FP Avoidance tuning; add some test cases to analyze __BIGNUM_EMAILS against the masscheck corpora
git-svn-id: https://svn.apache.org/repos/asf/spamassassin/trunk@1887000 13f79535-47bb-0310-9956-ffa450edef68
diff --git a/rulesrc/sandbox/jhardin/20_misc_testing.cf b/rulesrc/sandbox/jhardin/20_misc_testing.cf
index 3593354..3e5e81f 100644
--- a/rulesrc/sandbox/jhardin/20_misc_testing.cf
+++ b/rulesrc/sandbox/jhardin/20_misc_testing.cf
@@ -715,6 +715,10 @@
#body LOTSA_EMAILS /\b(?:thousand|million)\se-?mail(?:\saddresse)?s?\b/i
#tflags LOTSA_EMAILS nopublish
+body __BIGNUM_EMAIL_TST1 /\b(?:thousand|million|\d[,1-9]{0,6}(?:[,0]+k?|k))\s(?:(?!and|or|your|place|baby)\w+\s)?(?:e-?mail)\b/i
+body __BIGNUM_EMAIL_TST2 /\b(?:thousand|million|\d[,1-9]{0,6}(?:[,0]+k?|k))\s(?:(?!and|or|your|place|baby)\w+\s)?(?:e-?mails)\b/i
+body __BIGNUM_EMAIL_TST3 /\b(?:thousand|million|\d[,1-9]{0,6}(?:[,0]+k?|k))\s(?:(?!and|or|your|place|baby)\w+\s)?(?:e-?mail:)/i
+body __BIGNUM_EMAIL_TST4 /\b(?:thousand|million|\d[,1-9]{0,6}(?:[,0]+k?|k))\s(?:(?!and|or|your|place|baby)\w+\s)?(?:e-?mail(?!:)s?)\b/i
body __BIGNUM_EMAILS /\b(?:thousand|million|\d[,1-9]{0,6}(?:[,0]+k?|k))\s(?:(?!and|or|your|place|baby)\w+\s)?(?:e-?mail(?:s|\saddresses)|fax numbers|leads|names)\b/i
tflags __BIGNUM_EMAILS multiple maxhits=5
meta BIGNUM_EMAILS __BIGNUM_EMAILS && !BIGNUM_EMAILS_MANY && !__BIGNUM_EMAILS_FREEM && !__RDNS_HEX && !MIME_QP_LONG_LINE
@@ -2757,7 +2761,7 @@
header __HAS_PHP_SCRIPT exists:X-PHP-Script
header __HAS_PHP_ORIG_SCRIPT exists:X-PHP-Originating-Script
-meta XM_RECPTID __HAS_XM_RECPTID && !__TAG_EXISTS_SCRIPT && !__REPLYTO_NOREPLY && !__ENVFROM_AMAZONSES
+meta XM_RECPTID __HAS_XM_RECPTID && !__TAG_EXISTS_SCRIPT && !__REPLYTO_NOREPLY && !__ENVFROM_AMAZONSES && !__DOS_DIRECT_TO_MX && !__FRAUD_PTX
describe XM_RECPTID Has spammy message header
score XM_RECPTID 3.000 # limit