sa-updates/20_uri_tests.cf

   1 # SpamAssassin rules file: URI tests
   2 #
   3 # Please don't modify this file as your changes will be overwritten with
   4 # the next update. Use /etc/mail/spamassassin/local.cf instead.
   5 # See 'perldoc Mail::SpamAssassin::Conf' for details.
   6 #
   7 # <@LICENSE>
   8 # Licensed to the Apache Software Foundation (ASF) under one or more
   9 # contributor license agreements.  See the NOTICE file distributed with
  10 # this work for additional information regarding copyright ownership.
  11 # The ASF licenses this file to you under the Apache License, Version 2.0
  12 # (the "License"); you may not use this file except in compliance with
  13 # the License.  You may obtain a copy of the License at:
  14 #
  15 #     http://www.apache.org/licenses/LICENSE-2.0
  16 #
  17 # Unless required by applicable law or agreed to in writing, software
  18 # distributed under the License is distributed on an "AS IS" BASIS,
  19 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  20 # See the License for the specific language governing permissions and
  21 # limitations under the License.
  22 # </@LICENSE>
  23 #
  24 ###########################################################################
  25
  26 require_version 3.004006
  27
  28 # possible IDN spoofing attack: https://web.archive.org/web/20141006091906/https://www.shmoo.com/idn/homograph.txt
  29 # not expecting any hits on this (yet)
  30 uri HIGH_CODEPAGE_URI          /^https?:\/\/[^\/]*\&\#(?:\d{4,}|[3456789]\d\d);/i
  31 tflags HIGH_CODEPAGE_URI       userconf
  32
  33 ###########################################################################
  34
  35 # Redirector URI patterns
  36 redirector_pattern      /^http:\/\/chkpt\.zdnet\.com\/chkpt\/\w+\/(.*)$/i
  37 redirector_pattern      /^http:\/\/www(?:\d+)?\.nate\.com\/r\/\w+\/(.*)$/i
  38 redirector_pattern      /^http:\/\/.+\.gov\/(?:.*\/)?externalLink\.jhtml\?.*url=(.*?)(?:&.*)?$/i
  39 redirector_pattern      /^http:\/\/redir\.internet\.com\/.+?\/.+?\/(.*)$/i
  40 redirector_pattern      /^http:\/\/(?:.*?\.)?adtech\.de\/.*(?:;|\|)link=(.*?)(?:;|$)/i
  41 redirector_pattern      m'^http.*?/redirect\.php\?.*(?<=[?&])goto=(.*?)(?:$|[&\#])'i
  42 redirector_pattern      m'^https?:/*(?:[^/]+\.)?emf\d\.com/r\.cfm.*?&r=(.*)'i
  43
  44 uri NUMERIC_HTTP_ADDR           m{^https?://[\d.]+(?:[:/?\#]|$)}i
  45 describe NUMERIC_HTTP_ADDR      Uses a numeric IP address in URL
  46
  47 # Theo sez:
  48 # Have gotten FPs off this, and whitespace can't be in the host, so...
  49 # %    Visit my homepage: http://i.like.foo.com %
  50 # Also ignore some bad parses like http://foo.bar%20http://foo.bar
  51 uri HTTP_ESCAPED_HOST           /^https?\:\/\/[^\/\s\?\&\#\']*(?!%(?:20|3[cCeE])(?:https?:|mailto:))%[0-9a-fA-F][0-9a-fA-F]/
  52 describe HTTP_ESCAPED_HOST      Uses %-escapes inside a URL's hostname
  53
  54 # look for URI with escaped 0-9, A-Z, or a-z characters (all other safe
  55 # characters have been well-tested, but are sometimes unnecessarily escaped
  56 # in nonspam; requiring "http" or "https" also reduces false positives).
  57 uri HTTP_EXCESSIVE_ESCAPES      /^https?:\/\/\S*%(?:3\d|[46][1-9a-f]|[57][\da])/i
  58 describe HTTP_EXCESSIVE_ESCAPES Completely unnecessary %-escapes inside a URL
  59
  60 # bug 1801
  61 uri IP_LINK_PLUS        m{^https?://\d+\.\d+\.\d+\.\d+.{0,20}(?:cgi|click|ads|id=)}i
  62 describe IP_LINK_PLUS   Dotted-decimal IP address followed by CGI
  63
  64 # allow ports 80 and 443 which are http and https, respectively
  65 # we don't want to hit http://www.cnn.com:USArticle1840@www.liquidshirts.com/
  66 # though, which actually doesn't have a weird port in it.
  67 uri WEIRD_PORT                  m{https?://[^/?\s]+?:\d+(?<!:80)(?<!:443)(?<!:8080)(?:/|\s|$)}
  68 describe WEIRD_PORT             Uses non-standard port number for HTTP
  69
  70 # Matt Cline
  71 # Pretty good for most folks, except for jm: I have a really stupid
  72 # e-commerce bunch obfuscating their URLs with this for some reason. screw 'em
  73 # jm: hesitant to remove this outright; it should be good against phishers
  74 #uri      HTTP_ENTITIES_HOST    m{https?://[^\s\">/]*\&\#[\da-f]+}i
  75 #describe HTTP_ENTITIES_HOST    URI obscured with character entities
  76
  77 uri YAHOO_RD_REDIR              m{^https?\://rd\.yahoo\.com/(?:[0-9]{4}|partner\b|dir\b)}i
  78 describe YAHOO_RD_REDIR         Has Yahoo Redirect URI
  79
  80 uri YAHOO_DRS_REDIR             m{^https?://drs\.yahoo\.com/}i
  81 describe YAHOO_DRS_REDIR        Has Yahoo Redirect URI
  82
  83 # "www" hidden as "%77%77%77", "ww%77", etc.
  84 # note: *not* anchored to start of string, to catch use of redirectors
  85 uri HTTP_77                     /http:\/\/.{0,2}\%77/
  86 describe HTTP_77                Contains an URL-encoded hostname (HTTP77)
  87
  88 # a.com.b.c
  89 uri SPOOF_COM2OTH       m{^https?://(?:\w+\.)+?com\.(?!(?:[a-z]{2}\.)?s3\.amazonaws\.com|\w+\.psmtp\.com)(?:\w+\.){2}}i
  90 describe SPOOF_COM2OTH  URI contains ".com" in middle
  91
  92 # a.com.b.com
  93 uri __SPOOF_COM2COM     m{^https?://(?:\w+\.)+?com\.(?!(?:[a-z]{2}\.)?s3\.amazonaws\.com|\w+\.psmtp\.com)(?:\w+\.)+?com\b}i
  94 meta SPOOF_COM2COM      __SPOOF_COM2COM && !SPOOF_COM2OTH
  95 describe SPOOF_COM2COM  URI contains ".com" in middle and end
  96
  97 # a.net.b.com
  98 uri SPOOF_NET2COM       m{^https?://(?:\w+\.)+?(?:net|org)\.(?!(?:[a-z]{2}\.)?s3\.amazonaws\.com)(?:\w+\.)+?com\b}i
  99 describe SPOOF_NET2COM  URI contains ".net" or ".org", then ".com"
 100
 101 uri URI_HEX             m%^https?://[^/?&\#]*\b(?![0-9a-f]{0,12}[a-f]{3})[0-9a-f]{6,}\b%i
 102 describe URI_HEX        URI hostname has long hexadecimal sequence
 103
 104 uri URI_NOVOWEL         m%^https?://[^/?&\#]*[bcdfgjklmnpqrstvwxz]{7}%i
 105 describe URI_NOVOWEL    URI hostname has long non-vowel sequence
 106 tflags URI_NOVOWEL      userconf     # lock scores low
 107
 108 uri URI_UNSUBSCRIBE             /\b(?:gone|opened|out)\.php/i
 109 describe URI_UNSUBSCRIBE        URI contains suspicious unsubscribe link
 110
 111
 112 # bug 3896: URIs in various TLDs, other than 3rd level www
 113 uri URI_NO_WWW_INFO_CGI /^(?:https?:\/\/)?[^\/]+(?<!\/www)\.[^.]{7,}\.info\/(?=\S{15,})\S*\?/i
 114 describe URI_NO_WWW_INFO_CGI    CGI in .info TLD other than third-level "www"
 115
 116 uri URI_NO_WWW_BIZ_CGI  /^(?:https?:\/\/)?[^\/]+(?<!\/www)\.[^.]{7,}\.biz\/(?=\S{15,})\S*\?/i
 117 describe URI_NO_WWW_BIZ_CGI     CGI in .biz TLD other than third-level "www"
 118
 119 ###########################################################################
 120
 121 uri NORMAL_HTTP_TO_IP          m{^https?://(?!1(?:0|27|69\.254|72\.(?:1[6-9]|2\d|3[01])|92\.168)\.)\d+\.\d+\.\d+\.\d+\b(?![.-])}i
 122 describe NORMAL_HTTP_TO_IP     URI host has a public dotted-decimal IPv4 address
 123