|
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 $osug; |
|
51 } |
|
52 |
|
53 protected function &_getMatches($lang, $str) { |
|
54 $lang = preg_replace('/[^a-z\-]/i', '', $lang); |
|
55 $str = preg_replace('/[\x00-\x1F\x7F]/', '', $str); |
|
56 $server = "www.google.com"; |
|
57 $port = 443; |
|
58 $path = "/tbproxy/spell?lang=" . $lang . "&hl=en"; |
|
59 $host = "www.google.com"; |
|
60 $url = "https://" . $server; |
|
61 |
|
62 // Setup XML request |
|
63 $xml = '<?xml version="1.0" encoding="utf-8" ?><spellrequest textalreadyclipped="0" ignoredups="0" ignoredigits="1" ignoreallcaps="1"><text>' . $str . '</text></spellrequest>'; |
|
64 |
|
65 $header = "POST ".$path." HTTP/1.0 \r\n"; |
|
66 $header .= "MIME-Version: 1.0 \r\n"; |
|
67 $header .= "Content-type: application/PTI26 \r\n"; |
|
68 $header .= "Content-length: ".strlen($xml)." \r\n"; |
|
69 $header .= "Content-transfer-encoding: text \r\n"; |
|
70 $header .= "Request-number: 1 \r\n"; |
|
71 $header .= "Document-type: Request \r\n"; |
|
72 $header .= "Interface-Version: Test 1.4 \r\n"; |
|
73 $header .= "Connection: close \r\n\r\n"; |
|
74 $header .= $xml; |
|
75 |
|
76 // Use curl if it exists |
|
77 if (function_exists('curl_init')) { |
|
78 // Use curl |
|
79 $ch = curl_init(); |
|
80 curl_setopt($ch, CURLOPT_URL,$url); |
|
81 curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1); |
|
82 curl_setopt($ch, CURLOPT_CUSTOMREQUEST, $header); |
|
83 curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, FALSE); |
|
84 $xml = curl_exec($ch); |
|
85 curl_close($ch); |
|
86 } else { |
|
87 // Use raw sockets |
|
88 $fp = fsockopen("ssl://" . $server, $port, $errno, $errstr, 30); |
|
89 if ($fp) { |
|
90 // Send request |
|
91 fwrite($fp, $header); |
|
92 |
|
93 // Read response |
|
94 $xml = ""; |
|
95 while (!feof($fp)) |
|
96 $xml .= fgets($fp, 128); |
|
97 |
|
98 fclose($fp); |
|
99 } else |
|
100 echo "Could not open SSL connection to google."; |
|
101 } |
|
102 |
|
103 // Grab and parse content |
|
104 $matches = array(); |
|
105 preg_match_all('/<c o="([^"]*)" l="([^"]*)" s="([^"]*)">([^<]*)<\/c>/', $xml, $matches, PREG_SET_ORDER); |
|
106 |
|
107 return $matches; |
|
108 } |
|
109 |
|
110 protected function _unhtmlentities($string) { |
|
111 $string = preg_replace('~&#x([0-9a-f]+);~ei', 'chr(hexdec("\\1"))', $string); |
|
112 $string = preg_replace('~&#([0-9]+);~e', 'chr(\\1)', $string); |
|
113 |
|
114 $trans_tbl = get_html_translation_table(HTML_ENTITIES); |
|
115 $trans_tbl = array_flip($trans_tbl); |
|
116 |
|
117 return strtr($string, $trans_tbl); |
|
118 } |
|
119 } |
|
120 |
|
121 // Patch in multibyte support |
|
122 if (!function_exists('mb_substr')) { |
|
123 function mb_substr($str, $start, $len = '', $encoding="UTF-8"){ |
|
124 $limit = strlen($str); |
|
125 |
|
126 for ($s = 0; $start > 0;--$start) {// found the real start |
|
127 if ($s >= $limit) |
|
128 break; |
|
129 |
|
130 if ($str[$s] <= "\x7F") |
|
131 ++$s; |
|
132 else { |
|
133 ++$s; // skip length |
|
134 |
|
135 while ($str[$s] >= "\x80" && $str[$s] <= "\xBF") |
|
136 ++$s; |
|
137 } |
|
138 } |
|
139 |
|
140 if ($len == '') |
|
141 return substr($str, $s); |
|
142 else |
|
143 for ($e = $s; $len > 0; --$len) {//found the real end |
|
144 if ($e >= $limit) |
|
145 break; |
|
146 |
|
147 if ($str[$e] <= "\x7F") |
|
148 ++$e; |
|
149 else { |
|
150 ++$e;//skip length |
|
151 |
|
152 while ($str[$e] >= "\x80" && $str[$e] <= "\xBF" && $e < $limit) |
|
153 ++$e; |
|
154 } |
|
155 } |
|
156 |
|
157 return substr($str, $s, $e - $s); |
|
158 } |
|
159 } |
|
160 |
|
161 ?> |