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