wb-2_10_x / branches / main / modules / output_filter / filters / filterEmail.php @ 5
1 |
<?php
|
---|---|
2 |
/**
|
3 |
* protect email addresses (replace '@' and '.' and obfuscate address
|
4 |
* @param string $content
|
5 |
* @return string
|
6 |
*/
|
7 |
function doFilterEmail($content) { |
8 |
$aFilterSettings = getOutputFilterSettings();
|
9 |
$key = preg_replace('=^.*?filter([^\.\/\\\\]+)(\.[^\.]+)?$=is', '\1', __FILE__); |
10 |
if ($aFilterSettings[$key]) { |
11 |
if (
|
12 |
$aFilterSettings['OutputFilterMode'] == 0 && |
13 |
$aFilterSettings['at_replacement'] == '@' && |
14 |
$aFilterSettings['dot_replacement'] == '.' |
15 |
) { |
16 |
return $content; |
17 |
} |
18 |
$bNeedMdcr = false; |
19 |
// test if js-decryption is available
|
20 |
$script = str_replace('\\', '/',str_replace(WB_PATH,'', dirname(__DIR__)).'/js/mdcr.js'); |
21 |
if(is_readable(WB_PATH.$script)) { |
22 |
$scriptLink = "\t".'<script src="'.WB_URL.$script.'" type="text/javascript"></script>'."\n"; |
23 |
$aFilterSettings['OutputFilterMode'] |= pow(2, 2); // n | 2^2 |
24 |
} |
25 |
/* *** obfuscate mailto addresses by js:mdcr *** */
|
26 |
// work out the defined output filter mode: possible output filter modes: [0], 1, 2, 3, 6, 7
|
27 |
// 2^0 * (0.. disable, 1.. enable) filtering of mail addresses in text
|
28 |
// 2^1 * (0.. disable, 1.. enable) filtering of mail addresses in mailto links
|
29 |
// 2^2 * (0.. disable, 1.. enable) Javascript mailto encryption (only if mailto filtering enabled)
|
30 |
|
31 |
// first search part to find all mailto email addresses
|
32 |
// $pattern = '#(<a[^<]*href\s*?=\s*?"\s*?mailto\s*?:\s*?)([A-Z0-9._%+-]+@(?:[A-Z0-9-]+\.)+[A-Z]{2,4})([^"]*?)"([^>]*>)(.*?)</a>';
|
33 |
// bugfix 20160417
|
34 |
$pattern = '#(<a[^<]*href\s*?=\s*?"\s*?mailto\s*?:\s*?)([A-Z0-9._%+-]+@(?:[A-Z0-9-]+\.)+[A-Z]{2,4})([^"]*?)"([^>]*>\s*)(.*?)</a>'; |
35 |
// second part to find all non mailto email addresses
|
36 |
$pattern .= '|(value\s*=\s*"|\')??\b([A-Z0-9._%+-]+@(?:[A-Z0-9-]+\.)+[A-Z]{2,4})\b#i'; |
37 |
/*
|
38 |
(<a[^<]*href\s*?=\s*?"\s*?mailto\s*?:\s*?)([A-Z0-9._%+-]+@(?:[A-Z0-9-]+\.)+[A-Z]{2,4})([^"]*?)"([^>]*>)(.*?)</a>|(value\s*=\s*"|\')??\b([A-Z0-9._%+-]+@(?:[A-Z0-9-]+\.)+[A-Z]{2,4})\b
|
39 |
Sub 1:\b(<a.[^<]*href\s*?=\s*?"\s*?mailto\s*?:\s*?) --> "<a id="yyy" class="xxx" href = " mailto :" ignoring white spaces
|
40 |
Sub 2:([A-Z0-9._%+-]+@(?:[A-Z0-9-]+\.)+[A-Z]{2,4}) --> the email address in the mailto: part of the mail link
|
41 |
Sub 3:([^"]*?)" --> possible ?Subject&cc... stuff attached to the mail address
|
42 |
Sub 4:([^>]*>) --> all class or id statements after the mailto but before closing ..>
|
43 |
Sub 5:(.*?)</a>\b --> the mailto text; all characters between >xxxxx</a>
|
44 |
Sub 6:|\b([A-Z0-9._%+-]+@(?:[A-Z0-9-]+\.)+[A-Z]{2,4})\b --> email addresses which may appear in the text (require word boundaries)
|
45 |
*/
|
46 |
// find all email addresses embedded in the content and filter them using a callback function
|
47 |
$content = preg_replace_callback( |
48 |
$pattern,
|
49 |
/* ************************************************************************** */
|
50 |
function ($match) use ($aFilterSettings, $bNeedMdcr) { |
51 |
// check if required arguments are defined
|
52 |
$search = array('@', '.'); |
53 |
$replace = array($aFilterSettings['at_replacement'] ,$aFilterSettings['dot_replacement']); |
54 |
// check if the match contains the expected number of subpatterns (6|8)
|
55 |
switch (count($match)) { |
56 |
case 8: |
57 |
/** OUTPUT FILTER FOR EMAIL ADDRESSES EMBEDDED IN TEXT **/
|
58 |
// 1.. text mails only, 3.. text mails + mailto (no JS), 7 text mails + mailto (JS)
|
59 |
if(!in_array($aFilterSettings['OutputFilterMode'], array(1,3,5,7))){ return $match[0]; } |
60 |
// do not filter mail addresses included in input tags (<input ... value = "test@mail)
|
61 |
if (strpos($match[6], 'value') !== false) return $match[0]; { |
62 |
// filtering of non mailto email addresses enabled
|
63 |
return str_replace($search, $replace, $match[0]); |
64 |
} |
65 |
break;
|
66 |
case 6: |
67 |
/** OUTPUT FILTER FOR EMAIL ADDRESSES EMBEDDED IN MAILTO LINKS **/
|
68 |
// 2.. mailto only (no JS), 3.. text mails + mailto (no JS), 6.. mailto only (JS), 7.. all filters active
|
69 |
if(!in_array($aFilterSettings['OutputFilterMode'], array(2,3,6,7))){ return $match[0]; } |
70 |
// check if last part of the a href link: >xxxx</a> contains a email address we need to filter
|
71 |
$pattern = '#[A-Z0-9._%+-]+@(?:[A-Z0-9-]+\.)+[A-Z]{2,4}#i'; |
72 |
if(preg_match_all($pattern, $match[5], $matches)) { |
73 |
foreach($matches as $submatch) { |
74 |
foreach($submatch as $value) { |
75 |
// replace all . and all @ in email address parts by (dot) and (at) strings
|
76 |
$match[5] = str_replace($value, str_replace($search, $replace, $value), $match[5]); |
77 |
} |
78 |
} |
79 |
} |
80 |
// check if Javascript encryption routine is enabled
|
81 |
if(in_array($aFilterSettings['OutputFilterMode'], array(6,7))) { |
82 |
/** USE JAVASCRIPT ENCRYPTION FOR MAILTO LINKS **/
|
83 |
// extract possible class and id attribute from ahref link
|
84 |
preg_match('/class\s*?=\s*?("|\')(.*?)\1/ix', $match[0], $class_attr); |
85 |
$class_attr = empty($class_attr) ? '' : 'class="' . $class_attr[2] . '" '; |
86 |
preg_match('/id\s*?=\s*?("|\')(.*?)\1/ix', $match[0], $id_attr); |
87 |
$id_attr = empty($id_attr) ? '' : 'id="' . $id_attr[2] . '" '; |
88 |
// preprocess mailto link parts for further usage
|
89 |
$search = array('@', '.', '_', '-'); $replace = array('F', 'Z', 'X', 'K'); |
90 |
$email_address = str_replace($search, $replace, strtolower($match[2])); |
91 |
$email_subject = rawurlencode(html_entity_decode($match[3])); |
92 |
// create a random encryption key for the Caesar cipher
|
93 |
mt_srand((double)microtime()*1000000); // (PHP < 4.2.0) |
94 |
$shift = mt_rand(1, 25); |
95 |
// encrypt the email using an adapted Caesar cipher
|
96 |
$encrypted_email = ""; |
97 |
for($i = strlen($email_address) -1; $i > -1; $i--) { |
98 |
if(preg_match('#[FZXK0-9]#', $email_address[$i], $characters)) { |
99 |
$encrypted_email .= $email_address[$i]; |
100 |
} else {
|
101 |
$encrypted_email .= chr((ord($email_address[$i]) -97 + $shift) % 26 + 97); |
102 |
} |
103 |
} |
104 |
$encrypted_email .= chr($shift + 97); |
105 |
// build the encrypted Javascript mailto link
|
106 |
$mailto_link = "<a {$class_attr}{$id_attr}href=\"javascript:mdcr('$encrypted_email','$email_subject')\">" .$match[5] ."</a>"; |
107 |
$bNeedMdcr = true; |
108 |
return $mailto_link; |
109 |
} else {
|
110 |
/** DO NOT USE JAVASCRIPT ENCRYPTION FOR MAILTO LINKS **/
|
111 |
// as minimum protection, replace @ in the mailto part by (at)
|
112 |
// dots are not transformed as this would transform my.name@domain.com into: my(dot)name(at)domain(dot)com
|
113 |
// rebuild the mailto link from the subpatterns (at the missing characters " and </a>")
|
114 |
return $match[1].str_replace('@', $aFilterSettings['at_replacement'],$match[2]).$match[3].'"'.$match[4].$match[5].'</a>'; |
115 |
// if you want to protect both, @ and dots, comment out the line above and remove the comment from the line below
|
116 |
// return $match[1] .str_replace($search, $replace, $match[2]) .$match[3] .'"' .$match[4] .$match[5] .'</a>';
|
117 |
} |
118 |
break;
|
119 |
default:
|
120 |
// number of subpatterns do not match the requirements ... do nothing
|
121 |
return $match[0]; |
122 |
break;
|
123 |
} |
124 |
}, |
125 |
/* ************************************************************************** */
|
126 |
$content
|
127 |
); |
128 |
} |
129 |
if ($bNeedMdcr) { |
130 |
// test if js-decryption is installed
|
131 |
if (!preg_match('/<head.*<.*src=\".*\/mdcr.js.*>.*<\/head/siU', $content) ) { |
132 |
// try to insert js-decrypt into <head> if available
|
133 |
$script = str_replace('\\', '/',str_replace(WB_PATH,'', dirname(__DIR__)).'/js/mdcr.js'); |
134 |
if(is_readable(WB_PATH.$script)) { |
135 |
$scriptLink = "\t".'<script src="'.WB_URL.$script.'" type="text/javascript"></script>'."\n"; |
136 |
$regex = '/(.*)(<\s*?\/\s*?head\s*>.*)/isU'; |
137 |
$replace = '$1'.$scriptLink.'$2'; |
138 |
$content = preg_replace ($regex, $replace, $content); |
139 |
} |
140 |
} |
141 |
} |
142 |
return $content; |
143 |
} |