# SARE HTML Ruleset for SpamAssassin - ruleset eng
# Version: 01.03.07
# Created: 2004-03-31 
# Modified: 2005-07-02
# Usage instructions, documentation, and change history in 70_sare_html0.cf 
#@@# Revision History:  Full Revision History stored in 70_sare_html.log
#@@# 01.03.07: July 2 2005
#@@#           Minor score tweaks based on recent mass-checks

# License: Artistic - see http://www.rulesemporium.com/license.txt 
# Current Maintainer: Bob Menschel - RMSA@Menschel.net
# Current Home: http://www.rulesemporium.com/rules/70_sare_html_eng.cf 
#
########  ######################   ##################################################

########  ######################   ##################################################
#   Use of accented HTML entities
########  ######################   ##################################################

# Accented HTML entities are generally not used in English-language emails.

# Note: OE will render these entities without the trailing ; so we must not check for this!
# Taken from: http://nwalsh.com/perl/dtdparse/html40/dtdent/

# & aacute; = á # & eacute; = é # & iacute; = í # & oacute; = ó # & uacute; = ú # & yacute; = ý
# & acirc;  = â # & ecirc;  = ê # & icirc;  = î # & ocirc;  = ô # & ucirc;  = û 
# & agrave; = à # & egrave; = è # & igrave; = ì # & ograve; = ò # & ugrave; = ù
# & atilde; = ã # & ntilde; = ñ # & otilde; = õ
# & auml;   = ä # & euml;   = ë # & iuml;   = ï # & ouml;   = ö # & uuml;   = ü # & yuml;   = ÿ
# & aring;  = å # & ccedil; = ç # & cent;   = ¢ # & aelig;  = æ # & micro;  = µ # & yen;    = ¥
# & iexcl;  = ¡ # & iquest; = ¿ # & oslash; = ø

rawbody   __SARE_HTML_ENT_ACUTE    /&[aeiouy]acute/i
rawbody   __SARE_HTML_ENT_CIRC     /&[aeiou]circ/i
rawbody   __SARE_HTML_ENT_GRAVE    /&[aeiou]grave/i
rawbody   __SARE_HTML_ENT_TILDE    /&[ano]tilde/i
rawbody   __SARE_HTML_ENT_UML      /&[aeiouy]uml/i
rawbody   __SARE_HTML_ENT_OTHER    /&(?:aring|ccedil|cent|aelig|micro|yen|iexcl|iquest|oslash)/i

meta      SARE_HTML_ENT_4          ((__SARE_HTML_ENT_ACUTE + __SARE_HTML_ENT_CIRC + __SARE_HTML_ENT_GRAVE + __SARE_HTML_ENT_TILDE + __SARE_HTML_ENT_UML + __SARE_HTML_ENT_OTHER) >= 4)
describe  SARE_HTML_ENT_4          email contains a number of diff types of accents
score     SARE_HTML_ENT_4          0.615
#counts   SARE_HTML_ENT_4          198s/6h of 269462 corpus (128310s/141152h RM) 06/17/05
#max      SARE_HTML_ENT_4          643s/0h of 84989 corpus (62394s/22595h RM) 06/07/04
#counts   SARE_HTML_ENT_4          35s/0h of 13446 corpus (11335s/2111h MY) 06/02/04
#counts   SARE_HTML_ENT_4          6s/16h of 54067 corpus (16890s/37177h JH-3.01) 06/18/05
#max      SARE_HTML_ENT_4          227s/11h of 38858 corpus (15368s/23490h JH-SA3.0rc1) 08/22/04
#counts   SARE_HTML_ENT_4          2s/0h of 47221 corpus (42968s/4253h MY) 06/18/05
#counts   SARE_HTML_ENT_4          0s/0h of 4676 corpus (808s/3868h ft) 05/28/05
#counts   SARE_HTML_ENT_4          5s/0h of 11260 corpus (6568s/4692h CT) 06/17/05

# EOF