I noticed this in the changes:
sub ReadNorm {
# read some values from rebuildspamdb
my ( $fh ) = @_;
my $this = $Con{$fh};
return if -e "$base/normfile";
open(H,"<$base/normfile");
my $norm=split(" ",<H>); close H;
return if !$norm;
if ( $norm > 3) {
$this->{badnorm}=$norm;
mlog( 0, "Baysian check disabled, SpamDB has bad norm($norm)" );
}
}
I see where it says $norm > 3
Here is my last spamdb rebuild from this morning:
Mon Apr 28 06:00:00 2008: RebuildSpamDB 1.3.5(2.5) started
Do Not Collect RedRe Messages: Enabled
**Messages matching the RedRe will be removed from the corpus!**
Use Subject as Maillog Names: False
Maxbytes: 8000
Maxfiles: 14000
Mon Apr 28 06:00:01 2008: Analyzing c:\ASSP/errors/spam 48
Mon Apr 28 06:00:03 2008: Analyzing c:\ASSP/errors/notspam 95
Mon Apr 28 06:02:01 2008: Analyzing c:\ASSP/spam 12822
Mon Apr 28 06:10:16 2008: Analyzing c:\ASSP/notspam 13243
Found 1501573 spam words, 5622466 non-spam words.
Generating weighted keys...
norm=0.2671
norm should be between 0.5 and 1.5
more spam words needed
Mon Apr 28 06:10:22 2008: Saving rebuilt SPAM database
Mon Apr 28 06:10:23 2008: Analyzing ended
spam.db: 1150976 bytes
Mon Apr 28 06:10:23 2008: Cleaning whitelist
whitelist before: 2135
whitelist after: 2130
Mon Apr 28 06:10:23 2008: RebuildSpamDB 1.3.5(2.5) ended
total time processing=623 second(s)
Uploading Griplist via Direct Connection
submitted 40019 bytes
Might be a stupid question, but will my Bayesian check be disabled or is it
just for norms being above 3?
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
Assp-test mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/assp-test
