You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@spamassassin.apache.org by fe...@apache.org on 2006/12/15 22:10:19 UTC

svn commit: r487662 - in /spamassassin/rules/branches/3.1: 20_uri_tests.cf 30_text_de.cf 30_text_fr.cf 30_text_nl.cf 30_text_pl.cf 50_scores.cf

Author: felicity
Date: Fri Dec 15 13:10:18 2006
New Revision: 487662

URL: http://svn.apache.org/viewvc?view=rev&rev=487662
Log:
bug 5246: drop BIZ_TLD and INFO_TLD rules due to bad hit rates

Modified:
    spamassassin/rules/branches/3.1/20_uri_tests.cf
    spamassassin/rules/branches/3.1/30_text_de.cf
    spamassassin/rules/branches/3.1/30_text_fr.cf
    spamassassin/rules/branches/3.1/30_text_nl.cf
    spamassassin/rules/branches/3.1/30_text_pl.cf
    spamassassin/rules/branches/3.1/50_scores.cf

Modified: spamassassin/rules/branches/3.1/20_uri_tests.cf
URL: http://svn.apache.org/viewvc/spamassassin/rules/branches/3.1/20_uri_tests.cf?view=diff&rev=487662&r1=487661&r2=487662
==============================================================================
--- spamassassin/rules/branches/3.1/20_uri_tests.cf (original)
+++ spamassassin/rules/branches/3.1/20_uri_tests.cf Fri Dec 15 13:10:18 2006
@@ -118,13 +118,6 @@
 uri BARGAIN_URL			/bargain([sz]|-\S+)?\.(?:com|biz)/
 describe BARGAIN_URL		Includes a link to a likely spammer domain
 
-# these are somewhat loose, but results are good
-uri BIZ_TLD  			/\.biz(?::\d+)?(?:\/|$)/i
-describe BIZ_TLD		Contains an URL in the BIZ top-level domain    
-
-uri INFO_TLD  			/\.info(?::\d+)?(?:\/|$)/i
-describe INFO_TLD		Contains an URL in the INFO top-level domain    
-
 # Matt Cline
 # Pretty good for most folks, except for jm: I have a really stupid
 # e-commerce bunch obfuscating their URLs with this for some reason. screw 'em

Modified: spamassassin/rules/branches/3.1/30_text_de.cf
URL: http://svn.apache.org/viewvc/spamassassin/rules/branches/3.1/30_text_de.cf?view=diff&rev=487662&r1=487661&r2=487662
==============================================================================
--- spamassassin/rules/branches/3.1/30_text_de.cf (original)
+++ spamassassin/rules/branches/3.1/30_text_de.cf Fri Dec 15 13:10:18 2006
@@ -564,8 +564,6 @@
 lang de describe USERPASS Hyperlink enthält Benutzername und (eventuell) ein Kennwort
 lang de describe URI_IS_POUND Dateiname besteht nur aus "#"; vielleicht ein JavaScript-Trick
 lang de describe BARGAIN_URL Spamverdächtiger Hyperlink
-lang de describe BIZ_TLD Hyperlink mit ".biz"-Domain
-lang de describe INFO_TLD Hyperlink mit ".info"-Domain
 lang de describe YAHOO_RD_REDIR URL mit Umleitung über Yahoo
 lang de describe YAHOO_DRS_REDIR URL mit Umleitung über Yahoo
 lang de describe URI_OFFERS Hyperlink zu Firmenangebot

Modified: spamassassin/rules/branches/3.1/30_text_fr.cf
URL: http://svn.apache.org/viewvc/spamassassin/rules/branches/3.1/30_text_fr.cf?view=diff&rev=487662&r1=487661&r2=487662
==============================================================================
--- spamassassin/rules/branches/3.1/30_text_fr.cf (original)
+++ spamassassin/rules/branches/3.1/30_text_fr.cf Fri Dec 15 13:10:18 2006
@@ -98,7 +98,6 @@
 lang fr describe BE_BOSS		Contient "Be your own boss" (soyez votre propre patron)
 lang fr describe BILLION_DOLLARS        Evoque des millions ou milliards de dollars
 lang fr describe BILL_1618              Revendique la conformité avec l'arrêt 1816 du sénat US
-lang fr describe BIZ_TLD		Contient une URL dans un domaine en .biz
 lang fr describe BLANK_LINES_70_80	Le corps du message a 70 à 80% de lignes vides
 lang fr describe BLANK_LINES_80_90	Le corps du message a 80 à 90% de lignes vides
 lang fr describe BLANK_LINES_90_100	Le corps du message a 90 à 100% de lignes vides

Modified: spamassassin/rules/branches/3.1/30_text_nl.cf
URL: http://svn.apache.org/viewvc/spamassassin/rules/branches/3.1/30_text_nl.cf?view=diff&rev=487662&r1=487661&r2=487662
==============================================================================
--- spamassassin/rules/branches/3.1/30_text_nl.cf (original)
+++ spamassassin/rules/branches/3.1/30_text_nl.cf Fri Dec 15 13:10:18 2006
@@ -408,7 +408,6 @@
 lang nl describe USERPASS                        URL bevat een gebruikersnaam en (evt ook) een wachtwoord
 lang nl describe URI_IS_POUND                    Bestandsnaam bevat alleen '\#'; duidt op een javascript geintje
 lang nl describe BARGAIN_URL                     Bevat een link naar een spam-domein
-lang nl describe BIZ_TLD                         Bevat een URL in het BIZ top-level domein    
 lang nl describe YAHOO_RD_REDIR                  Bevat een Yahoo Redirect URI
 lang nl describe YAHOO_DRS_REDIR                 Bevat een Yahoo Redirect URI
 lang nl describe URI_OFFERS                      Bevat een link naar aanbiedingen van een bedrijf

Modified: spamassassin/rules/branches/3.1/30_text_pl.cf
URL: http://svn.apache.org/viewvc/spamassassin/rules/branches/3.1/30_text_pl.cf?view=diff&rev=487662&r1=487661&r2=487662
==============================================================================
--- spamassassin/rules/branches/3.1/30_text_pl.cf (original)
+++ spamassassin/rules/branches/3.1/30_text_pl.cf Fri Dec 15 13:10:18 2006
@@ -84,7 +84,6 @@
 lang pl describe BEST_PORN		Prawdopodobnie porno - 'Najlepsze, Najwiêksze, Najwiêcej Porno'
 lang pl describe BILL_1618		Twierdzi zgodno¶æ z poprawk± Ustawy Senackiej 1618
 lang pl describe BILLION_DOLLARS	Tre¶æ: o niesamowitej ilo¶ci pieniêdzy
-lang pl describe BIZ_TLD		Zawiera URL w domenie Top-Level typu .biz
 lang pl describe BLANK_LINES_70_80	Tre¶æ zawiera 70-80% pustych linii
 lang pl describe BLANK_LINES_80_90	Tre¶æ zawiera 80-90% pustych linii
 lang pl describe BLANK_LINES_90_100	Tre¶æ zawiera 90-100% pustych linii

Modified: spamassassin/rules/branches/3.1/50_scores.cf
URL: http://svn.apache.org/viewvc/spamassassin/rules/branches/3.1/50_scores.cf?view=diff&rev=487662&r1=487661&r2=487662
==============================================================================
--- spamassassin/rules/branches/3.1/50_scores.cf (original)
+++ spamassassin/rules/branches/3.1/50_scores.cf Fri Dec 15 13:10:18 2006
@@ -57,7 +57,6 @@
 score BE_BOSS 1.268 1.271 1.480 1.539
 score BILLION_DOLLARS 0
 score BILL_1618 1.440 1.405 1.665 1.800
-score BIZ_TLD 1.719 1.169 2.035 2.013
 score BLANK_LINES_70_80 1.499 1.236 1.757 1.805
 score BLANK_LINES_80_90 0.272 0.107 0.810 0
 score BLANK_LINES_90_100 0 # n=3
@@ -347,7 +346,6 @@
 score HTTP_ESCAPED_HOST 0 0 0.124 0
 score HTTP_EXCESSIVE_ESCAPES 1.329 1.146 1.145 1.572
 score IMPOTENCE 2.093 0.592 2.443 0.627
-score INFO_TLD 1.373 0.813 1.457 1.273
 score INTERRUPTUS 1.154 0.533 1.106 0.182
 score INVALID_DATE 1.700 1.760 2.005 2.193
 score INVALID_DATE_TZ_ABSURD 1.360 1.346 1.573 1.700