Project

General

Profile

1
<?php
2

    
3
// $Id: functions-utf8.php 915 2009-01-21 19:27:01Z Ruebenwurzel $
4

    
5
/*
6

    
7
 Website Baker Project <http://www.websitebaker.org/>
8
 Copyright (C) 2004-2009, Ryan Djurovich
9

    
10
 Website Baker is free software; you can redistribute it and/or modify
11
 it under the terms of the GNU General Public License as published by
12
 the Free Software Foundation; either version 2 of the License, or
13
 (at your option) any later version.
14

    
15
 Website Baker is distributed in the hope that it will be useful,
16
 but WITHOUT ANY WARRANTY; without even the implied warranty of
17
 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
18
 GNU General Public License for more details.
19

    
20
 You should have received a copy of the GNU General Public License
21
 along with Website Baker; if not, write to the Free Software
22
 Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
23

    
24
*/
25

    
26
/*
27
 * A part of this file is based on 'utf8.php' from the DokuWiki-project.
28
 * (http://www.splitbrain.org/projects/dokuwiki):
29
 **
30
 * UTF8 helper functions
31
 * @license    LGPL (http://www.gnu.org/copyleft/lesser.html)
32
 * @author     Andreas Gohr <andi@splitbrain.org>
33
 **
34
 * modified for use with Website Baker
35
 * from thorn, Jan. 2008
36
 *
37
 * most of the original functions appeared to be to slow with large strings, so i replaced them with my own ones
38
 * thorn, Mar. 2008
39
 */
40

    
41
// Functions we use in Website Baker:
42
//   entities_to_7bit()
43
//   entities_to_umlauts2()
44
//   umlauts_to_entities2()
45

    
46
if(!defined('WB_URL')) {
47
	header('Location: ../index.php');
48
	exit(0);
49
}
50

    
51
/*
52
 * check for mb_string support
53
 */
54
//define('UTF8_NOMBSTRING',1); // uncomment this to forbid use of mb_string-functions
55
if(!defined('UTF8_MBSTRING')){
56
  if(function_exists('mb_substr') && !defined('UTF8_NOMBSTRING')){
57
    define('UTF8_MBSTRING',1);
58
  }else{
59
    define('UTF8_MBSTRING',0);
60
  }
61
}
62

    
63
if(UTF8_MBSTRING){ mb_internal_encoding('UTF-8'); }
64

    
65
require_once(WB_PATH.'/framework/charsets_table.php');
66

    
67
/*
68
 * Checks if a string contains 7bit ASCII only
69
 *
70
 * @author thorn
71
 */
72
function utf8_isASCII($str){
73
	if(preg_match('/[\x80-\xFF]/', $str))
74
		return false;
75
	else
76
		return true;
77
}
78

    
79
/*
80
 * Tries to detect if a string is in Unicode encoding
81
 *
82
 * @author <bmorel@ssi.fr>
83
 * @link   http://www.php.net/manual/en/function.utf8-encode.php
84
 */
85
function utf8_check($Str) {
86
 for ($i=0; $i<strlen($Str); $i++) {
87
  $b = ord($Str[$i]);
88
  if ($b < 0x80) continue; # 0bbbbbbb
89
  elseif (($b & 0xE0) == 0xC0) $n=1; # 110bbbbb
90
  elseif (($b & 0xF0) == 0xE0) $n=2; # 1110bbbb
91
  elseif (($b & 0xF8) == 0xF0) $n=3; # 11110bbb
92
  elseif (($b & 0xFC) == 0xF8) $n=4; # 111110bb
93
  elseif (($b & 0xFE) == 0xFC) $n=5; # 1111110b
94
  else return false; # Does not match any model
95
  for ($j=0; $j<$n; $j++) { # n bytes matching 10bbbbbb follow ?
96
   if ((++$i == strlen($Str)) || ((ord($Str[$i]) & 0xC0) != 0x80))
97
   return false;
98
  }
99
 }
100
 return true;
101
}
102

    
103
/*
104
 * Romanize a non-latin string
105
 *
106
 * @author Andreas Gohr <andi@splitbrain.org>
107
 */
108
function utf8_romanize($string){
109
  if(utf8_isASCII($string)) return $string; //nothing to do
110

    
111
  global $UTF8_ROMANIZATION;
112
  return strtr($string,$UTF8_ROMANIZATION);
113
}
114

    
115
/*
116
 * Removes special characters (nonalphanumeric) from a UTF-8 string
117
 *
118
 * This function adds the controlchars 0x00 to 0x19 to the array of
119
 * stripped chars (they are not included in $UTF8_SPECIAL_CHARS2)
120
 *
121
 * @author Andreas Gohr <andi@splitbrain.org>
122
 * @param  string $string     The UTF8 string to strip of special chars
123
 * @param  string $repl       Replace special with this string
124
 * @param  string $additional Additional chars to strip (used in regexp char class)
125
 */
126
function utf8_stripspecials($string,$repl='',$additional=''){
127
  global $UTF8_SPECIAL_CHARS2;
128

    
129
  static $specials = null;
130
  if(is_null($specials)){
131
    $specials = preg_quote($UTF8_SPECIAL_CHARS2, '/');
132
  }
133

    
134
  return preg_replace('/['.$additional.'\x00-\x19'.$specials.']/u',$repl,$string);
135
}
136

    
137
/*
138
 * added functions - thorn
139
 */
140

    
141
/*
142
 * faster replacement for utf8_entities_to_umlauts()
143
 * not all features of utf8_entities_to_umlauts() --> utf8_unhtml() are supported!
144
 * @author thorn
145
 */
146
function utf8_fast_entities_to_umlauts($str) {
147
	if(UTF8_MBSTRING) {
148
		// we need this for use with mb_convert_encoding
149
		$str = str_replace(array('&amp;','&gt;','&lt;','&quot;','&#039;','&nbsp;'), array('&amp;amp;','&amp;gt;','&amp;lt;','&amp;quot;','&amp;#39;','&amp;nbsp;'), $str);
150
		// we need two mb_convert_encoding()-calls - is this a bug?
151
		// mb_convert_encoding("ö&ouml;", 'UTF-8', 'HTML-ENTITIES'); // with string in utf-8-encoding doesn't work. Result: "öö"
152
		// Work-around: convert all umlauts to entities first ("ö&ouml;"->"&ouml;&ouml;"), then all entities to umlauts ("&ouml;&ouml;"->"öö")
153
		return(mb_convert_encoding(mb_convert_encoding($str, 'HTML-ENTITIES', 'UTF-8'),'UTF-8', 'HTML-ENTITIES'));
154
	} else {
155
		global $named_entities;global $numbered_entities;
156
		$str = str_replace($named_entities, $numbered_entities, $str);
157
		$str = preg_replace("/&#([0-9]+);/e", "code_to_utf8($1)", $str);
158
	}
159
	return($str);
160
}
161
// support-function for utf8_fast_entities_to_umlauts()
162
function code_to_utf8($num) {
163
	if ($num <= 0x7F) {
164
		return chr($num);
165
	} elseif ($num <= 0x7FF) {
166
		return chr(($num >> 6) + 192) . chr(($num & 63) + 128);
167
	} elseif ($num <= 0xFFFF) {
168
		 return chr(($num >> 12) + 224) . chr((($num >> 6) & 63) + 128) . chr(($num & 63) + 128);
169
	} elseif ($num <= 0x1FFFFF) {
170
		return chr(($num >> 18) + 240) . chr((($num >> 12) & 63) + 128) . chr((($num >> 6) & 63) + 128) . chr(($num & 63) + 128);
171
	}
172
	return "?";
173
}
174

    
175
/*
176
 * faster replacement for utf8_umlauts_to_entities()
177
 * not all features of utf8_umlauts_to_entities() --> utf8_tohtml() are supported!
178
 * @author thorn
179
 */
180
function utf8_fast_umlauts_to_entities($string, $named_entities=true) {
181
	if(UTF8_MBSTRING)
182
		return(mb_convert_encoding($string, 'HTML-ENTITIES', 'UTF-8'));
183
	else {
184
		global $named_entities;global $numbered_entities;
185
		$new = "";
186
		$i=0;
187
		$len=strlen($string);
188
		if($len==0) return $string;
189
		do {
190
			if(ord($string{$i}) <= 127) $ud = $string{$i++};
191
			elseif(ord($string{$i}) <= 223) $ud = (ord($string{$i++})-192)*64 + (ord($string{$i++})-128);
192
			elseif(ord($string{$i}) <= 239) $ud = (ord($string{$i++})-224)*4096 + (ord($string{$i++})-128)*64 + (ord($string{$i++})-128);
193
			elseif(ord($string{$i}) <= 247) $ud = (ord($string{$i++})-240)*262144 + (ord($string{$i++})-128)*4096 + (ord($string{$i++})-128)*64 + (ord($string{$i++})-128);
194
			else $ud = ord($string{$i++}); // error!
195
			if($ud > 127) {
196
				$new .= "&#$ud;";
197
			} else {
198
				$new .= $ud;
199
			}
200
		} while($i < $len);
201
		$string = $new;
202
		if($named_entities)
203
			$string = str_replace($numbered_entities, $named_entities, $string);
204
	}
205
	return($string);
206
}
207

    
208
/*
209
 * Converts from various charsets to UTF-8
210
 *
211
 * Will convert a string from various charsets to UTF-8.
212
 * HTML-entities may be converted, too.
213
 * In case of error the returned string is unchanged, and a message is emitted.
214
 * Supported charsets are:
215
 * direct: iso_8859_1 iso_8859_2 iso_8859_3 iso_8859_4 iso_8859_5
216
 *         iso_8859_6 iso_8859_7 iso_8859_8 iso_8859_9 iso_8859_10 iso_8859_11
217
 * mb_convert_encoding: all wb charsets (except those from 'direct'); but not GB2312
218
 * iconv:  all wb charsets (except those from 'direct')
219
 *
220
 * @param  string  A string in supported encoding
221
 * @param  string  The charset to convert from, defaults to DEFAULT_CHARSET
222
 * @return string  A string in UTF-8-encoding, with all entities decoded, too.
223
 *                 String is unchanged in case of error.
224
 * @author thorn
225
 */
226
function charset_to_utf8($str, $charset_in=DEFAULT_CHARSET, $decode_entities=true) {
227
	global $iso_8859_2_to_utf8, $iso_8859_3_to_utf8, $iso_8859_4_to_utf8, $iso_8859_5_to_utf8, $iso_8859_6_to_utf8, $iso_8859_7_to_utf8, $iso_8859_8_to_utf8, $iso_8859_9_to_utf8, $iso_8859_10_to_utf8, $iso_8859_11_to_utf8;
228
	$charset_in = strtoupper($charset_in);
229
	if ($charset_in == "") { $charset_in = 'UTF-8'; }
230
	$wrong_ISO8859 = false;
231
	$converted = false;
232

    
233
	if((!function_exists('iconv') && !UTF8_MBSTRING && ($charset_in=='BIG5' || $charset_in=='ISO-2022-JP' || $charset_in=='ISO-2022-KR')) || (!function_exists('iconv') && $charset_in=='GB2312')) {
234
		// Nothing we can do here :-(
235
		// Charset is one of those obscure ISO-2022... or BIG5, GB2312 or something
236
		// and we can't use mb_convert_encoding() or iconv();
237
		// Emit an error-message.
238
		trigger_error("Can't convert from $charset_in without mb_convert_encoding() or iconv(). Use UTF-8 instead.", E_USER_WARNING);
239
		return($str);
240
	}
241

    
242
	// check if we have UTF-8 or a plain ASCII string
243
	if($charset_in == 'UTF-8' || utf8_isASCII($str)) {
244
		// we have utf-8. Just replace HTML-entities and return
245
		if($decode_entities && preg_match('/&[#0-9a-zA-Z]+;/',$str))
246
			return(utf8_fast_entities_to_umlauts($str));
247
		else // nothing to do
248
			return($str);
249
	}
250
	
251
	// Convert $str to utf8
252
	if(substr($charset_in,0,8) == 'ISO-8859') {
253
		switch($charset_in) {
254
			case 'ISO-8859-1': $str=utf8_encode($str); break;
255
			case 'ISO-8859-2': $str=strtr($str, $iso_8859_2_to_utf8); break;
256
			case 'ISO-8859-3': $str=strtr($str, $iso_8859_3_to_utf8); break;
257
			case 'ISO-8859-4': $str=strtr($str, $iso_8859_4_to_utf8); break;
258
			case 'ISO-8859-5': $str=strtr($str, $iso_8859_5_to_utf8); break;
259
			case 'ISO-8859-6': $str=strtr($str, $iso_8859_6_to_utf8); break;
260
			case 'ISO-8859-7': $str=strtr($str, $iso_8859_7_to_utf8); break;
261
			case 'ISO-8859-8': $str=strtr($str, $iso_8859_8_to_utf8); break;
262
			case 'ISO-8859-9': $str=strtr($str, $iso_8859_9_to_utf8); break;
263
			case 'ISO-8859-10': $str=strtr($str, $iso_8859_10_to_utf8); break;
264
			case 'ISO-8859-11': $str=strtr($str, $iso_8859_11_to_utf8); break;
265
			default: $wrong_ISO8859 = true;
266
		}
267
		if(!$wrong_ISO8859)
268
			$converted = true;
269
	}
270
	if(!$converted && UTF8_MBSTRING && $charset_in != 'GB2312') {
271
		// $charset is neither UTF-8 nor a known ISO-8859...
272
		// Try mb_convert_encoding() - but there's no GB2312 encoding in php's mb_* functions
273
		$str = mb_convert_encoding($str, 'UTF-8', $charset_in);
274
		$converted = true;
275
	} elseif(!$converted) { // Try iconv
276
		if(function_exists('iconv')) {
277
			$str = iconv($charset_in, 'UTF-8', $str);
278
			$converted = true;
279
		}
280
	}
281
	if($converted) {
282
		// we have utf-8, now replace HTML-entities and return
283
		if($decode_entities && preg_match('/&[#0-9a-zA-Z]+;/',$str))
284
			$str = utf8_fast_entities_to_umlauts($str);
285
		return($str);
286
	}
287
	
288
	// Nothing we can do here :-(
289
	// Charset is one of those obscure ISO-2022... or BIG5, GB2312 or something
290
	// and we can't use mb_convert_encoding() or iconv();
291
	// Emit an error-message.
292
	trigger_error("Can't convert from $charset_in without mb_convert_encoding() or iconv(). Use UTF-8 instead.", E_USER_WARNING);
293
	
294
	return $str;
295
}
296

    
297
/*
298
 * Converts from UTF-8 to various charsets
299
 *
300
 * Will convert a string from UTF-8 to various charsets.
301
 * HTML-entities will not! be converted.
302
 * In case of error the returned string is unchanged, and a message is emitted.
303
 * Supported charsets are:
304
 * direct: iso_8859_1 iso_8859_2 iso_8859_3 iso_8859_4 iso_8859_5
305
 *         iso_8859_6 iso_8859_7 iso_8859_8 iso_8859_9 iso_8859_10 iso_8859_11
306
 * mb_convert_encoding: all wb charsets (except those from 'direct'); but not GB2312
307
 * iconv:  all wb charsets (except those from 'direct')
308
 *
309
 * @param  string  An UTF-8 encoded string
310
 * @param  string  The charset to convert to, defaults to DEFAULT_CHARSET
311
 * @return string  A string in a supported encoding, with all entities decoded, too.
312
 *                 String is unchanged in case of error.
313
 * @author thorn
314
 */
315
function utf8_to_charset($str, $charset_out=DEFAULT_CHARSET) {
316
	global $utf8_to_iso_8859_2, $utf8_to_iso_8859_3, $utf8_to_iso_8859_4, $utf8_to_iso_8859_5, $utf8_to_iso_8859_6, $utf8_to_iso_8859_7, $utf8_to_iso_8859_8, $utf8_to_iso_8859_9, $utf8_to_iso_8859_10, $utf8_to_iso_8859_11;
317
	$charset_out = strtoupper($charset_out);
318
	$wrong_ISO8859 = false;
319
	$converted = false;
320

    
321
	if((!function_exists('iconv') && !UTF8_MBSTRING && ($charset_out=='BIG5' || $charset_out=='ISO-2022-JP' || $charset_out=='ISO-2022-KR')) || (!function_exists('iconv') && $charset_out=='GB2312')) {
322
		// Nothing we can do here :-(
323
		// Charset is one of those obscure ISO-2022... or BIG5, GB2312 or something
324
		// and we can't use mb_convert_encoding() or iconv();
325
		// Emit an error-message.
326
		trigger_error("Can't convert into $charset_out without mb_convert_encoding() or iconv(). Use UTF-8 instead.", E_USER_WARNING);
327
		return($str);
328
	}
329
	
330
	// the string comes from charset_to_utf8(), so we can skip this
331
	// replace HTML-entities first
332
	//if(preg_match('/&[#0-9a-zA-Z]+;/',$str))
333
	//	$str = utf8_entities_to_umlauts($str);
334
	
335
	// check if we need to convert
336
	if($charset_out == 'UTF-8' || utf8_isASCII($str)) {
337
		// Nothing to do. Just return
338
			return($str);
339
	}
340
	
341
	// Convert $str to $charset_out
342
	if(substr($charset_out,0,8) == 'ISO-8859') {
343
		switch($charset_out) {
344
			case 'ISO-8859-1': $str=utf8_decode($str); break;
345
			case 'ISO-8859-2': $str=strtr($str, $utf8_to_iso_8859_2); break;
346
			case 'ISO-8859-3': $str=strtr($str, $utf8_to_iso_8859_3); break;
347
			case 'ISO-8859-4': $str=strtr($str, $utf8_to_iso_8859_4); break;
348
			case 'ISO-8859-5': $str=strtr($str, $utf8_to_iso_8859_5); break;
349
			case 'ISO-8859-6': $str=strtr($str, $utf8_to_iso_8859_6); break;
350
			case 'ISO-8859-7': $str=strtr($str, $utf8_to_iso_8859_7); break;
351
			case 'ISO-8859-8': $str=strtr($str, $utf8_to_iso_8859_8); break;
352
			case 'ISO-8859-9': $str=strtr($str, $utf8_to_iso_8859_9); break;
353
			case 'ISO-8859-10': $str=strtr($str, $utf8_to_iso_8859_10); break;
354
			case 'ISO-8859-11': $str=strtr($str, $utf8_to_iso_8859_11); break;
355
			default: $wrong_ISO8859 = true;
356
		}
357
		if(!$wrong_ISO8859)
358
			$converted = true;
359
	}
360
	if(!$converted && UTF8_MBSTRING && $charset_out != 'GB2312') {
361
		// $charset is neither UTF-8 nor a known ISO-8859...
362
		// Try mb_convert_encoding() - but there's no GB2312 encoding in php's mb_* functions
363
		$str = mb_convert_encoding($str, $charset_out, 'UTF-8');
364
		$converted = true;
365
	} elseif(!$converted) { // Try iconv
366
		if(function_exists('iconv')) {
367
			$str = iconv('UTF-8', $charset_out, $str);
368
			$converted = true;
369
		}
370
	}
371
	if($converted) {
372
		return($str);
373
	}
374
	
375
	// Nothing we can do here :-(
376
	// Charset is one of those obscure ISO-2022... or BIG5, GB2312 or something
377
	// and we can't use mb_convert_encoding() or iconv();
378
	// Emit an error-message.
379
	trigger_error("Can't convert into $charset_out without mb_convert_encoding() or iconv(). Use UTF-8 instead.", E_USER_WARNING);
380
	
381
	return $str;
382
}
383

    
384
/*
385
 * convert Filenames to ASCII
386
 *
387
 * Convert all non-ASCII characters and all HTML-entities to their plain 7bit equivalents
388
 * Characters without an equivalent will be converted to hex-values.
389
 * The name entities_to_7bit() is somewhat misleading, but kept for compatibility-reasons.
390
 *
391
 * @param  string  Filename to convert (all encodings from charset_to_utf8() are allowed)
392
 * @return string  ASCII encoded string, to use as filename in wb's page_filename() and media_filename
393
 * @author thorn
394
 */
395
function entities_to_7bit($str) {
396
	// convert to UTF-8
397
	$str = charset_to_utf8($str);
398
	if(!utf8_check($str))
399
		return($str);
400
	// replace some specials
401
	$str = utf8_stripspecials($str, '_');
402
	// translate non-ASCII characters to ASCII
403
	$str = utf8_romanize($str);
404
	// missed some? - Many UTF-8-chars can't be romanized
405
	// convert to HTML-entities, and replace entites by hex-numbers
406
	$str = utf8_fast_umlauts_to_entities($str, false);
407
	$str = str_replace('&#039;', '&apos;', $str);
408
	$str = preg_replace('/&#([0-9]+);/e', "dechex('$1')",  $str);
409
	// maybe there are some &gt; &lt; &apos; &quot; &amp; &nbsp; left, replace them too
410
	$str = str_replace(array('&gt;', '&lt;', '&apos;', '\'', '&quot;', '&amp;'), '', $str);
411
	$str = str_replace('&amp;', '', $str);
412
	
413
	return($str);
414
}
415

    
416
/*
417
 * Convert a string from mixed html-entities/umlauts to pure $charset_out-umlauts
418
 * 
419
 * Will replace all numeric and named entities except
420
 * &gt; &lt; &apos; &quot; &#039; &nbsp;
421
 * @author thorn
422
 */
423
function entities_to_umlauts2($string, $charset_out=DEFAULT_CHARSET) {
424
	$string = charset_to_utf8($string, DEFAULT_CHARSET, true);
425
	//if(utf8_check($string)) // this check is to much time-consuming (this may fail only if AddDefaultCharset is set)
426
		$string = utf8_to_charset($string, $charset_out);
427
	return ($string);
428
}
429

    
430
/*
431
 * Convert a string from mixed html-entities/umlauts to pure ASCII with HTML-entities
432
 * 
433
 * Will convert a string in $charset_in encoding to a pure ASCII string with HTML-entities.
434
 * @author thorn
435
 */
436
function umlauts_to_entities2($string, $charset_in=DEFAULT_CHARSET) {
437
	$string = charset_to_utf8($string, $charset_in, false);
438
	//if(utf8_check($string)) // this check is to much time-consuming (this may fail only if AddDefaultCharset is set)
439
		$string = utf8_fast_umlauts_to_entities($string, false);
440
	return($string);
441
}
442

    
443
?>
(10-10/14)