3 ///////////////////////////////////////////////////////////////////////////
5 // NOTICE OF COPYRIGHT //
7 // Moodle - Modular Object-Oriented Dynamic Learning Environment //
8 // http://moodle.com //
10 // Copyright (C) 2001-3001 Martin Dougiamas http://dougiamas.com //
11 // (C) 2001-3001 Eloy Lafuente (stronk7) http://contiento.com //
13 // This program is free software; you can redistribute it and/or modify //
14 // it under the terms of the GNU General Public License as published by //
15 // the Free Software Foundation; either version 2 of the License, or //
16 // (at your option) any later version. //
18 // This program is distributed in the hope that it will be useful, //
19 // but WITHOUT ANY WARRANTY; without even the implied warranty of //
20 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the //
21 // GNU General Public License for more details: //
23 // http://www.gnu.org/copyleft/gpl.html //
25 ///////////////////////////////////////////////////////////////////////////
28 require_once($CFG->libdir
.'/typo3/class.t3lib_cs.php');
29 require_once($CFG->libdir
.'/typo3/class.t3lib_div.php');
31 /// If ICONV is available, lets Typo3 library use it for convert
32 if (extension_loaded('iconv')) {
33 $GLOBALS['TYPO3_CONF_VARS']['SYS']['t3lib_cs_convMethod'] = 'iconv';
34 /// Else if mbstring is available, lets Typo3 library use it
35 } else if (extension_loaded('mbstring')) {
36 $GLOBALS['TYPO3_CONF_VARS']['SYS']['t3lib_cs_convMethod'] = 'mbstring';
37 /// Else if recode is available, lets Typo3 library use it
38 } else if (extension_loaded('recode')) {
39 $GLOBALS['TYPO3_CONF_VARS']['SYS']['t3lib_cs_convMethod'] = 'recode';
41 $GLOBALS['TYPO3_CONF_VARS']['SYS']['t3lib_cs_convMethod'] = '';
44 /// If mbstring is available, lets Typo3 library use it for functions
45 if (extension_loaded('mbstring')) {
46 $GLOBALS['TYPO3_CONF_VARS']['SYS']['t3lib_cs_utils'] = 'mbstring';
48 $GLOBALS['TYPO3_CONF_VARS']['SYS']['t3lib_cs_utils'] = '';
51 /// And this directory must exist to allow Typo to cache conversion
52 /// tables when using internal functions
53 make_upload_directory('temp/typo3temp/cs');
55 /// Default mask for Typo
56 $GLOBALS['TYPO3_CONF_VARS']['BE']['fileCreateMask'] = $CFG->directorypermissions
;
58 /// This full path constants must be defined too, transforming backslashes
59 /// to forward slashed beacuse Typo3 requires it.
60 define ('PATH_t3lib', str_replace('\\','/',$CFG->libdir
.'/typo3/'));
61 define ('PATH_typo3', str_replace('\\','/',$CFG->libdir
.'/typo3/'));
62 define ('PATH_site', str_replace('\\','/',$CFG->dataroot
.'/temp/'));
63 define ('TYPO3_OS', stristr(PHP_OS
,'win')&&!stristr(PHP_OS
,'darwin')?
'WIN':'');
66 /// As we implement the singleton pattern to use this class (only one instance
67 /// is shared globally), we need this helper function
69 /// IMPORTANT Note: Typo3 libraries always expect lowercase charsets to use 100%
70 /// its capabilities so, don't forget to make the conversion
71 /// from every wrapper function!
73 function textlib_get_instance () {
75 if (!is_object($instance)) {
76 $instance = new textlib();
82 * This class is used to manipulate strings under Moodle 1.6 an later. As
83 * utf-8 text become mandatory a pool of safe functions under this encoding
84 * become necessary. The name of the methods is exactly the
85 * same than their PHP originals.
87 * A big part of this class acts as a wrapper over the Typo3 charset library,
88 * really a cool group of utilities to handle texts and encoding conversion.
90 * Take a look to its own copyright and license details.
97 * Standard constructor of the class. All it does is to instantiate
98 * a new t3lib_cs object to have all their functions ready.
100 * Instead of istantiating a lot of objects of this class everytime
101 * some of their functions is going to be used, you can invoke the:
102 * textlib_get_instance() function, avoiding the creation of them
103 * (following the singleton pattern)
106 /// Instantiate a conversor object some of the methods in typo3
107 /// reference to $this and cannot be executed in a static context
108 $this->typo3cs
= new t3lib_cs();
112 * Converts the text between different encodings. It will use iconv, mbstring
113 * or internal (typo3) methods to try such conversion. Returns false if fails.
115 function convert($text, $fromCS, $toCS='utf-8') {
116 /// Normalize charsets
117 $fromCS = $this->typo3cs
->parse_charset($fromCS);
118 $toCS = $this->typo3cs
->parse_charset($toCS);
119 /// Avoid some notices from Typo3 code
120 $oldlevel = error_reporting(E_PARSE
);
121 /// Call Typo3 conv() function. It will do all the work
122 $result = $this->typo3cs
->conv($text, $fromCS, $toCS);
123 /// Restore original debug level
124 error_reporting($oldlevel);
129 * Multibyte safe substr() function, uses mbstring if available.
131 function substr($text, $start, $len=null, $charset='utf-8') {
132 /// Normalize charset
133 $charset = $this->typo3cs
->parse_charset($charset);
134 /// Avoid some notices from Typo3 code
135 $oldlevel = error_reporting(E_PARSE
);
136 /// Call Typo3 substr() function. It will do all the work
137 $result = $this->typo3cs
->substr($charset,$text,$start,$len);
138 /// Restore original debug level
139 error_reporting($oldlevel);
144 * Multibyte safe strlen() function, uses mbstring if available.
146 function strlen($text, $charset='utf-8') {
147 /// Normalize charset
148 $charset = $this->typo3cs
->parse_charset($charset);
149 /// Avoid some notices from Typo3 code
150 $oldlevel = error_reporting(E_PARSE
);
151 /// Call Typo3 strlen() function. It will do all the work
152 $result = $this->typo3cs
->strlen($charset,$text);
153 /// Restore original debug level
154 error_reporting($oldlevel);
159 * Multibyte safe strtolower() function, uses mbstring if available.
161 function strtolower($text, $charset='utf-8') {
162 /// Normalize charset
163 $charset = $this->typo3cs
->parse_charset($charset);
164 /// Avoid some notices from Typo3 code
165 $oldlevel = error_reporting(E_PARSE
);
166 /// Call Typo3 conv_case() function. It will do all the work
167 $result = $this->typo3cs
->conv_case($charset,$text,'toLower');
168 /// Restore original debug level
169 error_reporting($oldlevel);
174 * Multibyte safe strtoupper() function, uses mbstring if available.
176 function strtoupper($text, $charset='utf-8') {
177 /// Normalize charset
178 $charset = $this->typo3cs
->parse_charset($charset);
179 /// Avoid some notices from Typo3 code
180 $oldlevel = error_reporting(E_PARSE
);
181 /// Call Typo3 conv_case() function. It will do all the work
182 $result = $this->typo3cs
->conv_case($charset,$text,'toUpper');
183 /// Restore original debug level
184 error_reporting($oldlevel);
189 * UTF-8 ONLY safe strpos() function, uses mbstring if available.
191 function strpos($haystack,$needle,$offset=0) {
192 /// Call Typo3 utf8_strpos() function. It will do all the work
193 return $this->typo3cs
->utf8_strpos($haystack,$needle,$offset);
197 * UTF-8 ONLY safe strrpos() function, uses mbstring if available.
199 function strrpos($haystack,$needle) {
200 /// Call Typo3 utf8_strrpos() function. It will do all the work
201 return $this->typo3cs
->utf8_strrpos($haystack,$needle);
205 * Try to convert upper unicode characters to plain ascii,
206 * the returned string may cantain unconverted unicode characters.
208 function specialtoascii($text,$charset='utf-8') {
209 /// Normalize charset
210 $charset = $this->typo3cs
->parse_charset($charset);
211 /// Avoid some notices from Typo3 code
212 $oldlevel = error_reporting(E_PARSE
);
213 $result = $this->typo3cs
->specCharsToASCII($charset,$text);
214 /// Restore original debug level
215 error_reporting($oldlevel);
220 * Generate a correct base64 encoded header to be used in MIME mail messages.
221 * This function seems to be 100% compliant with RFC1342. Credits go to:
222 * paravoid (http://www.php.net/manual/en/function.mb-encode-mimeheader.php#60283).
224 function encode_mimeheader($text, $charset='utf-8') {
226 return (string)$text;
228 /// Normalize charset
229 $charset = $this->typo3cs
->parse_charset($charset);
230 /// If the text is pure ASCII, we don't need to encode it
231 if ($this->convert($text, $charset, 'ascii') == $text) {
234 /// Although RFC says that line feed should be \r\n, it seems that
235 /// some mailers double convert \r, so we are going to use \n alone
237 /// Define start and end of every chunk
238 $start = "=?$charset?B?";
240 /// Acumulate results
242 /// Max line length is 75 (including start and end)
243 $length = 75 - strlen($start) - strlen($end);
245 $multilength = $this->strlen($text, $charset);
246 /// Detect if strlen and friends supported
247 if ($multilength === false) {
248 if ($charset == 'GB18030' or $charset == 'gb18030') {
249 while (strlen($text)) {
250 // try to encode first 22 chars - we expect most chars are two bytes long
251 if (preg_match('/^(([\x00-\x7f])|([\x81-\xfe][\x40-\x7e])|([\x81-\xfe][\x80-\xfe])|([\x81-\xfe][\x30-\x39]..)){1,22}/m', $text, $matches)) {
252 $chunk = $matches[0];
253 $encchunk = base64_encode($chunk);
254 if (strlen($encchunk) > $length) {
255 // find first 11 chars - each char in 4 bytes - worst case scenario
256 preg_match('/^(([\x00-\x7f])|([\x81-\xfe][\x40-\x7e])|([\x81-\xfe][\x80-\xfe])|([\x81-\xfe][\x30-\x39]..)){1,11}/m', $text, $matches);
257 $chunk = $matches[0];
258 $encchunk = base64_encode($chunk);
260 $text = substr($text, strlen($chunk));
261 $encoded .= ' '.$start.$encchunk.$end.$linefeed;
266 $encoded = trim($encoded);
272 $ratio = $multilength / strlen($text);
274 $magic = $avglength = floor(3 * $length * $ratio / 4);
275 /// basic infinite loop protection
276 $maxiterations = strlen($text)*2;
278 /// Iterate over the string in magic chunks
279 for ($i=0; $i <= $multilength; $i+
=$magic) {
280 if ($iteration++
> $maxiterations) {
281 return false; // probably infinite loop
285 /// Ensure the chunk fits in length, reduding magic if necessary
288 $chunk = $this->substr($text, $i, $magic, $charset);
289 $chunk = base64_encode($chunk);
291 } while (strlen($chunk) > $length);
292 /// This chunk doen't break any multi-byte char. Use it.
294 $encoded .= ' '.$start.$chunk.$end.$linefeed;
296 /// Strip the first space and the last linefeed
297 $encoded = substr($encoded, 1, -strlen($linefeed));
303 * Converts all the numeric entities &#nnnn; or &#xnnn; to UTF-8
304 * Original from laurynas dot butkus at gmail at:
305 * http://php.net/manual/en/function.html-entity-decode.php#75153
306 * with some custom mods to provide more functionality
308 * @param string $str input string
309 * @param boolean $htmlent convert also html entities (defaults to true)
311 * NOTE: we could have used typo3 entities_to_utf8() here
312 * but the direct alternative used runs 400% quicker
313 * and uses 0.5Mb less memory, so, let's use it
314 * (tested agains 10^6 conversions)
316 function entities_to_utf8($str, $htmlent=true) {
318 static $trans_tbl; /// Going to use static translit table
320 /// Replace numeric entities
321 $result = preg_replace('~&#x([0-9a-f]+);~ei', 'textlib::code2utf8(hexdec("\\1"))', $str);
322 $result = preg_replace('~&#([0-9]+);~e', 'textlib::code2utf8(\\1)', $result);
324 /// Replace literal entities (if desired)
326 /// Generate/create $trans_tbl
327 if (!isset($trans_tbl)) {
328 $trans_tbl = array();
329 foreach (get_html_translation_table(HTML_ENTITIES
) as $val=>$key) {
330 $trans_tbl[$key] = utf8_encode($val);
333 $result = strtr($result, $trans_tbl);
335 /// Return utf8-ised string
340 * Converts all Unicode chars > 127 to numeric entities &#nnnn; or &#xnnn;.
342 * @param string input string
343 * @param boolean output decadic only number entities
344 * @param boolean remove all nonumeric entities
345 * @return string converted string
347 function utf8_to_entities($str, $dec=false, $nonnum=false) {
348 /// Avoid some notices from Typo3 code
349 $oldlevel = error_reporting(E_PARSE
);
351 $str = $this->typo3cs
->entities_to_utf8($str, true);
353 $result = $this->typo3cs
->utf8_to_entities($str);
355 $result = preg_replace('/&#x([0-9a-f]+);/ie', "'&#'.hexdec('$1').';'", $result);
357 /// Restore original debug level
358 error_reporting($oldlevel);
363 * Removes the BOM from unicode string - see http://unicode.org/faq/utf_bom.html
365 function trim_utf8_bom($str) {
366 $bom = "\xef\xbb\xbf";
367 if (strpos($str, $bom) === 0) {
368 return substr($str, strlen($bom));
374 * Returns encoding options for select boxes, utf-8 and platform encoding first
375 * @return array encodings
377 function get_encodings() {
378 $encodings = array();
379 $encodings['UTF-8'] = 'UTF-8';
380 $winenc = strtoupper(get_string('localewincharset'));
382 $encodings[$winenc] = $winenc;
384 $nixenc = strtoupper(get_string('oldcharset'));
385 $encodings[$nixenc] = $nixenc;
387 foreach ($this->typo3cs
->synonyms
as $enc) {
388 $enc = strtoupper($enc);
389 $encodings[$enc] = $enc;
395 * Returns the utf8 string corresponding to the unicode value
396 * (from php.net, courtesy - romans@void.lv)
398 * @param int $num one unicode value
399 * @return string the UTF-8 char corresponding to the unicode value
401 function code2utf8($num) {
406 return chr(($num >> 6) +
192) . chr(($num & 63) +
128);
409 return chr(($num >> 12) +
224) . chr((($num >> 6) & 63) +
128) . chr(($num & 63) +
128);
411 if ($num < 2097152) {
412 return chr(($num >> 18) +
240) . chr((($num >> 12) & 63) +
128) . chr((($num >> 6) & 63) +
128) . chr(($num & 63) +
128);