Author: quinlan
Date: Tue May 25 12:39:19 2004
New Revision: 20444
Modified:
incubator/spamassassin/trunk/rules/70_testing.cf
Log:
do some checking for T_HAS_TEMPLATE* rules since we're not catching all
the ham yet
remove T_SPECIAL_BY_CHOICE* because it sucks
add some more T_MANY_MX variants
Modified: incubator/spamassassin/trunk/rules/70_testing.cf
==============================================================================
--- incubator/spamassassin/trunk/rules/70_testing.cf (original)
+++ incubator/spamassassin/trunk/rules/70_testing.cf Tue May 25 12:39:19 2004
@@ -35,6 +35,9 @@
body T_SA_MARKED_UP /Spam detection software, running on the system /
# more general template rule to try
+# this seems to be evolving into two types of exclusions:
+# - freely changeable fields like Subject:, To:, Cc:, and From:
+# - fields that have binary garbage that may match
header __HEADER_TEMPLATE ALL =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
header __UIDL_TEMPLATE_LIKE X-UIDL =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
header __XFACE_TEMPLATE_LIKE X-Face =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
@@ -43,6 +46,10 @@
header __SUBJECT_TEMPLATE_LIKE Subject =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
meta T_HAS_TEMPLATE_0 __HEADER_TEMPLATE
meta T_HAS_TEMPLATE_1 (__HEADER_TEMPLATE && !__UIDL_TEMPLATE_LIKE && !__XFACE_TEMPLATE_LIKE && !__TOCC_TEMPLATE_LIKE && !__FROM_TEMPLATE_LIKE && !__SUBJECT_TEMPLATE_LIKE)
+# check Message-IDs sure and In-Reply-To for both Message-ID and name
+header __REF_TEMPLATE_LIKE References =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
+header __REP_TEMPLATE_LIKE In-Reply-To =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
+header __MID_TEMPLATE_LIKE MESSAGEID =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
########################################################################
@@ -129,19 +136,6 @@
# worth it?
body T_SAVE_BUCKS /\bsave \$/i
-# concept: you have a choice + something great
-body T_SPECIAL_BY_CHOICE_0 /\b(?:wish|want|question|unsubscrib\w+|discontinue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-body T_SPECIAL_BY_CHOICE_1 /\b(?:wish|want|question|unsub\w+|discontinue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-# A-F change "discontinue" to "continue"
-body T_SPECIAL_BY_CHOICE_A /\b(?:wish|want|question|unsubscrib\w+|continue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-body T_SPECIAL_BY_CHOICE_B /\b(?:wish|want|question|unsub\w+|continue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-# C-D drop "want"
-body T_SPECIAL_BY_CHOICE_C /\b(?:wish|question|unsubscrib\w+|continue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-body T_SPECIAL_BY_CHOICE_D /\b(?:wish|question|unsub\w+|continue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-# E-F drop "offer"
-body T_SPECIAL_BY_CHOICE_E /\b(?:wish|want|question|unsubscrib\w+|continue|remove).{0,64}\b(?:gift|special|promotion|coupon|discount|reward)/i
-body T_SPECIAL_BY_CHOICE_F /\b(?:wish|want|question|unsub\w+|continue|remove).{0,64}\b(?:gift|special|promotion|coupon|discount|reward)/i
-
# removed some old ones
header T_MANY_MX_5 From:addr =~ /\@mx\d{2,4}\..{1,20}\./i
header T_MANY_MX_7 From:addr =~ /\@\w\d+\..{1,30}\./i
@@ -159,6 +153,46 @@
header T_MANY_MX_19 From:addr =~ /\@\w{1,4}\d+\..*\./i
header T_MANY_MX_20 From:addr =~ /\@\w{0,5}\d+\..*\./i
header T_MANY_MX_21 From:addr =~ /\@\w{1,5}\d+\..*\./i
+#
+header T_MANY_MX_22 From:addr =~ /\@\w\d{2,}\..*\./i
+header T_MANY_MX_23 From:addr =~ /\@\w*\d{2,}\..*\./i
+header T_MANY_MX_24 From:addr =~ /\@\w+\d{2,}\..*\./i
+header T_MANY_MX_25 From:addr =~ /\@\w?\d{2,}\..*\./i
+header T_MANY_MX_26 From:addr =~ /\@\w{0,2}\d{2,}\..*\./i
+header T_MANY_MX_27 From:addr =~ /\@\w{1,2}\d{2,}\..*\./i
+header T_MANY_MX_28 From:addr =~ /\@\w{0,3}\d{2,}\..*\./i
+header T_MANY_MX_29 From:addr =~ /\@\w{1,3}\d{2,}\..*\./i
+header T_MANY_MX_30 From:addr =~ /\@\w{0,4}\d{2,}\..*\./i
+header T_MANY_MX_31 From:addr =~ /\@\w{1,4}\d{2,}\..*\./i
+header T_MANY_MX_32 From:addr =~ /\@\w{0,5}\d{2,}\..*\./i
+header T_MANY_MX_33 From:addr =~ /\@\w{1,5}\d{2,}\..*\./i
+#
+header T_MANY_MX_34 ALL =~ /\@\w\d+\..*\./i
+header T_MANY_MX_35 ALL =~ /\@\w*\d+\..*\./i
+header T_MANY_MX_36 ALL =~ /\@\w+\d+\..*\./i
+header T_MANY_MX_37 ALL =~ /\@\w?\d+\..*\./i
+header T_MANY_MX_38 ALL =~ /\@\w{0,2}\d+\..*\./i
+header T_MANY_MX_39 ALL =~ /\@\w{1,2}\d+\..*\./i
+header T_MANY_MX_40 ALL =~ /\@\w{0,3}\d+\..*\./i
+header T_MANY_MX_41 ALL =~ /\@\w{1,3}\d+\..*\./i
+header T_MANY_MX_42 ALL =~ /\@\w{0,4}\d+\..*\./i
+header T_MANY_MX_43 ALL =~ /\@\w{1,4}\d+\..*\./i
+header T_MANY_MX_44 ALL =~ /\@\w{0,5}\d+\..*\./i
+header T_MANY_MX_45 ALL =~ /\@\w{1,5}\d+\..*\./i
+#
+header T_MANY_MX_46 ALL =~ /\@\w\d{2,}\..*\./i
+header T_MANY_MX_47 ALL =~ /\@\w*\d{2,}\..*\./i
+header T_MANY_MX_48 ALL =~ /\@\w+\d{2,}\..*\./i
+header T_MANY_MX_49 ALL =~ /\@\w?\d{2,}\..*\./i
+header T_MANY_MX_50 ALL =~ /\@\w{0,2}\d{2,}\..*\./i
+header T_MANY_MX_51 ALL =~ /\@\w{1,2}\d{2,}\..*\./i
+header T_MANY_MX_52 ALL =~ /\@\w{0,3}\d{2,}\..*\./i
+header T_MANY_MX_53 ALL =~ /\@\w{1,3}\d{2,}\..*\./i
+header T_MANY_MX_54 ALL =~ /\@\w{0,4}\d{2,}\..*\./i
+header T_MANY_MX_55 ALL =~ /\@\w{1,4}\d{2,}\..*\./i
+header T_MANY_MX_56 ALL =~ /\@\w{0,5}\d{2,}\..*\./i
+header T_MANY_MX_57 ALL =~ /\@\w{1,5}\d{2,}\..*\./i
+
# dip at length of 5, so stop here
# Updated test rules for fraud, uses existing rules and updated rules
Date: Tue May 25 12:39:19 2004
New Revision: 20444
Modified:
incubator/spamassassin/trunk/rules/70_testing.cf
Log:
do some checking for T_HAS_TEMPLATE* rules since we're not catching all
the ham yet
remove T_SPECIAL_BY_CHOICE* because it sucks
add some more T_MANY_MX variants
Modified: incubator/spamassassin/trunk/rules/70_testing.cf
==============================================================================
--- incubator/spamassassin/trunk/rules/70_testing.cf (original)
+++ incubator/spamassassin/trunk/rules/70_testing.cf Tue May 25 12:39:19 2004
@@ -35,6 +35,9 @@
body T_SA_MARKED_UP /Spam detection software, running on the system /
# more general template rule to try
+# this seems to be evolving into two types of exclusions:
+# - freely changeable fields like Subject:, To:, Cc:, and From:
+# - fields that have binary garbage that may match
header __HEADER_TEMPLATE ALL =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
header __UIDL_TEMPLATE_LIKE X-UIDL =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
header __XFACE_TEMPLATE_LIKE X-Face =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
@@ -43,6 +46,10 @@
header __SUBJECT_TEMPLATE_LIKE Subject =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
meta T_HAS_TEMPLATE_0 __HEADER_TEMPLATE
meta T_HAS_TEMPLATE_1 (__HEADER_TEMPLATE && !__UIDL_TEMPLATE_LIKE && !__XFACE_TEMPLATE_LIKE && !__TOCC_TEMPLATE_LIKE && !__FROM_TEMPLATE_LIKE && !__SUBJECT_TEMPLATE_LIKE)
+# check Message-IDs sure and In-Reply-To for both Message-ID and name
+header __REF_TEMPLATE_LIKE References =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
+header __REP_TEMPLATE_LIKE In-Reply-To =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
+header __MID_TEMPLATE_LIKE MESSAGEID =~ /%[A-Z][A-Z_-]+[^%A-Za-z_-]/
########################################################################
@@ -129,19 +136,6 @@
# worth it?
body T_SAVE_BUCKS /\bsave \$/i
-# concept: you have a choice + something great
-body T_SPECIAL_BY_CHOICE_0 /\b(?:wish|want|question|unsubscrib\w+|discontinue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-body T_SPECIAL_BY_CHOICE_1 /\b(?:wish|want|question|unsub\w+|discontinue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-# A-F change "discontinue" to "continue"
-body T_SPECIAL_BY_CHOICE_A /\b(?:wish|want|question|unsubscrib\w+|continue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-body T_SPECIAL_BY_CHOICE_B /\b(?:wish|want|question|unsub\w+|continue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-# C-D drop "want"
-body T_SPECIAL_BY_CHOICE_C /\b(?:wish|question|unsubscrib\w+|continue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-body T_SPECIAL_BY_CHOICE_D /\b(?:wish|question|unsub\w+|continue|remove).{0,64}\b(?:gift|offer|special|promotion|coupon|discount|reward)/i
-# E-F drop "offer"
-body T_SPECIAL_BY_CHOICE_E /\b(?:wish|want|question|unsubscrib\w+|continue|remove).{0,64}\b(?:gift|special|promotion|coupon|discount|reward)/i
-body T_SPECIAL_BY_CHOICE_F /\b(?:wish|want|question|unsub\w+|continue|remove).{0,64}\b(?:gift|special|promotion|coupon|discount|reward)/i
-
# removed some old ones
header T_MANY_MX_5 From:addr =~ /\@mx\d{2,4}\..{1,20}\./i
header T_MANY_MX_7 From:addr =~ /\@\w\d+\..{1,30}\./i
@@ -159,6 +153,46 @@
header T_MANY_MX_19 From:addr =~ /\@\w{1,4}\d+\..*\./i
header T_MANY_MX_20 From:addr =~ /\@\w{0,5}\d+\..*\./i
header T_MANY_MX_21 From:addr =~ /\@\w{1,5}\d+\..*\./i
+#
+header T_MANY_MX_22 From:addr =~ /\@\w\d{2,}\..*\./i
+header T_MANY_MX_23 From:addr =~ /\@\w*\d{2,}\..*\./i
+header T_MANY_MX_24 From:addr =~ /\@\w+\d{2,}\..*\./i
+header T_MANY_MX_25 From:addr =~ /\@\w?\d{2,}\..*\./i
+header T_MANY_MX_26 From:addr =~ /\@\w{0,2}\d{2,}\..*\./i
+header T_MANY_MX_27 From:addr =~ /\@\w{1,2}\d{2,}\..*\./i
+header T_MANY_MX_28 From:addr =~ /\@\w{0,3}\d{2,}\..*\./i
+header T_MANY_MX_29 From:addr =~ /\@\w{1,3}\d{2,}\..*\./i
+header T_MANY_MX_30 From:addr =~ /\@\w{0,4}\d{2,}\..*\./i
+header T_MANY_MX_31 From:addr =~ /\@\w{1,4}\d{2,}\..*\./i
+header T_MANY_MX_32 From:addr =~ /\@\w{0,5}\d{2,}\..*\./i
+header T_MANY_MX_33 From:addr =~ /\@\w{1,5}\d{2,}\..*\./i
+#
+header T_MANY_MX_34 ALL =~ /\@\w\d+\..*\./i
+header T_MANY_MX_35 ALL =~ /\@\w*\d+\..*\./i
+header T_MANY_MX_36 ALL =~ /\@\w+\d+\..*\./i
+header T_MANY_MX_37 ALL =~ /\@\w?\d+\..*\./i
+header T_MANY_MX_38 ALL =~ /\@\w{0,2}\d+\..*\./i
+header T_MANY_MX_39 ALL =~ /\@\w{1,2}\d+\..*\./i
+header T_MANY_MX_40 ALL =~ /\@\w{0,3}\d+\..*\./i
+header T_MANY_MX_41 ALL =~ /\@\w{1,3}\d+\..*\./i
+header T_MANY_MX_42 ALL =~ /\@\w{0,4}\d+\..*\./i
+header T_MANY_MX_43 ALL =~ /\@\w{1,4}\d+\..*\./i
+header T_MANY_MX_44 ALL =~ /\@\w{0,5}\d+\..*\./i
+header T_MANY_MX_45 ALL =~ /\@\w{1,5}\d+\..*\./i
+#
+header T_MANY_MX_46 ALL =~ /\@\w\d{2,}\..*\./i
+header T_MANY_MX_47 ALL =~ /\@\w*\d{2,}\..*\./i
+header T_MANY_MX_48 ALL =~ /\@\w+\d{2,}\..*\./i
+header T_MANY_MX_49 ALL =~ /\@\w?\d{2,}\..*\./i
+header T_MANY_MX_50 ALL =~ /\@\w{0,2}\d{2,}\..*\./i
+header T_MANY_MX_51 ALL =~ /\@\w{1,2}\d{2,}\..*\./i
+header T_MANY_MX_52 ALL =~ /\@\w{0,3}\d{2,}\..*\./i
+header T_MANY_MX_53 ALL =~ /\@\w{1,3}\d{2,}\..*\./i
+header T_MANY_MX_54 ALL =~ /\@\w{0,4}\d{2,}\..*\./i
+header T_MANY_MX_55 ALL =~ /\@\w{1,4}\d{2,}\..*\./i
+header T_MANY_MX_56 ALL =~ /\@\w{0,5}\d{2,}\..*\./i
+header T_MANY_MX_57 ALL =~ /\@\w{1,5}\d{2,}\..*\./i
+
# dip at length of 5, so stop here
# Updated test rules for fraud, uses existing rules and updated rules