to work with flex 2.5.36 generated
    lexers (for instance, on Fedora 18 "Spherical Cow") that flip the
    type of yyleng from int to size_t. We use a signed long internally.
  * Fix XML form of Bulgarian FAQ so that it validates;
    and validate XHTML at build time.
  * Mark Berkeley DB 5.2.42 and 5.3.21 supported.
Comments 0
openSUSE Build Service is sponsored by