Commit | Line | Data |
---|---|---|
fb916bf0 PS |
1 | <?php |
2 | /** | |
3 | * $Id: editor_plugin_src.js 201 2007-02-12 15:56:56Z spocke $ | |
4 | * | |
5 | * @package MCManager.includes | |
6 | * @author Moxiecode | |
7 | * @copyright Copyright � 2004-2007, Moxiecode Systems AB, All rights reserved. | |
8 | */ | |
9 | ||
10 | class GoogleSpell extends SpellChecker { | |
11 | /** | |
12 | * Spellchecks an array of words. | |
13 | * | |
14 | * @param {String} $lang Language code like sv or en. | |
15 | * @param {Array} $words Array of words to spellcheck. | |
16 | * @return {Array} Array of misspelled words. | |
17 | */ | |
18 | function &checkWords($lang, $words) { | |
19 | $wordstr = implode(' ', $words); | |
20 | $matches = $this->_getMatches($lang, $wordstr); | |
21 | $words = array(); | |
22 | ||
23 | for ($i=0; $i<count($matches); $i++) | |
24 | $words[] = $this->_unhtmlentities(mb_substr($wordstr, $matches[$i][1], $matches[$i][2], "UTF-8")); | |
25 | ||
26 | return $words; | |
27 | } | |
28 | ||
29 | /** | |
30 | * Returns suggestions of for a specific word. | |
31 | * | |
32 | * @param {String} $lang Language code like sv or en. | |
33 | * @param {String} $word Specific word to get suggestions for. | |
34 | * @return {Array} Array of suggestions for the specified word. | |
35 | */ | |
36 | function &getSuggestions($lang, $word) { | |
37 | $sug = array(); | |
38 | $osug = array(); | |
39 | $matches = $this->_getMatches($lang, $word); | |
40 | ||
41 | if (count($matches) > 0) | |
42 | $sug = explode("\t", utf8_encode($this->_unhtmlentities($matches[0][4]))); | |
43 | ||
44 | // Remove empty | |
45 | foreach ($sug as $item) { | |
46 | if ($item) | |
47 | $osug[] = $item; | |
48 | } | |
49 | ||
50 | return $osug; | |
51 | } | |
52 | ||
53 | function &_getMatches($lang, $str) { | |
54 | $server = "www.google.com"; | |
55 | $port = 443; | |
56 | $path = "/tbproxy/spell?lang=" . $lang . "&hl=en"; | |
57 | $host = "www.google.com"; | |
58 | $url = "https://" . $server; | |
59 | ||
60 | // Setup XML request | |
61 | $xml = '<?xml version="1.0" encoding="utf-8" ?><spellrequest textalreadyclipped="0" ignoredups="0" ignoredigits="1" ignoreallcaps="1"><text>' . $str . '</text></spellrequest>'; | |
62 | ||
63 | $header = "POST ".$path." HTTP/1.0 \r\n"; | |
64 | $header .= "MIME-Version: 1.0 \r\n"; | |
65 | $header .= "Content-type: application/PTI26 \r\n"; | |
66 | $header .= "Content-length: ".strlen($xml)." \r\n"; | |
67 | $header .= "Content-transfer-encoding: text \r\n"; | |
68 | $header .= "Request-number: 1 \r\n"; | |
69 | $header .= "Document-type: Request \r\n"; | |
70 | $header .= "Interface-Version: Test 1.4 \r\n"; | |
71 | $header .= "Connection: close \r\n\r\n"; | |
72 | $header .= $xml; | |
73 | ||
74 | // Use curl if it exists | |
75 | if (function_exists('curl_init')) { | |
76 | // Use curl | |
77 | $ch = curl_init(); | |
78 | curl_setopt($ch, CURLOPT_URL,$url); | |
79 | curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1); | |
80 | curl_setopt($ch, CURLOPT_CUSTOMREQUEST, $header); | |
81 | curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, FALSE); | |
8f9f70fb PS |
82 | if (!empty($this->_config['GoogleSpell.proxyhost'])) { |
83 | if (!empty($this->_config['GoogleSpell.proxytype']) and ($this->_config['GoogleSpell.proxytype'] === 'SOCKS5')) { | |
84 | curl_setopt($ch, CURLOPT_PROXYTYPE, CURLPROXY_SOCKS5); | |
85 | } else { | |
86 | curl_setopt($ch, CURLOPT_PROXYTYPE, CURLPROXY_HTML); | |
87 | curl_setopt($ch, CURLOPT_HTTPPROXYTUNNEL, FALSE); | |
88 | } | |
89 | if (empty($this->_config['GoogleSpell.proxyport'])) { | |
90 | curl_setopt($ch, CURLOPT_PROXY, $this->_config['GoogleSpell.proxyhost']); | |
91 | } else { | |
92 | curl_setopt($ch, CURLOPT_PROXY, $this->_config['GoogleSpell.proxyhost'].':'.$this->_config['GoogleSpell.proxyport']); | |
93 | } | |
94 | if (!empty($this->_config['GoogleSpell.proxyuser']) and !empty($this->_config['GoogleSpell.proxypassword'])) { | |
95 | curl_setopt($ch, CURLOPT_PROXYUSERPWD, $this->_config['GoogleSpell.proxyuser'].':'.$this->_config['GoogleSpell.proxypassword']); | |
96 | if (defined('CURLOPT_PROXYAUTH')) { | |
97 | // any proxy authentication if PHP 5.1 | |
98 | curl_setopt($ch, CURLOPT_PROXYAUTH, CURLAUTH_BASIC | CURLAUTH_NTLM); | |
99 | } | |
100 | } | |
101 | } | |
fb916bf0 PS |
102 | $xml = curl_exec($ch); |
103 | curl_close($ch); | |
104 | } else { | |
105 | // Use raw sockets | |
106 | $fp = fsockopen("ssl://" . $server, $port, $errno, $errstr, 30); | |
107 | if ($fp) { | |
108 | // Send request | |
109 | fwrite($fp, $header); | |
110 | ||
111 | // Read response | |
112 | $xml = ""; | |
113 | while (!feof($fp)) | |
114 | $xml .= fgets($fp, 128); | |
115 | ||
116 | fclose($fp); | |
117 | } else | |
118 | echo "Could not open SSL connection to google."; | |
119 | } | |
120 | ||
121 | // Grab and parse content | |
122 | $matches = array(); | |
123 | preg_match_all('/<c o="([^"]*)" l="([^"]*)" s="([^"]*)">([^<]*)<\/c>/', $xml, $matches, PREG_SET_ORDER); | |
124 | ||
125 | return $matches; | |
126 | } | |
127 | ||
128 | function _unhtmlentities($string) { | |
129 | $string = preg_replace('~&#x([0-9a-f]+);~ei', 'chr(hexdec("\\1"))', $string); | |
130 | $string = preg_replace('~&#([0-9]+);~e', 'chr(\\1)', $string); | |
131 | ||
132 | $trans_tbl = get_html_translation_table(HTML_ENTITIES); | |
133 | $trans_tbl = array_flip($trans_tbl); | |
134 | ||
135 | return strtr($string, $trans_tbl); | |
136 | } | |
137 | } | |
138 | ||
139 | // Patch in multibyte support | |
140 | if (!function_exists('mb_substr')) { | |
141 | function mb_substr($str, $start, $len = '', $encoding="UTF-8"){ | |
142 | $limit = strlen($str); | |
143 | ||
144 | for ($s = 0; $start > 0;--$start) {// found the real start | |
145 | if ($s >= $limit) | |
146 | break; | |
147 | ||
148 | if ($str[$s] <= "\x7F") | |
149 | ++$s; | |
150 | else { | |
151 | ++$s; // skip length | |
152 | ||
153 | while ($str[$s] >= "\x80" && $str[$s] <= "\xBF") | |
154 | ++$s; | |
155 | } | |
156 | } | |
157 | ||
158 | if ($len == '') | |
159 | return substr($str, $s); | |
160 | else | |
161 | for ($e = $s; $len > 0; --$len) {//found the real end | |
162 | if ($e >= $limit) | |
163 | break; | |
164 | ||
165 | if ($str[$e] <= "\x7F") | |
166 | ++$e; | |
167 | else { | |
168 | ++$e;//skip length | |
169 | ||
170 | while ($str[$e] >= "\x80" && $str[$e] <= "\xBF" && $e < $limit) | |
171 | ++$e; | |
172 | } | |
173 | } | |
174 | ||
175 | return substr($str, $s, $e - $s); | |
176 | } | |
177 | } | |
178 | ||
179 | ?> |