PageRenderTime 52ms CodeModel.GetById 14ms RepoModel.GetById 1ms app.codeStats 0ms

/moodle/lib/typo3/class.t3lib_cs.php

#
PHP | 2348 lines | 1521 code | 212 blank | 615 comment | 374 complexity | 5c4344bd2d09886d246b3632c4de121d MD5 | raw file
Possible License(s): GPL-3.0, LGPL-2.1, BSD-3-Clause, AGPL-3.0, MPL-2.0-no-copyleft-exception, LGPL-3.0, Apache-2.0

Large files files are truncated, but you can click here to view the full file

    

  1. <?php
    2. 

  2. /***************************************************************
    3. 

  3.  *  Copyright notice
    4. 

  4.  *
    5. 

  5.  *  (c) 2003-2011 Kasper Skĺrhřj (kasperYYYY@typo3.com)
    6. 

  6.  *  All rights reserved
    7. 

  7.  *
    8. 

  8.  *  This script is part of the Typo3 project. The Typo3 project is
    9. 

  9.  *  free software; you can redistribute it and/or modify
    10. 

  10.  *  it under the terms of the GNU General Public License as published by
    11. 

  11.  *  the Free Software Foundation; either version 2 of the License, or
    12. 

  12.  *  (at your option) any later version.
    13. 

  13.  *
    14. 

  14.  *  The GNU General Public License can be found at
    15. 

  15.  *  http://www.gnu.org/copyleft/gpl.html.
    16. 

  16.  *
    17. 

  17.  *  This script is distributed in the hope that it will be useful,
    18. 

  18.  *  but WITHOUT ANY WARRANTY; without even the implied warranty of
    19. 

  19.  *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    20. 

  20.  *  GNU General Public License for more details.
    21. 

  21.  *
    22. 

  22.  *  This copyright notice MUST APPEAR in all copies of the script!
    23. 

  23.  ***************************************************************/
    24. 

  24. /**
    25. 

  25.  * Class for conversion between charsets.
    26. 

  26.  *
    27. 

  27.  * $Id: class.t3lib_cs.php,v 1.15 2011/10/27 17:26:27 moodlerobot Exp $
    28. 

  28.  *
    29. 

  29.  * @author	Kasper Skĺrhřj <kasperYYYY@typo3.com>
    30. 

  30.  * @author	Martin Kutschker <martin.t.kutschker@blackbox.net>
    31. 

  31.  */
    32. 

  32. /**
    33. 

  33.  * [CLASS/FUNCTION INDEX of SCRIPT]
    34. 

  34.  *
    35. 

  35.  *
    36. 

  36.  *
    37. 

  37.  *  136: class t3lib_cs
    38. 

  38.  *  488:	 function parse_charset($charset)
    39. 

  39.  *  507:	 function get_locale_charset($locale)
    40. 

  40.  *
    41. 

  41.  *			  SECTION: Charset Conversion functions
    42. 

  42.  *  560:	 function conv($str,$fromCS,$toCS,$useEntityForNoChar=0)
    43. 

  43.  *  600:	 function convArray(&$array,$fromCS,$toCS,$useEntityForNoChar=0)
    44. 

  44.  *  617:	 function utf8_encode($str,$charset)
    45. 

  45.  *  663:	 function utf8_decode($str,$charset,$useEntityForNoChar=0)
    46. 

  46.  *  706:	 function utf8_to_entities($str)
    47. 

  47.  *  739:	 function entities_to_utf8($str,$alsoStdHtmlEnt=0)
    48. 

  48.  *  773:	 function utf8_to_numberarray($str,$convEntities=0,$retChar=0)
    49. 

  49.  *  823:	 function UnumberToChar($cbyte)
    50. 

  50.  *  868:	 function utf8CharToUnumber($str,$hex=0)
    51. 

  51.  *
    52. 

  52.  *			  SECTION: Init functions
    53. 

  53.  *  911:	 function initCharset($charset)
    54. 

  54.  *  973:	 function initUnicodeData($mode=null)
    55. 

  55.  * 1198:	 function initCaseFolding($charset)
    56. 

  56.  * 1260:	 function initToASCII($charset)
    57. 

  57.  *
    58. 

  58.  *			  SECTION: String operation functions
    59. 

  59.  * 1331:	 function substr($charset,$string,$start,$len=null)
    60. 

  60.  * 1384:	 function strlen($charset,$string)
    61. 

  61.  * 1414:	 function crop($charset,$string,$len,$crop='')
    62. 

  62.  * 1467:	 function strtrunc($charset,$string,$len)
    63. 

  63.  * 1501:	 function conv_case($charset,$string,$case)
    64. 

  64.  * 1527:	 function specCharsToASCII($charset,$string)
    65. 

  65.  *
    66. 

  66.  *			  SECTION: Internal string operation functions
    67. 

  67.  * 1567:	 function sb_char_mapping($str,$charset,$mode,$opt='')
    68. 

  68.  *
    69. 

  69.  *			  SECTION: Internal UTF-8 string operation functions
    70. 

  70.  * 1622:	 function utf8_substr($str,$start,$len=null)
    71. 

  71.  * 1655:	 function utf8_strlen($str)
    72. 

  72.  * 1676:	 function utf8_strtrunc($str,$len)
    73. 

  73.  * 1698:	 function utf8_strpos($haystack,$needle,$offset=0)
    74. 

  74.  * 1723:	 function utf8_strrpos($haystack,$needle)
    75. 

  75.  * 1745:	 function utf8_char2byte_pos($str,$pos)
    76. 

  76.  * 1786:	 function utf8_byte2char_pos($str,$pos)
    77. 

  77.  * 1809:	 function utf8_char_mapping($str,$mode,$opt='')
    78. 

  78.  *
    79. 

  79.  *			  SECTION: Internal EUC string operation functions
    80. 

  80.  * 1885:	 function euc_strtrunc($str,$len,$charset)
    81. 

  81.  * 1914:	 function euc_substr($str,$start,$charset,$len=null)
    82. 

  82.  * 1939:	 function euc_strlen($str,$charset)
    83. 

  83.  * 1966:	 function euc_char2byte_pos($str,$pos,$charset)
    84. 

  84.  * 2007:	 function euc_char_mapping($str,$charset,$mode,$opt='')
    85. 

  85.  *
    86. 

  86.  * TOTAL FUNCTIONS: 35
    87. 

  87.  * (This index is automatically created/updated by the extension "extdeveval")
    88. 

  88.  *
    89. 

  89.  */
    90. 

  90. 

  91. 

  92. /**
    93. 

  93.  * Notes on UTF-8
    94. 

  94.  *
    95. 

  95.  * Functions working on UTF-8 strings:
    96. 

  96.  *
    97. 

  97.  * - strchr/strstr
    98. 

  98.  * - strrchr
    99. 

  99.  * - substr_count
    100. 

  100.  * - implode/explode/join
    101. 

  101.  *
    102. 

  102.  * Functions nearly working on UTF-8 strings:
    103. 

  103.  *
    104. 

  104.  * - strlen: returns the length in BYTES, if you need the length in CHARACTERS use utf8_strlen
    105. 

  105.  * - trim/ltrim/rtrim: the second parameter 'charlist' won't work for characters not contained in 7-bit ASCII
    106. 

  106.  * - strpos/strrpos: they return the BYTE position, if you need the CHARACTER position use utf8_strpos/utf8_strrpos
    107. 

  107.  * - htmlentities: charset support for UTF-8 only since PHP 4.3.0
    108. 

  108.  * - preg_*: Support compiled into PHP by default nowadays, but could be unavailable, need to use modifier
    109. 

  109.  *
    110. 

  110.  * Functions NOT working on UTF-8 strings:
    111. 

  111.  *
    112. 

  112.  * - str*cmp
    113. 

  113.  * - stristr
    114. 

  114.  * - stripos
    115. 

  115.  * - substr
    116. 

  116.  * - strrev
    117. 

  117.  * - split/spliti
    118. 

  118.  * - ...
    119. 

  119.  *
    120. 

  120.  */
    121. 

  121. /**
    122. 

  122.  * Class for conversion between charsets
    123. 

  123.  *
    124. 

  124.  * @author	Kasper Skĺrhřj <kasperYYYY@typo3.com>
    125. 

  125.  * @author	Martin Kutschker <martin.t.kutschker@blackbox.net>
    126. 

  126.  * @package TYPO3
    127. 

  127.  * @subpackage t3lib
    128. 

  128.  */
    129. 

  129. class t3lib_cs {
    130. 

  130. 	var $noCharByteVal = 63; // ASCII Value for chars with no equivalent.
    131. 

  131. 

  132. 		// This is the array where parsed conversion tables are stored (cached)
    133. 

  133. 	var $parsedCharsets = array();
    134. 

  134. 

  135. 		// An array where case folding data will be stored (cached)
    136. 

  136. 	var $caseFolding = array();
    137. 

  137. 

  138. 		// An array where charset-to-ASCII mappings are stored (cached)
    139. 

  139. 	var $toASCII = array();
    140. 

  140. 

  141. 		// This tells the converter which charsets has two bytes per char:
    142. 

  142. 	var $twoByteSets = array(
    143. 

  143. 		'ucs-2' => 1, // 2-byte Unicode
    144. 

  144. 	);
    145. 

  145. 

  146. 		// This tells the converter which charsets has four bytes per char:
    147. 

  147. 	var $fourByteSets = array(
    148. 

  148. 		'ucs-4' => 1, // 4-byte Unicode
    149. 

  149. 		'utf-32' => 1, // 4-byte Unicode (limited to the 21-bits of UTF-16)
    150. 

  150. 	);
    151. 

  151. 

  152. 		// This tells the converter which charsets use a scheme like the Extended Unix Code:
    153. 

  153. 	var $eucBasedSets = array(
    154. 

  154. 		'gb2312' => 1, // Chinese, simplified.
    155. 

  155. 		'big5' => 1, // Chinese, traditional.
    156. 

  156. 		'euc-kr' => 1, // Korean
    157. 

  157. 		'shift_jis' => 1, // Japanese - WARNING: Shift-JIS includes half-width katakana single-bytes characters above 0x80!
    158. 

  158. 	);
    159. 

  159. 

  160. 		// see	http://developer.apple.com/documentation/macos8/TextIntlSvcs/TextEncodingConversionManager/TEC1.5/TEC.b0.html
    161. 

  161. 		// http://czyborra.com/charsets/iso8859.html
    162. 

  162. 	var $synonyms = array(
    163. 

  163. 		'us' => 'ascii',
    164. 

  164. 		'us-ascii' => 'ascii',
    165. 

  165. 		'cp819' => 'iso-8859-1',
    166. 

  166. 		'ibm819' => 'iso-8859-1',
    167. 

  167. 		'iso-ir-100' => 'iso-8859-1',
    168. 

  168. 		'iso-ir-101' => 'iso-8859-2',
    169. 

  169. 		'iso-ir-109' => 'iso-8859-3',
    170. 

  170. 		'iso-ir-110' => 'iso-8859-4',
    171. 

  171. 		'iso-ir-144' => 'iso-8859-5',
    172. 

  172. 		'iso-ir-127' => 'iso-8859-6',
    173. 

  173. 		'iso-ir-126' => 'iso-8859-7',
    174. 

  174. 		'iso-ir-138' => 'iso-8859-8',
    175. 

  175. 		'iso-ir-148' => 'iso-8859-9',
    176. 

  176. 		'iso-ir-157' => 'iso-8859-10',
    177. 

  177. 		'iso-ir-179' => 'iso-8859-13',
    178. 

  178. 		'iso-ir-199' => 'iso-8859-14',
    179. 

  179. 		'iso-ir-203' => 'iso-8859-15',
    180. 

  180. 		'csisolatin1' => 'iso-8859-1',
    181. 

  181. 		'csisolatin2' => 'iso-8859-2',
    182. 

  182. 		'csisolatin3' => 'iso-8859-3',
    183. 

  183. 		'csisolatin5' => 'iso-8859-9',
    184. 

  184. 		'csisolatin8' => 'iso-8859-14',
    185. 

  185. 		'csisolatin9' => 'iso-8859-15',
    186. 

  186. 		'csisolatingreek' => 'iso-8859-7',
    187. 

  187. 		'iso-celtic' => 'iso-8859-14',
    188. 

  188. 		'latin1' => 'iso-8859-1',
    189. 

  189. 		'latin2' => 'iso-8859-2',
    190. 

  190. 		'latin3' => 'iso-8859-3',
    191. 

  191. 		'latin5' => 'iso-8859-9',
    192. 

  192. 		'latin6' => 'iso-8859-10',
    193. 

  193. 		'latin8' => 'iso-8859-14',
    194. 

  194. 		'latin9' => 'iso-8859-15',
    195. 

  195. 		'l1' => 'iso-8859-1',
    196. 

  196. 		'l2' => 'iso-8859-2',
    197. 

  197. 		'l3' => 'iso-8859-3',
    198. 

  198. 		'l5' => 'iso-8859-9',
    199. 

  199. 		'l6' => 'iso-8859-10',
    200. 

  200. 		'l8' => 'iso-8859-14',
    201. 

  201. 		'l9' => 'iso-8859-15',
    202. 

  202. 		'cyrillic' => 'iso-8859-5',
    203. 

  203. 		'arabic' => 'iso-8859-6',
    204. 

  204. 		'tis-620' => 'iso-8859-11',
    205. 

  205. 		'win874' => 'windows-874',
    206. 

  206. 		'win1250' => 'windows-1250',
    207. 

  207. 		'win1251' => 'windows-1251',
    208. 

  208. 		'win1252' => 'windows-1252',
    209. 

  209. 		'win1253' => 'windows-1253',
    210. 

  210. 		'win1254' => 'windows-1254',
    211. 

  211. 		'win1255' => 'windows-1255',
    212. 

  212. 		'win1256' => 'windows-1256',
    213. 

  213. 		'win1257' => 'windows-1257',
    214. 

  214. 		'win1258' => 'windows-1258',
    215. 

  215. 		'cp1250' => 'windows-1250',
    216. 

  216. 		'cp1251' => 'windows-1251',
    217. 

  217. 		'cp1252' => 'windows-1252',
    218. 

  218. 		'ms-ee' => 'windows-1250',
    219. 

  219. 		'ms-ansi' => 'windows-1252',
    220. 

  220. 		'ms-greek' => 'windows-1253',
    221. 

  221. 		'ms-turk' => 'windows-1254',
    222. 

  222. 		'winbaltrim' => 'windows-1257',
    223. 

  223. 		'koi-8ru' => 'koi-8r',
    224. 

  224. 		'koi8r' => 'koi-8r',
    225. 

  225. 		'cp878' => 'koi-8r',
    226. 

  226. 		'mac' => 'macroman',
    227. 

  227. 		'macintosh' => 'macroman',
    228. 

  228. 		'euc-cn' => 'gb2312',
    229. 

  229. 		'x-euc-cn' => 'gb2312',
    230. 

  230. 		'euccn' => 'gb2312',
    231. 

  231. 		'cp936' => 'gb2312',
    232. 

  232. 		'big-5' => 'big5',
    233. 

  233. 		'cp950' => 'big5',
    234. 

  234. 		'eucjp' => 'euc-jp',
    235. 

  235. 		'sjis' => 'shift_jis',
    236. 

  236. 		'shift-jis' => 'shift_jis',
    237. 

  237. 		'cp932' => 'shift_jis',
    238. 

  238. 		'cp949' => 'euc-kr',
    239. 

  239. 		'utf7' => 'utf-7',
    240. 

  240. 		'utf8' => 'utf-8',
    241. 

  241. 		'utf16' => 'utf-16',
    242. 

  242. 		'utf32' => 'utf-32',
    243. 

  243. 		'utf8' => 'utf-8',
    244. 

  244. 		'ucs2' => 'ucs-2',
    245. 

  245. 		'ucs4' => 'ucs-4',
    246. 

  246. 	);
    247. 

  247. 

  248. 		// mapping of iso-639-1 language codes to script names
    249. 

  249. 	var $lang_to_script = array(
    250. 

  250. 			// iso-639-1 language codes, see http://www.loc.gov/standards/iso639-2/php/code_list.php
    251. 

  251. 		'ar' => 'arabic',
    252. 

  252. 		'bg' => 'cyrillic', // Bulgarian
    253. 

  253. 		'bs' => 'east_european', // Bosnian
    254. 

  254. 		'cs' => 'east_european', // Czech
    255. 

  255. 		'da' => 'west_european', // Danish
    256. 

  256. 		'de' => 'west_european', // German
    257. 

  257. 		'es' => 'west_european', // Spanish
    258. 

  258. 		'et' => 'estonian',
    259. 

  259. 		'eo' => 'unicode', // Esperanto
    260. 

  260. 		'eu' => 'west_european', // Basque
    261. 

  261. 		'fa' => 'arabic', // Persian
    262. 

  262. 		'fi' => 'west_european', // Finish
    263. 

  263. 		'fo' => 'west_european', // Faroese
    264. 

  264. 		'fr' => 'west_european', // French
    265. 

  265. 		'ga' => 'west_european', // Irish
    266. 

  266. 		'gl' => 'west_european', // Galician
    267. 

  267. 		'gr' => 'greek',
    268. 

  268. 		'he' => 'hebrew', // Hebrew (since 1998)
    269. 

  269. 		'hi' => 'unicode', // Hindi
    270. 

  270. 		'hr' => 'east_european', // Croatian
    271. 

  271. 		'hu' => 'east_european', // Hungarian
    272. 

  272. 		'iw' => 'hebrew', // Hebrew (til 1998)
    273. 

  273. 		'is' => 'west_european', // Icelandic
    274. 

  274. 		'it' => 'west_european', // Italian
    275. 

  275. 		'ja' => 'japanese',
    276. 

  276. 		'ka' => 'unicode', // Georgian
    277. 

  277. 		'kl' => 'west_european', // Greenlandic
    278. 

  278. 		'km' => 'unicode', // Khmer
    279. 

  279. 		'ko' => 'korean',
    280. 

  280. 		'lt' => 'lithuanian',
    281. 

  281. 		'lv' => 'west_european', // Latvian/Lettish
    282. 

  282. 		'nl' => 'west_european', // Dutch
    283. 

  283. 		'no' => 'west_european', // Norwegian
    284. 

  284. 		'nb' => 'west_european', // Norwegian Bokmal
    285. 

  285. 		'nn' => 'west_european', // Norwegian Nynorsk
    286. 

  286. 		'pl' => 'east_european', // Polish
    287. 

  287. 		'pt' => 'west_european', // Portuguese
    288. 

  288. 		'ro' => 'east_european', // Romanian
    289. 

  289. 		'ru' => 'cyrillic', // Russian
    290. 

  290. 		'sk' => 'east_european', // Slovak
    291. 

  291. 		'sl' => 'east_european', // Slovenian
    292. 

  292. 		'sr' => 'cyrillic', // Serbian
    293. 

  293. 		'sv' => 'west_european', // Swedish
    294. 

  294. 		'sq' => 'albanian', // Albanian
    295. 

  295. 		'th' => 'thai',
    296. 

  296. 		'uk' => 'cyrillic', // Ukranian
    297. 

  297. 		'vi' => 'vietnamese',
    298. 

  298. 		'zh' => 'chinese',
    299. 

  299. 			// MS language codes, see http://msdn.microsoft.com/library/default.asp?url=/library/en-us/vclib/html/_crt_language_strings.asp
    300. 

  300. 			// http://msdn.microsoft.com/library/default.asp?url=/library/en-us/wceinternational5/html/wce50conLanguageIdentifiersandLocales.asp
    301. 

  301. 		'ara' => 'arabic',
    302. 

  302. 		'bgr' => 'cyrillic', // Bulgarian
    303. 

  303. 		'cat' => 'west_european', // Catalan
    304. 

  304. 		'chs' => 'simpl_chinese',
    305. 

  305. 		'cht' => 'trad_chinese',
    306. 

  306. 		'csy' => 'east_european', // Czech
    307. 

  307. 		'dan' => 'west_european', // Danisch
    308. 

  308. 		'deu' => 'west_european', // German
    309. 

  309. 		'dea' => 'west_european', // German (Austrian)
    310. 

  310. 		'des' => 'west_european', // German (Swiss)
    311. 

  311. 		'ena' => 'west_european', // English (Australian)
    312. 

  312. 		'enc' => 'west_european', // English (Canadian)
    313. 

  313. 		'eng' => 'west_european', // English
    314. 

  314. 		'enz' => 'west_european', // English (New Zealand)
    315. 

  315. 		'enu' => 'west_european', // English (United States)
    316. 

  316. 		'euq' => 'west_european', // Basque
    317. 

  317. 		'fos' => 'west_european', // Faroese
    318. 

  318. 		'far' => 'arabic', // Persian
    319. 

  319. 		'fin' => 'west_european', // Finish
    320. 

  320. 		'fra' => 'west_european', // French
    321. 

  321. 		'frb' => 'west_european', // French (Belgian)
    322. 

  322. 		'frc' => 'west_european', // French (Canadian)
    323. 

  323. 		'frs' => 'west_european', // French (Swiss)
    324. 

  324. 		'geo' => 'unicode', // Georgian
    325. 

  325. 		'glg' => 'west_european', // Galician
    326. 

  326. 		'ell' => 'greek',
    327. 

  327. 		'heb' => 'hebrew',
    328. 

  328. 		'hin' => 'unicode', // Hindi
    329. 

  329. 		'hun' => 'east_european', // Hungarian
    330. 

  330. 		'isl' => 'west_euorpean', // Icelandic
    331. 

  331. 		'ita' => 'west_european', // Italian
    332. 

  332. 		'its' => 'west_european', // Italian (Swiss)
    333. 

  333. 		'jpn' => 'japanese',
    334. 

  334. 		'khm' => 'unicode', // Khmer
    335. 

  335. 		'kor' => 'korean',
    336. 

  336. 		'lth' => 'lithuanian',
    337. 

  337. 		'lvi' => 'west_european', // Latvian/Lettish
    338. 

  338. 		'msl' => 'west_european', // Malay
    339. 

  339. 		'nlb' => 'west_european', // Dutch (Belgian)
    340. 

  340. 		'nld' => 'west_european', // Dutch
    341. 

  341. 		'nor' => 'west_european', // Norwegian (bokmal)
    342. 

  342. 		'non' => 'west_european', // Norwegian (nynorsk)
    343. 

  343. 		'plk' => 'east_european', // Polish
    344. 

  344. 		'ptg' => 'west_european', // Portuguese
    345. 

  345. 		'ptb' => 'west_european', // Portuguese (Brazil)
    346. 

  346. 		'rom' => 'east_european', // Romanian
    347. 

  347. 		'rus' => 'cyrillic', // Russian
    348. 

  348. 		'slv' => 'east_european', // Slovenian
    349. 

  349. 		'sky' => 'east_european', // Slovak
    350. 

  350. 		'srl' => 'east_european', // Serbian (Latin)
    351. 

  351. 		'srb' => 'cyrillic', // Serbian (Cyrillic)
    352. 

  352. 		'esp' => 'west_european', // Spanish (trad. sort)
    353. 

  353. 		'esm' => 'west_european', // Spanish (Mexican)
    354. 

  354. 		'esn' => 'west_european', // Spanish (internat. sort)
    355. 

  355. 		'sve' => 'west_european', // Swedish
    356. 

  356. 		'sqi' => 'albanian', // Albanian
    357. 

  357. 		'tha' => 'thai',
    358. 

  358. 		'trk' => 'turkish',
    359. 

  359. 		'ukr' => 'cyrillic', // Ukrainian
    360. 

  360. 			// English language names
    361. 

  361. 		'albanian' => 'albanian',
    362. 

  362. 		'arabic' => 'arabic',
    363. 

  363. 		'basque' => 'west_european',
    364. 

  364. 		'bosnian' => 'east_european',
    365. 

  365. 		'bulgarian' => 'east_european',
    366. 

  366. 		'catalan' => 'west_european',
    367. 

  367. 		'croatian' => 'east_european',
    368. 

  368. 		'czech' => 'east_european',
    369. 

  369. 		'danish' => 'west_european',
    370. 

  370. 		'dutch' => 'west_european',
    371. 

  371. 		'english' => 'west_european',
    372. 

  372. 		'esperanto' => 'unicode',
    373. 

  373. 		'estonian' => 'estonian',
    374. 

  374. 		'faroese' => 'west_european',
    375. 

  375. 		'farsi' => 'arabic',
    376. 

  376. 		'finnish' => 'west_european',
    377. 

  377. 		'french' => 'west_european',
    378. 

  378. 		'galician' => 'west_european',
    379. 

  379. 		'georgian' => 'unicode',
    380. 

  380. 		'german' => 'west_european',
    381. 

  381. 		'greek' => 'greek',
    382. 

  382. 		'greenlandic' => 'west_european',
    383. 

  383. 		'hebrew' => 'hebrew',
    384. 

  384. 		'hindi' => 'unicode',
    385. 

  385. 		'hungarian' => 'east_european',
    386. 

  386. 		'icelandic' => 'west_european',
    387. 

  387. 		'italian' => 'west_european',
    388. 

  388. 		'khmer' => 'unicode',
    389. 

  389. 		'latvian' => 'west_european',
    390. 

  390. 		'lettish' => 'west_european',
    391. 

  391. 		'lithuanian' => 'lithuanian',
    392. 

  392. 		'malay' => 'west_european',
    393. 

  393. 		'norwegian' => 'west_european',
    394. 

  394. 		'persian' => 'arabic',
    395. 

  395. 		'polish' => 'east_european',
    396. 

  396. 		'portuguese' => 'west_european',
    397. 

  397. 		'russian' => 'cyrillic',
    398. 

  398. 		'romanian' => 'east_european',
    399. 

  399. 		'serbian' => 'cyrillic',
    400. 

  400. 		'slovak' => 'east_european',
    401. 

  401. 		'slovenian' => 'east_european',
    402. 

  402. 		'spanish' => 'west_european',
    403. 

  403. 		'svedish' => 'west_european',
    404. 

  404. 		'that' => 'thai',
    405. 

  405. 		'turkish' => 'turkish',
    406. 

  406. 		'ukrainian' => 'cyrillic',
    407. 

  407. 	);
    408. 

  408. 

  409. 		// mapping of language (family) names to charsets on Unix
    410. 

  410. 	var $script_to_charset_unix = array(
    411. 

  411. 		'west_european' => 'iso-8859-1',
    412. 

  412. 		'estonian' => 'iso-8859-1',
    413. 

  413. 		'east_european' => 'iso-8859-2',
    414. 

  414. 		'baltic' => 'iso-8859-4',
    415. 

  415. 		'cyrillic' => 'iso-8859-5',
    416. 

  416. 		'arabic' => 'iso-8859-6',
    417. 

  417. 		'greek' => 'iso-8859-7',
    418. 

  418. 		'hebrew' => 'iso-8859-8',
    419. 

  419. 		'turkish' => 'iso-8859-9',
    420. 

  420. 		'thai' => 'iso-8859-11', // = TIS-620
    421. 

  421. 		'lithuanian' => 'iso-8859-13',
    422. 

  422. 		'chinese' => 'gb2312', // = euc-cn
    423. 

  423. 		'japanese' => 'euc-jp',
    424. 

  424. 		'korean' => 'euc-kr',
    425. 

  425. 		'simpl_chinese' => 'gb2312',
    426. 

  426. 		'trad_chinese' => 'big5',
    427. 

  427. 		'vietnamese' => '',
    428. 

  428. 		'unicode' => 'utf-8',
    429. 

  429. 		'albanian' => 'utf-8'
    430. 

  430. 	);
    431. 

  431. 

  432. 		// mapping of language (family) names to charsets on Windows
    433. 

  433. 	var $script_to_charset_windows = array(
    434. 

  434. 		'east_european' => 'windows-1250',
    435. 

  435. 		'cyrillic' => 'windows-1251',
    436. 

  436. 		'west_european' => 'windows-1252',
    437. 

  437. 		'greek' => 'windows-1253',
    438. 

  438. 		'turkish' => 'windows-1254',
    439. 

  439. 		'hebrew' => 'windows-1255',
    440. 

  440. 		'arabic' => 'windows-1256',
    441. 

  441. 		'baltic' => 'windows-1257',
    442. 

  442. 		'estonian' => 'windows-1257',
    443. 

  443. 		'lithuanian' => 'windows-1257',
    444. 

  444. 		'vietnamese' => 'windows-1258',
    445. 

  445. 		'thai' => 'cp874',
    446. 

  446. 		'korean' => 'cp949',
    447. 

  447. 		'chinese' => 'gb2312',
    448. 

  448. 		'japanese' => 'shift_jis',
    449. 

  449. 		'simpl_chinese' => 'gb2312',
    450. 

  450. 		'trad_chinese' => 'big5',
    451. 

  451. 		'albanian' => 'windows-1250',
    452. 

  452. 		'unicode' => 'utf-8'
    453. 

  453. 	);
    454. 

  454. 

  455. 		// mapping of locale names to charsets
    456. 

  456. 	var $locale_to_charset = array(
    457. 

  457. 		'japanese.euc' => 'euc-jp',
    458. 

  458. 		'ja_jp.ujis' => 'euc-jp',
    459. 

  459. 		'korean.euc' => 'euc-kr',
    460. 

  460. 		'sr@Latn' => 'iso-8859-2',
    461. 

  461. 		'zh_cn' => 'gb2312',
    462. 

  462. 		'zh_hk' => 'big5',
    463. 

  463. 		'zh_tw' => 'big5',
    464. 

  464. 	);
    465. 

  465. 

  466. 		// TYPO3 specific: Array with the system charsets used for each system language in TYPO3:
    467. 

  467. 		// Empty values means "iso-8859-1"
    468. 

  468. 	var $charSetArray = array(
    469. 

  469. 		'dk' => '',
    470. 

  470. 		'de' => '',
    471. 

  471. 		'no' => '',
    472. 

  472. 		'it' => '',
    473. 

  473. 		'fr' => '',
    474. 

  474. 		'es' => '',
    475. 

  475. 		'nl' => '',
    476. 

  476. 		'cz' => 'windows-1250',
    477. 

  477. 		'pl' => 'iso-8859-2',
    478. 

  478. 		'si' => 'windows-1250',
    479. 

  479. 		'fi' => '',
    480. 

  480. 		'tr' => 'iso-8859-9',
    481. 

  481. 		'se' => '',
    482. 

  482. 		'pt' => '',
    483. 

  483. 		'ru' => 'windows-1251',
    484. 

  484. 		'ro' => 'iso-8859-2',
    485. 

  485. 		'ch' => 'gb2312',
    486. 

  486. 		'sk' => 'windows-1250',
    487. 

  487. 		'lt' => 'windows-1257',
    488. 

  488. 		'is' => 'utf-8',
    489. 

  489. 		'hr' => 'windows-1250',
    490. 

  490. 		'hu' => 'iso-8859-2',
    491. 

  491. 		'gl' => '',
    492. 

  492. 		'th' => 'iso-8859-11',
    493. 

  493. 		'gr' => 'iso-8859-7',
    494. 

  494. 		'hk' => 'big5',
    495. 

  495. 		'eu' => '',
    496. 

  496. 		'bg' => 'windows-1251',
    497. 

  497. 		'br' => '',
    498. 

  498. 		'et' => 'iso-8859-4',
    499. 

  499. 		'ar' => 'iso-8859-6',
    500. 

  500. 		'he' => 'utf-8',
    501. 

  501. 		'ua' => 'windows-1251',
    502. 

  502. 		'jp' => 'shift_jis',
    503. 

  503. 		'lv' => 'utf-8',
    504. 

  504. 		'vn' => 'utf-8',
    505. 

  505. 		'ca' => 'iso-8859-15',
    506. 

  506. 		'ba' => 'iso-8859-2',
    507. 

  507. 		'kr' => 'euc-kr',
    508. 

  508. 		'eo' => 'utf-8',
    509. 

  509. 		'my' => '',
    510. 

  510. 		'hi' => 'utf-8',
    511. 

  511. 		'fo' => 'utf-8',
    512. 

  512. 		'fa' => 'utf-8',
    513. 

  513. 		'sr' => 'utf-8',
    514. 

  514. 		'sq' => 'utf-8',
    515. 

  515. 		'ge' => 'utf-8',
    516. 

  516. 		'ga' => '',
    517. 

  517. 		'km' => 'utf-8',
    518. 

  518. 		'qc' => '',
    519. 

  519. 	);
    520. 

  520. 

  521. 		// TYPO3 specific: Array with the iso names used for each system language in TYPO3:
    522. 

  522. 		// Missing keys means: same as Typo3
    523. 

  523. 	var $isoArray = array(
    524. 

  524. 		'ba' => 'bs',
    525. 

  525. 		'br' => 'pt_BR',
    526. 

  526. 		'ch' => 'zh_CN',
    527. 

  527. 		'cz' => 'cs',
    528. 

  528. 		'dk' => 'da',
    529. 

  529. 		'si' => 'sl',
    530. 

  530. 		'se' => 'sv',
    531. 

  531. 		'gl' => 'kl',
    532. 

  532. 		'gr' => 'el',
    533. 

  533. 		'hk' => 'zh_HK',
    534. 

  534. 		'kr' => 'ko',
    535. 

  535. 		'ua' => 'uk',
    536. 

  536. 		'jp' => 'ja',
    537. 

  537. 		'qc' => 'fr_CA',
    538. 

  538. 		'vn' => 'vi',
    539. 

  539. 		'ge' => 'ka',
    540. 

  540. 		'ga' => 'gl',
    541. 

  541. 	);
    542. 

  542. 

  543. 	/**
    544. 

  544. 	 * Normalize - changes input character set to lowercase letters.
    545. 

  545. 	 *
    546. 

  546. 	 * @param	string		Input charset
    547. 

  547. 	 * @return	string		Normalized charset
    548. 

  548. 	 * @author	Martin Kutschker <martin.t.kutschker@blackbox.net>
    549. 

  549. 	 */
    550. 

  550. 	function parse_charset($charset) {
    551. 

  551. 		$charset = trim(strtolower($charset));
    552. 

  552. 		if (isset($this->synonyms[$charset])) {
    553. 

  553. 			$charset = $this->synonyms[$charset];
    554. 

  554. 		}
    555. 

  555. 

  556. 		return $charset;
    557. 

  557. 	}
    558. 

  558. 

  559. 	/**
    560. 

  560. 	 * Get the charset of a locale.
    561. 

  561. 	 *
    562. 

  562. 	 * ln			language
    563. 

  563. 	 * ln_CN		 language / country
    564. 

  564. 	 * ln_CN.cs	  language / country / charset
    565. 

  565. 	 * ln_CN.cs@mod  language / country / charset / modifier
    566. 

  566. 	 *
    567. 

  567. 	 * @param	string		Locale string
    568. 

  568. 	 * @return	string		Charset resolved for locale string
    569. 

  569. 	 * @author	Martin Kutschker <martin.t.kutschker@blackbox.net>
    570. 

  570. 	 */
    571. 

  571. 	function get_locale_charset($locale) {
    572. 

  572. 		$locale = strtolower($locale);
    573. 

  573. 

  574. 			// exact locale specific charset?
    575. 

  575. 		if (isset($this->locale_to_charset[$locale])) {
    576. 

  576. 			return $this->locale_to_charset[$locale];
    577. 

  577. 		}
    578. 

  578. 

  579. 			// get modifier
    580. 

  580. 		list($locale, $modifier) = explode('@', $locale);
    581. 

  581. 

  582. 			// locale contains charset: use it
    583. 

  583. 		list($locale, $charset) = explode('.', $locale);
    584. 

  584. 		if ($charset) {
    585. 

  585. 			return $this->parse_charset($charset);
    586. 

  586. 		}
    587. 

  587. 

  588. 			// modifier is 'euro' (after charset check, because of xx.utf-8@euro)
    589. 

  589. 		if ($modifier == 'euro') {
    590. 

  590. 			return 'iso-8859-15';
    591. 

  591. 		}
    592. 

  592. 

  593. 			// get language
    594. 

  594. 		list($language, $country) = explode('_', $locale);
    595. 

  595. 		if (isset($this->lang_to_script[$language])) {
    596. 

  596. 			$script = $this->lang_to_script[$language];
    597. 

  597. 		}
    598. 

  598. 

  599. 		if (TYPO3_OS == 'WIN') {
    600. 

  600. 			$cs = $this->script_to_charset_windows[$script] ? $this->script_to_charset_windows[$script] : 'windows-1252';
    601. 

  601. 		} else {
    602. 

  602. 			$cs = $this->script_to_charset_unix[$script] ? $this->script_to_charset_unix[$script] : 'iso-8859-1';
    603. 

  603. 		}
    604. 

  604. 

  605. 		return $cs;
    606. 

  606. 	}
    607. 

  607. 

  608. 

  609. 	/********************************************
    610. 

  610. 	 *
    611. 

  611. 	 * Charset Conversion functions
    612. 

  612. 	 *
    613. 

  613. 	 ********************************************/
    614. 

  614. 

  615. 	/**
    616. 

  616. 	 * Convert from one charset to another charset.
    617. 

  617. 	 *
    618. 

  618. 	 * @param	string		Input string
    619. 

  619. 	 * @param	string		From charset (the current charset of the string)
    620. 

  620. 	 * @param	string		To charset (the output charset wanted)
    621. 

  621. 	 * @param	boolean		If set, then characters that are not available in the destination character set will be encoded as numeric entities
    622. 

  622. 	 * @return	string		Converted string
    623. 

  623. 	 * @see convArray()
    624. 

  624. 	 */
    625. 

  625. 	function conv($str, $fromCS, $toCS, $useEntityForNoChar = 0) {
    626. 

  626. 		if ($fromCS == $toCS) {
    627. 

  627. 			return $str;
    628. 

  628. 		}
    629. 

  629. 

  630. 			// PHP-libs don't support fallback to SGML entities, but UTF-8 handles everything
    631. 

  631. 		if ($toCS == 'utf-8' || !$useEntityForNoChar) {
    632. 

  632. 			switch ($GLOBALS['TYPO3_CONF_VARS']['SYS']['t3lib_cs_convMethod']) {
    633. 

  633. 				case 'mbstring':
    634. 

  634. 					$conv_str = mb_convert_encoding($str, $toCS, $fromCS);
    635. 

  635. 					if (FALSE !== $conv_str) {
    636. 

  636. 						return $conv_str;
    637. 

  637. 					} // returns false for unsupported charsets
    638. 

  638. 					break;
    639. 

  639. 

  640. 				case 'iconv':
    641. 

  641. 					$conv_str = iconv($fromCS, $toCS . '//TRANSLIT', $str);
    642. 

  642. 					if (FALSE !== $conv_str) {
    643. 

  643. 						return $conv_str;
    644. 

  644. 					}
    645. 

  645. 					break;
    646. 

  646. 

  647. 				case 'recode':
    648. 

  648. 					$conv_str = recode_string($fromCS . '..' . $toCS, $str);
    649. 

  649. 					if (FALSE !== $conv_str) {
    650. 

  650. 						return $conv_str;
    651. 

  651. 					}
    652. 

  652. 					break;
    653. 

  653. 			}
    654. 

  654. 			// fallback to TYPO3 conversion
    655. 

  655. 		}
    656. 

  656. 

  657. 		if ($fromCS != 'utf-8') {
    658. 

  658. 			$str = $this->utf8_encode($str, $fromCS);
    659. 

  659. 		}
    660. 

  660. 		if ($toCS != 'utf-8') {
    661. 

  661. 			$str = $this->utf8_decode($str, $toCS, $useEntityForNoChar);
    662. 

  662. 		}
    663. 

  663. 		return $str;
    664. 

  664. 	}
    665. 

  665. 

  666. 	/**
    667. 

  667. 	 * Convert all elements in ARRAY with type string from one charset to another charset.
    668. 

  668. 	 * NOTICE: Array is passed by reference!
    669. 

  669. 	 *
    670. 

  670. 	 * @param	string		Input array, possibly multidimensional
    671. 

  671. 	 * @param	string		From charset (the current charset of the string)
    672. 

  672. 	 * @param	string		To charset (the output charset wanted)
    673. 

  673. 	 * @param	boolean		If set, then characters that are not available in the destination character set will be encoded as numeric entities
    674. 

  674. 	 * @return	void
    675. 

  675. 	 * @see conv()
    676. 

  676. 	 */
    677. 

  677. 	function convArray(&$array, $fromCS, $toCS, $useEntityForNoChar = 0) {
    678. 

  678. 		foreach ($array as $key => $value) {
    679. 

  679. 			if (is_array($array[$key])) {
    680. 

  680. 				$this->convArray($array[$key], $fromCS, $toCS, $useEntityForNoChar);
    681. 

  681. 			} elseif (is_string($array[$key])) {
    682. 

  682. 				$array[$key] = $this->conv($array[$key], $fromCS, $toCS, $useEntityForNoChar);
    683. 

  683. 			}
    684. 

  684. 		}
    685. 

  685. 	}
    686. 

  686. 

  687. 	/**
    688. 

  688. 	 * Converts $str from $charset to UTF-8
    689. 

  689. 	 *
    690. 

  690. 	 * @param	string		String in local charset to convert to UTF-8
    691. 

  691. 	 * @param	string		Charset, lowercase. Must be found in csconvtbl/ folder.
    692. 

  692. 	 * @return	string		Output string, converted to UTF-8
    693. 

  693. 	 */
    694. 

  694. 	function utf8_encode($str, $charset) {
    695. 

  695. 

  696. 		if ($charset === 'utf-8') {
    697. 

  697. 			return $str;
    698. 

  698. 		}
    699. 

  699. 

  700. 			// Charset is case-insensitive.
    701. 

  701. 		if ($this->initCharset($charset)) { // Parse conv. table if not already...
    702. 

  702. 			$strLen = strlen($str);
    703. 

  703. 			$outStr = '';
    704. 

  704. 

  705. 			for ($a = 0; $a < $strLen; $a++) { // Traverse each char in string.
    706. 

  706. 				$chr = substr($str, $a, 1);
    707. 

  707. 				$ord = ord($chr);
    708. 

  708. 				if (isset($this->twoByteSets[$charset])) { // If the charset has two bytes per char
    709. 

  709. 					$ord2 = ord($str{$a + 1});
    710. 

  710. 					$ord = $ord << 8 | $ord2; // assume big endian
    711. 

  711. 

  712. 					if (isset($this->parsedCharsets[$charset]['local'][$ord])) { // If the local char-number was found in parsed conv. table then we use that, otherwise 127 (no char?)
    713. 

  713. 						$outStr .= $this->parsedCharsets[$charset]['local'][$ord];
    714. 

  714. 					} else {
    715. 

  715. 						$outStr .= chr($this->noCharByteVal);
    716. 

  716. 					} // No char exists
    717. 

  717. 					$a++;
    718. 

  718. 				} elseif ($ord > 127) { // If char has value over 127 it's a multibyte char in UTF-8
    719. 

  719. 					if (isset($this->eucBasedSets[$charset])) { // EUC uses two-bytes above 127; we get both and advance pointer and make $ord a 16bit int.
    720. 

  720. 						if ($charset != 'shift_jis' || ($ord < 0xA0 || $ord > 0xDF)) { // Shift-JIS: chars between 160 and 223 are single byte
    721. 

  721. 							$a++;
    722. 

  722. 							$ord2 = ord(substr($str, $a, 1));
    723. 

  723. 							$ord = $ord * 256 + $ord2;
    724. 

  724. 						}
    725. 

  725. 					}
    726. 

  726. 

  727. 					if (isset($this->parsedCharsets[$charset]['local'][$ord])) { // If the local char-number was found in parsed conv. table then we use that, otherwise 127 (no char?)
    728. 

  728. 						$outStr .= $this->parsedCharsets[$charset]['local'][$ord];
    729. 

  729. 					} else {
    730. 

  730. 						$outStr .= chr($this->noCharByteVal);
    731. 

  731. 					} // No char exists
    732. 

  732. 				} else {
    733. 

  733. 					$outStr .= $chr;
    734. 

  734. 				} // ... otherwise it's just ASCII 0-127 and one byte. Transparent
    735. 

  735. 			}
    736. 

  736. 			return $outStr;
    737. 

  737. 		}
    738. 

  738. 	}
    739. 

  739. 

  740. 	/**
    741. 

  741. 	 * Converts $str from UTF-8 to $charset
    742. 

  742. 	 *
    743. 

  743. 	 * @param	string		String in UTF-8 to convert to local charset
    744. 

  744. 	 * @param	string		Charset, lowercase. Must be found in csconvtbl/ folder.
    745. 

  745. 	 * @param	boolean		If set, then characters that are not available in the destination character set will be encoded as numeric entities
    746. 

  746. 	 * @return	string		Output string, converted to local charset
    747. 

  747. 	 */
    748. 

  748. 	function utf8_decode($str, $charset, $useEntityForNoChar = 0) {
    749. 

  749. 

  750. 		if ($charset === 'utf-8') {
    751. 

  751. 			return $str;
    752. 

  752. 		}
    753. 

  753. 

  754. 			// Charset is case-insensitive.
    755. 

  755. 		if ($this->initCharset($charset)) { // Parse conv. table if not already...
    756. 

  756. 			$strLen = strlen($str);
    757. 

  757. 			$outStr = '';
    758. 

  758. 			$buf = '';
    759. 

  759. 			for ($a = 0, $i = 0; $a < $strLen; $a++, $i++) { // Traverse each char in UTF-8 string.
    760. 

  760. 				$chr = substr($str, $a, 1);
    761. 

  761. 				$ord = ord($chr);
    762. 

  762. 				if ($ord > 127) { // This means multibyte! (first byte!)
    763. 

  763. 					if ($ord & 64) { // Since the first byte must have the 7th bit set we check that. Otherwise we might be in the middle of a byte sequence.
    764. 

  764. 

  765. 						$buf = $chr; // Add first byte
    766. 

  766. 						for ($b = 0; $b < 8; $b++) { // for each byte in multibyte string...
    767. 

  767. 							$ord = $ord << 1; // Shift it left and ...
    768. 

  768. 							if ($ord & 128) { // ... and with 8th bit - if that is set, then there are still bytes in sequence.
    769. 

  769. 								$a++; // Increase pointer...
    770. 

  770. 								$buf .= substr($str, $a, 1); // ... and add the next char.
    771. 

  771. 							} else {
    772. 

  772. 								break;
    773. 

  773. 							}
    774. 

  774. 						}
    775. 

  775. 

  776. 						if (isset($this->parsedCharsets[$charset]['utf8'][$buf])) { // If the UTF-8 char-sequence is found then...
    777. 

  777. 							$mByte = $this->parsedCharsets[$charset]['utf8'][$buf]; // The local number
    778. 

  778. 							if ($mByte > 255) { // If the local number is greater than 255 we will need to split the byte (16bit word assumed) in two chars.
    779. 

  779. 								$outStr .= chr(($mByte >> 8) & 255) . chr($mByte & 255);
    780. 

  780. 							} else {
    781. 

  781. 								$outStr .= chr($mByte);
    782. 

  782. 							}
    783. 

  783. 						} elseif ($useEntityForNoChar) { // Create num entity:
    784. 

  784. 							$outStr .= '&#' . $this->utf8CharToUnumber($buf, 1) . ';';
    785. 

  785. 						} else {
    786. 

  786. 							$outStr .= chr($this->noCharByteVal);
    787. 

  787. 						} // No char exists
    788. 

  788. 					} else {
    789. 

  789. 						$outStr .= chr($this->noCharByteVal);
    790. 

  790. 					} // No char exists (MIDDLE of MB sequence!)
    791. 

  791. 				} else {
    792. 

  792. 					$outStr .= $chr;
    793. 

  793. 				} // ... otherwise it's just ASCII 0-127 and one byte. Transparent
    794. 

  794. 			}
    795. 

  795. 			return $outStr;
    796. 

  796. 		}
    797. 

  797. 	}
    798. 

  798. 

  799. 	/**
    800. 

  800. 	 * Converts all chars > 127 to numeric entities.
    801. 

  801. 	 *
    802. 

  802. 	 * @param	string		Input string
    803. 

  803. 	 * @return	string		Output string
    804. 

  804. 	 */
    805. 

  805. 	function utf8_to_entities($str) {
    806. 

  806. 		$strLen = strlen($str);
    807. 

  807. 		$outStr = '';
    808. 

  808. 		$buf = '';
    809. 

  809. 		for ($a = 0; $a < $strLen; $a++) { // Traverse each char in UTF-8 string.
    810. 

  810. 			$chr = substr($str, $a, 1);
    811. 

  811. 			$ord = ord($chr);
    812. 

  812. 			if ($ord > 127) { // This means multibyte! (first byte!)
    813. 

  813. 				if ($ord & 64) { // Since the first byte must have the 7th bit set we check that. Otherwise we might be in the middle of a byte sequence.
    814. 

  814. 					$buf = $chr; // Add first byte
    815. 

  815. 					for ($b = 0; $b < 8; $b++) { // for each byte in multibyte string...
    816. 

  816. 						$ord = $ord << 1; // Shift it left and ...
    817. 

  817. 						if ($ord & 128) { // ... and with 8th bit - if that is set, then there are still bytes in sequence.
    818. 

  818. 							$a++; // Increase pointer...
    819. 

  819. 							$buf .= substr($str, $a, 1); // ... and add the next char.
    820. 

  820. 						} else {
    821. 

  821. 							break;
    822. 

  822. 						}
    823. 

  823. 					}
    824. 

  824. 

  825. 					$outStr .= '&#' . $this->utf8CharToUnumber($buf, 1) . ';';
    826. 

  826. 				} else {
    827. 

  827. 					$outStr .= chr($this->noCharByteVal);
    828. 

  828. 				} // No char exists (MIDDLE of MB sequence!)
    829. 

  829. 			} else {
    830. 

  830. 				$outStr .= $chr;
    831. 

  831. 			} // ... otherwise it's just ASCII 0-127 and one byte. Transparent
    832. 

  832. 		}
    833. 

  833. 

  834. 		return $outStr;
    835. 

  835. 	}
    836. 

  836. 

  837. 	/**
    838. 

  838. 	 * Converts numeric entities (UNICODE, eg. decimal (&#1234;) or hexadecimal (&#x1b;)) to UTF-8 multibyte chars
    839. 

  839. 	 *
    840. 

  840. 	 * @param	string		Input string, UTF-8
    841. 

  841. 	 * @param	boolean		If set, then all string-HTML entities (like &amp; or &pound; will be converted as well)
    842. 

  842. 	 * @return	string		Output string
    843. 

  843. 	 */
    844. 

  844. 	function entities_to_utf8($str, $alsoStdHtmlEnt = 0) {
    845. 

  845. 		if ($alsoStdHtmlEnt) {
    846. 

  846. 			$trans_tbl = array_flip(get_html_translation_table(HTML_ENTITIES)); // Getting them in iso-8859-1 - but thats ok since this is observed below.
    847. 

  847. 		}
    848. 

  848. 

  849. 		$token = md5(microtime());
    850. 

  850. 		$parts = explode($token, preg_replace('/(&([#[:alnum:]]*);)/', $token . '${2}' . $token, $str));
    851. 

  851. 		foreach ($parts as $k => $v) {
    852. 

  852. 			if ($k % 2) {
    853. 

  853. 				if (substr($v, 0, 1) == '#') { // Dec or hex entities:
    854. 

  854. 					if (substr($v, 1, 1) == 'x') {
    855. 

  855. 						$parts[$k] = $this->UnumberToChar(hexdec(substr($v, 2)));
    856. 

  856. 					} else {
    857. 

  857. 						$parts[$k] = $this->UnumberToChar(substr($v, 1));
    858. 

  858. 					}
    859. 

  859. 				} elseif ($alsoStdHtmlEnt && $trans_tbl['&' . $v . ';']) { // Other entities:
    860. 

  860. 					$parts[$k] = $this->utf8_encode($trans_tbl['&' . $v . ';'], 'iso-8859-1');
    861. 

  861. 				} else { // No conversion:
    862. 

  862. 					$parts[$k] = '&' . $v . ';';
    863. 

  863. 				}
    864. 

  864. 			}
    865. 

  865. 		}
    866. 

  866. 

  867. 		return implode('', $parts);
    868. 

  868. 	}
    869. 

  869. 

  870. 	/**
    871. 

  871. 	 * Converts all chars in the input UTF-8 string into integer numbers returned in an array
    872. 

  872. 	 *
    873. 

  873. 	 * @param	string		Input string, UTF-8
    874. 

  874. 	 * @param	boolean		If set, then all HTML entities (like &amp; or &pound; or &#123; or &#x3f5d;) will be detected as characters.
    875. 

  875. 	 * @param	boolean		If set, then instead of integer numbers the real UTF-8 char is returned.
    876. 

  876. 	 * @return	array		Output array with the char numbers
    877. 

  877. 	 */
    878. 

  878. 	function utf8_to_numberarray($str, $convEntities = 0, $retChar = 0) {
    879. 

  879. 			// If entities must be registered as well...:
    880. 

  880. 		if ($convEntities) {
    881. 

  881. 			$str = $this->entities_to_utf8($str, 1);
    882. 

  882. 		}
    883. 

  883. 			// Do conversion:
    884. 

  884. 		$strLen = strlen($str);
    885. 

  885. 		$outArr = array();
    886. 

  886. 		$buf = '';
    887. 

  887. 		for ($a = 0; $a < $strLen; $a++) { // Traverse each char in UTF-8 string.
    888. 

  888. 			$chr = substr($str, $a, 1);
    889. 

  889. 			$ord = ord($chr);
    890. 

  890. 			if ($ord > 127) { // This means multibyte! (first byte!)
    891. 

  891. 				if ($ord & 64) { // Since the first byte must have the 7th bit set we check that. Otherwise we might be in the middle of a byte sequence.
    892. 

  892. 					$buf = $chr; // Add first byte
    893. 

  893. 					for ($b = 0; $b < 8; $b++) { // for each byte in multibyte string...
    894. 

  894. 						$ord = $ord << 1; // Shift it left and ...
    895. 

  895. 						if ($ord & 128) { // ... and with 8th bit - if that is set, then there are still bytes in sequence.
    896. 

  896. 							$a++; // Increase pointer...
    897. 

  897. 							$buf .= substr($str, $a, 1); // ... and add the next char.
    898. 

  898. 						} else {
    899. 

  899. 							break;
    900. 

  900. 						}
    901. 

  901. 					}
    902. 

  902. 

  903. 					$outArr[] = $retChar ? $buf : $this->utf8CharToUnumber($buf);
    904. 

  904. 				} else {
    905. 

  905. 					$outArr[] = $retChar ? chr($this->noCharByteVal) : $this->noCharByteVal;
    906. 

  906. 				} // No char exists (MIDDLE of MB sequence!)
    907. 

  907. 			} else {
    908. 

  908. 				$outArr[] = $retChar ? chr($ord) : $ord;
    909. 

  909. 			} // ... otherwise it's just ASCII 0-127 and one byte. Transparent
    910. 

  910. 		}
    911. 

  911. 

  912. 		return $outArr;
    913. 

  913. 	}
    914. 

  914. 

  915. 	/**
    916. 

  916. 	 * Converts a UNICODE number to a UTF-8 multibyte character
    917. 

  917. 	 * Algorithm based on script found at From: http://czyborra.com/utf/
    918. 

  918. 	 * Unit-tested by Kasper
    919. 

  919. 	 *
    920. 

  920. 	 * The binary representation of the character's integer value is thus simply spread across the bytes and the number of high bits set in the lead byte announces the number of bytes in the multibyte sequence:
    921. 

  921. 	 *
    922. 

  922. 	 *  bytes | bits | representation
    923. 

  923. 	 *	  1 |	7 | 0vvvvvvv
    924. 

  924. 	 *	  2 |   11 | 110vvvvv 10vvvvvv
    925. 

  925. 	 *	  3 |   16 | 1110vvvv 10vvvvvv 10vvvvvv
    926. 

  926. 	 *	  4 |   21 | 11110vvv 10vvvvvv 10vvvvvv 10vvvvvv
    927. 

  927. 	 *	  5 |   26 | 111110vv 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv
    928. 

  928. 	 *	  6 |   31 | 1111110v 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv
    929. 

  929. 	 *
    930. 

  930. 	 * @param	integer		UNICODE integer
    931. 

  931. 	 * @return	string		UTF-8 multibyte character string
    932. 

  932. 	 * @see utf8CharToUnumber()
    933. 

  933. 	 */
    934. 

  934. 	function UnumberToChar($cbyte) {
    935. 

  935. 		$str = '';
    936. 

  936. 

  937. 		if ($cbyte < 0x80) {
    938. 

  938. 			$str .= chr($cbyte);
    939. 

  939. 		} else {
    940. 

  940. 			if ($cbyte < 0x800) {
    941. 

  941. 				$str .= chr(0xC0 | ($cbyte >> 6));
    942. 

  942. 				$str .= chr(0x80 | ($cbyte & 0x3F));
    943. 

  943. 			} else {
    944. 

  944. 				if ($cbyte < 0x10000) {
    945. 

  945. 					$str .= chr(0xE0 | ($cbyte >> 12));
    946. 

  946. 					$str .= chr(0x80 | (($cbyte >> 6) & 0x3F));
    947. 

  947. 					$str .= chr(0x80 | ($cbyte & 0x3F));
    948. 

  948. 				} else {
    949. 

  949. 					if ($cbyte < 0x200000) {
    950. 

  950. 						$str .= chr(0xF0 | ($cbyte >> 18));
    951. 

  951. 						$str .= chr(0x80 | (($cbyte >> 12) & 0x3F));
    952. 

  952. 						$str .= chr(0x80 | (($cbyte >> 6) & 0x3F));
    953. 

  953. 						$str .= chr(0x80 | ($cbyte & 0x3F));
    954. 

  954. 					} else {
    955. 

  955. 						if ($cbyte < 0x4000000) {
    956. 

  956. 							$str .= chr(0xF8 | ($cbyte >> 24));
    957. 

  957. 							$str .= chr(0x80 | (($cbyte >> 18) & 0x3F));
    958. 

  958. 							$str .= chr(0x80 | (($cbyte >> 12) & 0x3F));
    959. 

  959. 							$str .= chr(0x80 | (($cbyte >> 6) & 0x3F));
    960. 

  960. 							$str .= chr(0x80 | ($cbyte & 0x3F));
    961. 

  961. 						} else {
    962. 

  962. 							if ($cbyte < 0x80000000) {
    963. 

  963. 								$str .= chr(0xFC | ($cbyte >> 30));
    964. 

  964. 								$str .= chr(0x80 | (($cbyte >> 24) & 0x3F));
    965. 

  965. 								$str .= chr(0x80 | (($cbyte >> 18) & 0x3F));
    966. 

  966. 								$str .= chr(0x80 | (($cbyte >> 12) & 0x3F));
    967. 

  967. 								$str .= chr(0x80 | (($cbyte >> 6) & 0x3F));
    968. 

  968. 								$str .= chr(0x80 | ($cbyte & 0x3F));
    969. 

  969. 							} else { // Cannot express a 32-bit character in UTF-8
    970. 

  970. 								$str .= chr($this->noCharByteVal);
    971. 

  971. 							}
    972. 

  972. 						}
    973. 

  973. 					}
    974. 

  974. 				}
    975. 

  975. 			}
    976. 

  976. 		}
    977. 

  977. 		return $str;
    978. 

  978. 	}
    979. 

  979. 

  980. 	/**
    981. 

  981. 	 * Converts a UTF-8 Multibyte character to a UNICODE number
    982. 

  982. 	 * Unit-tested by Kasper
    983. 

  983. 	 *
    984. 

  984. 	 * @param	string		UTF-8 multibyte character string
    985. 

  985. 	 * @param	boolean		If set, then a hex. number is returned.
    986. 

  986. 	 * @return	integer		UNICODE integer
    987. 

  987. 	 * @see UnumberToChar()
    988. 

  988. 	 */
    989. 

  989. 	function utf8CharToUnumber($str, $hex = 0) {
    990. 

  990. 		$ord = ord(substr($str, 0, 1)); // First char
    991. 

  991. 

  992. 		if (($ord & 192) == 192) { // This verifyes that it IS a multi byte string
    993. 

  993. 			$binBuf = '';
    994. 

  994. 			for ($b = 0; $b < 8; $b++) { // for each byte in multibyte string...
    995. 

  995. 				$ord = $ord << 1; // Shift it left and ...
    996. 

  996. 				if ($ord & 128) { // ... and with 8th bit - if that is set, then there are still bytes in sequence.
    997. 

  997. 					$binBuf .= substr('00000000' . decbin(ord(substr($str, $b + 1, 1))), -6);
    998. 

  998. 				} else {
    999. 

  999. 					break;
    1000. 

  1000. 				}
    1001. 

  1001. 			}
    1002. 

  1002. 			$binBuf = substr('00000000' . decbin(ord(substr($str, 0, 1))), -(6 - $b)) . $binBuf;
    1003. 

  1003. 

  1004. 			$int = bindec($binBuf);
    1005. 

  1005. 		} else {
    1006. 

  1006. 			$int = $ord;
    1007. 

  1007. 		}
    1008. 

  1008. 

  1009. 		return $hex ? 'x' . dechex($int) : $int;
    1010. 

  1010. 	}
    1011. 

  1011. 

  1012. 

  1013. 	/********************************************
    1014. 

  1014. 	 *
    1015. 

  1015. 	 * Init functions
    1016. 

  1016. 	 *
    1017. 

  1017. 	 ********************************************/
    1018. 

  1018. 

  1019. 	/**
    1020. 

  1020. 	 * This will initialize a charset for use if it's defined in the PATH_t3lib.'csconvtbl/' folder
    1021. 

  1021. 	 * This function is automatically called by the conversion functions
    1022. 

  1022. 	 *
    1023. 

  1023. 	 * PLEASE SEE: http://www.unicode.org/Public/MAPPINGS/
    1024. 

  1024. 	 *
    1025. 

  1025. 	 * @param	string		The charset to be initialized. Use lowercase charset always (the charset must match exactly with a filename in csconvtbl/ folder ([charset].tbl)
    1026. 

  1026. 	 * @return	integer		Returns '1' if already loaded. Returns FALSE if charset conversion table was not found. Returns '2' if the charset conversion table was found and parsed.
    1027. 

  1027. 	 * @access private
    1028. 

  1028. 	 */
    1029. 

  1029. 	function initCharset($charset) {
    1030. 

  1030. 			// Only process if the charset is not yet loaded:
    1031. 

  1031. 		if (empty($this->parsedCharsets[$charset]) || !is_array($this->parsedCharsets[$charset])) {
    1032. 

  1032. 

  1033. 				// Conversion table filename:
    1034. 

  1034. 			$charsetConvTableFile = PATH_t3lib . 'csconvtbl/' . $charset . '.tbl';
    1035. 

  1035. 

  1036. 				// If the conversion table is found:
    1037. 

  1037. 			if ($charset && t3lib_div::validPathStr($charsetConvTableFile) && @is_file($charsetConvTableFile)) {
    1038. 

  1038. 					// Cache file for charsets:
    1039. 

  1039. 					// Caching brought parsing time for gb2312 down from 2400 ms to 150 ms. For other charsets we are talking 11 ms down to zero.
    1040. 

  1040. 				$cacheFile = t3lib_div::getFileAbsFileName('typo3temp/cs/charset_' . $charset . '.tbl');
    1041. 

  1041. 				if ($cacheFile && @is_file($cacheFile)) {
    1042. 

  1042. 					$this->parsedCharsets[$charset] = unserialize(t3lib_div::getUrl($cacheFile));
    1043. 

  1043. 				} else {
    1044. 

  1044. 						// Parse conversion table into lines:
    1045. 

  1045. 					$lines = t3lib_div::trimExplode(LF, t3lib_div::getUrl($charsetConvTableFile), 1);
    1046. 

  1046. 						// Initialize the internal variable holding the conv. table:
    1047. 

  1047. 					$this->parsedCharsets[$charset] = array('local' => array(), 'utf8' => array());
    1048. 

  1048. 						// traverse the lines:
    1049. 

  1049. 					$detectedType = '';
    1050. 

  1050. 					foreach ($lines as $value) {
    1051. 

  1051. 						if (trim($value) && substr($value, 0, 1) != '#') { // Comment line or blanks are ignored.
    1052. 

  1052. 

  1053. 								// Detect type if not done yet: (Done on first real line)
    1054. 

  1054. 								// The "whitespaced" type is on the syntax 	"0x0A	0x000A	#LINE FEED" 	while 	"ms-token" is like 		"B9 = U+00B9 : SUPERSCRIPT ONE"
    1055. 

  1055. 							if (!$detectedType) {
    1056. 

  1056. 								$detectedType = preg_match('/[[:space:]]*0x([[:alnum:]]*)[[:space:]]+0x([[:alnum:]]*)[[:space:]]+/', $value) ? 'whitespaced' : 'ms-token';
    1057. 

  1057. 							}
    1058. 

  1058. 

  1059. 							if ($detectedType == 'ms-token') {
    1060. 

  1060. 								list($hexbyte, $utf8) = preg_split('/[=:]/', $value, 3);
    1061. 

  1061. 							} elseif ($detectedType == 'whitespaced') {
    1062. 

  1062. 								$regA = array();
    1063. 

  1063. 								preg_match('/[[:space:]]*0x([[:alnum:]]*)[[:space:]]+0x([[:alnum:]]*)[[:space:]]+/', $value, $regA);
    1064. 

  1064. 								$hexbyte = $regA[1];
    1065. 

  1065. 								$utf8 = 'U+' . $regA[2];
    1066. 

  1066. 							}
    1067. 

  1067. 							$decval = hexdec(trim($hexbyte));
    1068. 

  1068. 							if ($decval > 127) {
    1069. 

  1069. 								$utf8decval = hexdec(substr(trim($utf8), 2));
    1070. 

  1070. 								$this->parsedCharsets[$charset]['local'][$decval] = $this->UnumberToChar($utf8decval);
    1071. 

  1071. 								$this->parsedCharsets[$charset]['utf8'][$this->parsedCharsets[$charset]['local'][$decval]] = $decval;
    1072. 

  1072. 							}
    1073. 

  1073. 						}
    1074. 

  1074. 					}
    1075. 

  1075. 					if ($cacheFile) {
    1076. 

  1076. 						t3lib_div::writeFileToTypo3tempDir($cacheFile, serialize($this->parsedCharsets[$charset]));
    1077. 

  1077. 					}
    1078. 

  1078. 				}
    1079. 

  1079. 				return 2;
    1080. 

  1080. 			} else {
    1081. 

  1081. 				return FALSE;
    1082. 

  1082. 			}
    1083. 

  1083. 		} else {
    1084. 

  1084. 			return 1;
    1085. 

  1085. 		}
    1086. 

  1086. 	}
    1087. 

  1087. 

  1088. 	/**
    1089. 

  1089. 	 * This function initializes all UTF-8 character data tables.
    1090. 

  1090. 	 *
    1091. 

  1091. 	 * PLEASE SEE: http://www.unicode.org/Public/UNIDATA/
    1092. 

  1092. 	 *
    1093. 

  1093. 	 * @param	string		Mode ("case", "ascii", ...)
    1094. 

  1094. 	 * @return	integer		Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached).
    1095. 

  1095. 	 * @access private
    1096. 

  1096. 	 */
    1097. 

  1097. 	function initUnicodeData($mode = NULL) {
    1098. 

  1098. 			// cache files
    1099. 

  1099. 		$cacheFileCase = t3lib_div::getFileAbsFileName('typo3temp/cs/cscase_utf-8.tbl');
    1100. 

  1100. 		$cacheFileASCII = t3lib_div::getFileAbsFileName('typo3temp/cs/csascii_utf-8.tbl');
    1101. 

  1101. 

  1102. 			// Only process if the tables are not yet loaded
    1103. 

  1103. 		switch ($mode) {
    1104. 

  1104. 			case 'case':
    1105. 

  1105. 				if (is_array($this->caseFolding['utf-8'])) {
    1106. 

  1106. 					return 1;
    1107. 

  1107. 				}
    1108. 

  1108. 

  1109. 					// Use cached version if possible
    1110. 

  1110. 				if ($cacheFileCase && @is_file($cacheFileCase)) {
    1111. 

  1111. 					$this->caseFolding['utf-8'] = unserialize(t3lib_div::getUrl($cacheFileCase));
    1112. 

  1112. 					return 2;
    1113. 

  1113. 				}
    1114. 

  1114. 				break;
    1115. 

  1115. 

  1116. 			case 'ascii':
    1117. 

  1117. 				if (is_array($this->toASCII['utf-8'])) {
    1118. 

  1118. 					return 1;
    1119. 

  1119. 				}
    1120. 

  1120. 

  1121. 					// Use cached version if possible
    1122. 

  1122. 				if ($cacheFileASCII && @is_file($cacheFileASCII)) {
    1123. 

  1123. 					$this->toASCII['utf-8'] = unserialize(t3lib_div::getUrl($cacheFileASCII));
    1124. 

  1124. 					return 2;
    1125. 

  1125. 				}
    1126. 

  1126. 				break;
    1127. 

  1127. 		}
    1128. 

  1128. 

  1129. 			// process main Unicode data file
    1130. 

  1130. 		$unicodeDataFile = PATH_t3lib . 'unidata/UnicodeData.txt';
    1131. 

  1131. 		if (!(t3lib_div::validPathStr($unicodeDataFile) && @is_file($unicodeDataFile))) {
    1132. 

  1132. 			return FALSE;
    1133. 

  1133. 		}
    1134. 

  1134. 

  1135. 		$fh = fopen($unicodeDataFile, 'rb');
    1136. 

  1136. 		if (!$fh) {
    1137. 

  1137. 			return FALSE;
    1138. 

  1138. 		}
    1139. 

  1139. 

  1140. 			// key = utf8 char (single codepoint), value = utf8 string (codepoint sequence)
    1141. 

  1141. 			// note: we use the UTF-8 characters here and not the Unicode numbers to avoid conversion roundtrip in utf8_strtolower/-upper)
    1142. 

  1142. 		$this->caseFolding['utf-8'] = array();
    1143. 

  1143. 		$utf8CaseFolding =& $this->caseFolding['utf-8']; // a shorthand
    1144. 

  1144. 		$utf8CaseFolding['toUpper'] = array();
    1145. 

  1145. 		$utf8CaseFolding['toLower'] = array();
    1146. 

  1146. 		$utf8CaseFolding['toTitle'] = array();
    1147. 

  1147. 

  1148. 		$decomposition = array(); // array of temp. decompositions
    1149. 

  1149. 		$mark = array(); // array of chars that are marks (eg. composing accents)
    1150. 

  1150. 		$number = array(); // array of chars that are numbers (eg. digits)
    1151. 

  1151. 		$omit = array(); // array of chars to be omitted (eg. Russian hard sign)
    1152. 

  1152. 

  1153. 		while (!feof($fh)) {
    1154. 

  1154. 			$line = fgets($fh, 4096);
    1155. 

  1155. 				// has a lot of info
    1156. 

  1156. 			list($char, $name, $cat, , , $decomp, , , $num, , , , $upper, $lower, $title,) = explode(';', rtrim($line));
    1157. 

  1157. 

  1158. 			$ord = hexdec($char);
    1159. 

  1159. 			if ($ord > 0xFFFF) {
    1160. 

  1160. 				break;
    1161. 

  1161. 			} // only process the BMP
    1162. 

  1162. 

  1163. 			$utf8_char = $this->UnumberToChar($ord);
    1164. 

  1164. 

  1165. 			if ($upper) {
    1166. 

  1166. 				$utf8CaseFolding['toUpper'][$utf8_char] = $this->UnumberToChar(hexdec($upper));
    1167. 

  1167. 			}
    1168. 

  1168. 			if ($lower) {
    1169. 

  1169. 				$utf8CaseFolding['toLower'][$utf8_char] = $this->UnumberToChar(hexdec($lower));
    1170. 

  1170. 			}
    1171. 

  1171. 				// store "title" only when different from "upper" (only a few)
    1172. 

  1172. 			if ($title && $title != $upper) {
    1173. 

  1173. 				$utf8CaseFolding['toTitle'][$utf8_char] = $this->UnumberToChar(hexdec($title));
    1174. 

  1174. 			}
    1175. 

  1175. 

  1176. 			switch ($cat{0}) {
    1177. 

  1177. 				case 'M': // mark (accent, umlaut, ...)
    1178. 

  1178. 					$mark["U+$char"] = 1;
    1179. 

  1179. 					break;
    1180. 

  1180. 

  1181. 				case 'N': // numeric value
    1182. 

  1182. 					if ($ord > 0x80 && $num != '') {
    1183. 

  1183. 						$number["U+$char"] = $num;
    1184. 

  1184. 					}
    1185. 

  1185. 			}
    1186. 

  1186. 

  1187. 				// accented Latin letters without "official" decomposition
    1188. 

  1188. 			$match = array();
    1189. 

  1189. 			if (preg_match('/^LATIN (SMALL|CAPITAL) LETTER ([A-Z]) WITH/', $name, $match) && !$decomp) {
    1190. 

  1190. 				$c = ord($match[2]);
    1191. 

  1191. 				if ($match[1] == 'SMALL') {
    1192. 

  1192. 					$c += 32;
    1193. 

  1193. 				}
    1194. 

  1194. 

  1195. 				$decomposition["U+$char"] = array(dechex($c));
    1196. 

  1196. 				continue;
    1197. 

  1197. 			}
    1198. 

  1198. 

  1199. 			$match = array();
    1200. 

  1200. 			if (preg_match('/(<.*>)? *(.+)/', $decomp, $match)) {
    1201. 

  1201. 				switch ($match[1]) {
    1202. 

  1202. 					case '<circle>': // add parenthesis as circle replacement, eg (1)
    1203. 

  1203. 						$match[2] = '0028 ' . $match[2] . ' 0029';
    1204. 

  1204. 						break;
    1205. 

  1205. 

  1206. 					case '<square>': // add square brackets as square replacement, eg [1]
    1207. 

  1207. 						$match[2] = '005B ' . $match[2] . ' 005D';
    1208. 

  1208. 						break;
    1209. 

  1209. 

  1210. 					case '<compat>': // ignore multi char decompositions that start with a space
    1211. 

  1211. 						if (preg_match('/^0020 /', $match[2])) {
    1212. 

  1212. 							continue 2;
    1213. 

  1213. 						}
    1214. 

  1214. 						break;
    1215. 

  1215. 

  1216. 						// ignore Arabic and vertical layout presentation decomposition
    1217. 

  1217. 					case '<initial>':
    1218. 

  1218. 					case '<medial>':
    1219. 

  1219. 					case '<final>':
    1220. 

  1220. 					case '<isolated>':
    1221. 

  1221. 					case '<vertical>':
    1222. 

  1222. 						continue 2;
    1223. 

  1223. 				}
    1224. 

  1224. 				$decomposition["U+$char"] = explode(' ', $match[2]);
    1225. 

  1225. 			}
    1226. 

  1226. 		}
    1227. 

  1227. 		fclose($fh);
    1228. 

  1228. 

  1229. 			// process additional Unicode data for casing (allow folded characters to expand into a sequence)
    1230. 

  1230. 		$specialCasingFile = PATH_t3lib . 'unidata/SpecialCasing.txt';
    1231. 

  1231. 		if (t3lib_div::validPathStr($specialCasingFile) && @is_file($specialCasingFile)) {
    1232. 

  1232. 			$fh = fopen($specialCasingFile, 'rb');
    1233. 

  1233. 			if ($fh) {
    1234. 

  1234. 				while (!feof($fh)) {
    1235. 

  1235. 					$line = fgets($fh, 4096);
    1236. 

  1236. 					if ($line{0} != '#' && trim($line) != '') {
    1237. 

  1237. 

  1238. 						list($char, $lower, $title, $upper, $cond) = t3lib_div::trimExplode(';', $line);
    1239. 

  1239. 						if ($cond == '' || $cond{0} == '#') {
    1240. 

  1240. 							$utf8_char = $this->UnumberToChar(hexdec($char));
    1241. 

  1241. 							if ($char != $lower) {
    1242. 

  1242. 								$arr = explode(' ', $lower);
    1243. 

  1243. 								for ($i = 0; isset($arr[$i]); $i++) $arr[$i] = $this->UnumberToChar(hexdec($arr[$i]));
    1244. 

  1244. 								$utf8CaseFolding['toLower'][$utf8_char] = implode('', $arr);
    1245. 

  1245. 							}
    1246. 

  1246. 							if ($char != $title && $title != $upper) {
    1247. 

  1247. 								$arr = explode(' ', $title);
    1248. 

  1248. 								for ($i = 0; isset($arr[$i]); $i++) $arr[$i] = $this->UnumberToChar(hexdec($arr[$i]));
    1249. 

  1249. 								$utf8CaseFolding['toTitle'][$utf8_char] = implode('', $arr);
    1250. 

  1250. 							}
    1251. 

  1251. 							if ($char != $upper) {
    1252. 

  1252. 								$arr = explode(' ', $upper);
    1253. 

  1253. 								for ($i = 0; isset($arr[$i]); $i++) $arr[$i] = $this->UnumberToChar(hexdec($arr[$i]));
    1254. 

  1254. 								$utf8CaseFolding['toUpper'][$utf8_char] = implode('', $arr);
    1255. 

  1255. 							}
    1256. 

  1256. 						}
    1257. 

  1257. 					}
    1258. 

  1258. 				}
    1259. 

  1259. 				fclose($fh);
    1260. 

  1260. 			}
    1261. 

  1261. 		}
    1262. 

  1262. 

  1263. 			// process custom decompositions
    1264. 

  1264. 		$customTranslitFile = PATH_t3lib . 'unidata/Translit.txt';
    1265. 

  1265. 		if (t3lib_div::validPathStr($customTranslitFile) && @is_file($customTranslitFile)) {
    1266. 

  1266. 			$fh = fopen($customTranslitFile, 'rb');
    1267. 

  1267. 			if ($fh) {
    1268. 

  1268. 				while (!feof($fh)) {
    1269. 

  1269. 					$line = fgets($fh, 4096);
    1270. 

  1270. 					if ($line{0} != '#' && trim($line) != '') {
    1271. 

  1271. 						list($char, $translit) = t3lib_div::trimExplode(';', $line);
    1272. 

  1272. 						if (!$translit) {
    1273. 

  1273. 							$omit["U+$char"] = 1;
    1274. 

  1274. 						}
    1275. 

  1275. 						$decomposition["U+$char"] = explode(' ', $translit);
    1276. 

  1276. 

  1277. 					}
    1278. 

  1278. 				}
    1279. 

  1279. 				fclose($fh);
    1280. 

  1280. 			}
    1281. 

  1281. 		}
    1282. 

  1282. 

  1283. 			// decompose and remove marks; inspired by unac (Loic Dachary <loic@senga.org>)
    1284. 

  1284. 		foreach ($decomposition as $from => $to) {
    1285. 

  1285. 			$code_decomp = array();
    1286. 

  1286. 

  1287. 			while ($code_value = array_shift($to)) {
    1288. 

  1288. 				if (isset($decomposition["U+$code_value"])) { // do recursive decomposition
    1289. 

  1289. 					foreach (array_reverse($decomposition["U+$code_value"]) as $cv) {
    1290. 

  1290. 						array_unshift($to, $cv);
    1291. 

  1291. 					}
    1292. 

  1292. 				} elseif (!isset($mark["U+$code_value"])) { // remove mark
    1293. 

  1293. 					array_push($code_decomp, $code_value);
    1294. 

  1294. 				}
    1295. 

  1295. 			}
    1296. 

  1296. 			if (count($code_decomp) || isset($omit[$from])) {
    1297. 

  1297. 				$decomposition[$from] = $code_decomp;
    1298. 

  1298. 			} else {
    1299. 

  1299. 				unset($decomposition[$from]);
    1300. 

  1300. 			}
    1301. 

  1301. 		}
    1302. 

  1302. 

  1303. 			// create ascii only mapping
    1304. 

  1304. 		$this->toASCII['utf-8'] = array();
    1305. 

  1305. 		$ascii =& $this->toASCII['utf-8'];
    1306. 

  1306. 

  1307. 		foreach ($decomposition as $from => $to) {
    1308. 

  1308. 			$code_decomp = array();
    1309. 

  1309. 			while ($code_value = array_shift($to)) {
    1310. 

  1310. 				$ord = hexdec($code_value);
    1311. 

  1311. 				if ($ord > 127) {
    1312. 

  1312. 					continue 2;
    1313. 

  1313. 				} // skip decompositions containing non-ASCII chars
    1314. 

  1314. 				else
    1315. 

  1315. 				{
    1316. 

  1316. 					array_push($code_decomp, chr($ord));
    1317. 

  1317. 				}
    1318. 

  1318. 			}
    1319. 

  1319. 			$ascii[$this->UnumberToChar(hexdec($from))] = join('', $code_decomp);
    1320. 

  1320. 		}
    1321. 

  1321. 

  1322. 			// add numeric decompositions
    1323. 

  1323. 		foreach ($number as $from => $to) {
    1324. 

  1324. 			$utf8_char = $this->UnumberToChar(hexdec($from));
    1325. 

  1325. 			if (!isset($ascii[$utf8_char])) {
    1326. 

  1326. 				$ascii[$utf8_char] = $to;
    1327. 

  1327. 			}
    1328. 

  1328. 		}
    1329. 

  1329. 

  1330. 		if ($cacheFileCase) {
    1331. 

  1331. 			t3lib_div::writeFileToTypo3tempDir($cacheFileCase, serialize($utf8CaseFolding));
    1332. 

  1332. 		}
    1333. 

  1333. 

  1334. 		if ($cacheFileASCII) {
    1335. 

  1335. 			t3lib_div::writeFileToTypo3tempDir($cacheFileASCII, serialize($ascii));
    1336. 

  1336. 		}
    1337. 

  1337. 

  1338. 		return 3;
    1339. 

  1339. 	}
    1340. 

  1340. 

  1341. 	/**
    1342. 

  1342. 	 * This function initializes the folding table for a charset other than UTF-8.
    1343. 

  1343. 	 * This function is automatically called by the case folding functions.
    1344. 

  1344. 	 *
    1345. 

  1345. 	 * @param	string		Charset for which to initialize case folding.
    1346. 

  1346. 	 * @return	integer		Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached).
    1347. 

  1347. 	 * @access private
    1348. 

  1348. 	 */
    1349. 

  1349. 	function initCaseFolding($charset) {
    1350. 

  1350. 			// Only process if the case table is not yet loaded:
    1351. 

  1351. 		if (is_array($this->caseFolding[$charset])) {
    1352. 

  1352. 			return 1;
    1353. 

  1353. 		}
    1354. 

  1354. 

  1355. 			// Use cached version if possible
    1356. 

  1356. 		$cacheFile = t3lib_div::getFileAbsFileName('typo3temp/cs/cscase_' . $charset . '.tbl');
    1357. 

  1357. 		if ($cacheFile && @is_file($cacheFile)) {
    1358. 

  1358. 			$this->caseFolding[$charset] = unserialize(t3lib_div::getUrl($cacheFile));
    1359. 

  1359. 			return 2;
    1360. 

  1360. 		}
    1361. 

  1361. 

  1362. 			// init UTF-8 conversion for this charset
    1363. 

  1363. 		if (!$this->initCharset($charset)) {
    1364. 

  1364. 			return FALSE;
    1365. 

  1365. 		}
    1366. 

  1366. 

  1367. 			// UTF-8 case folding is used as the base conversion table
    1368. 

  1368. 		if (!$this->initUnicodeData('case')) {
    1369. 

  1369. 			return FALSE;
    1370. 

  1370. 		}
    1371. 

  1371. 

  1372. 		$nochar = chr($this->noCharByteVal);
    1373. 

  1373. 		foreach ($this->parsedCharsets[$charset]['local'] as $ci => $utf8) {
    1374. 

  1374. 				// reconvert to charset (don't use chr() of numeric value, might be muli-byte)
    1375. 

  1375. 			$c = $this->utf8_decode($utf8, $charset);
    1376. 

  1376. 

  1377. 				// $cc = $this->conv($this->caseFolding['utf-8']['toUpper'][$utf8], 'utf-8', $charset);
    1378. 

  1378. 			$cc = $this->utf8_decode($this->caseFolding['utf-8']['toUpper'][$utf8], $charset);
    1379. 

  1379. 			if ($cc != '' && $cc != $nochar) {
    1380. 

  1380. 				$this->caseFolding[$charset]['toUpper'][$c] = $cc;
    1381. 

  1381. 			}
    1382. 

  1382. 

  1383. 				// $cc = $this->conv($this->caseFolding['utf-8']['toLower'][$utf8], 'utf-8', $charset);
    1384. 

  1384. 			$cc = $this->utf8_decode($this->caseFolding['utf-8']['toLower'][$utf8], $charset);
    1385. 

  1385. 			if ($cc != '' && $cc != $nochar) {
    1386. 

  1386. 				$this->caseFolding[$charset]['toLower'][$c] = $cc;
    1387. 

  1387. 			}
    1388. 

  1388. 

  1389. 				// $cc = $this->conv($this->caseFolding['utf-8']['toTitle'][$utf8], 'utf-8', $charset);
    1390. 

  1390. 			$cc = $this->utf8_decode($this->caseFolding['utf-8']['toTitle'][$utf8], $charset);
    1391. 

  1391. 			if ($cc != '' && $cc != $nochar) {
    1392. 

  1392. 				$this->caseFolding[$charset]['toTitle'][$c] = $cc;
    1393. 

  1393. 			}
    1394. 

  1394. 		}
    1395. 

  1395. 

  1396. 			// add the ASCII case table
    1397. 

  1397. 		for ($i = ord('a'); $i <= ord('z'); $i++) {
    1398. 

  1398. 			$this->caseFolding[$charset]['toUpper'][chr($i)] = chr($i - 32);
    1399. 

  1399. 		}
    1400. 

  1400. 		for ($i = ord('A'); $i <= ord('Z'); $i++) {
    1401. 

  1401. 			$this->caseFolding[$charset]['toLower'][chr($i)] = chr($i + 32);
    1402. 

  1402. 		}
    1403. 

  1403. 

  1404. 		if ($cacheFile) {
    1405. 

  1405. 			t3lib_div::writeFileToTypo3tempDir($cacheFile, serialize($this->caseFolding[$charset]));
    1406. 

  1406. 		}
    1407. 

  1407. 

  1408. 		return 3;
    1409. 

  1409. 	}
    1410. 

  1410. 

  1411. 	/**
    1412. 

  1412. 	 * This function initializes the to-ASCII conversion table for a charset other than UTF-8.
    1413. 

  1413. 	 * This function is automatically called by the ASCII transliteration functions.
    1414. 

  1414. 	 *
    1415. 

  1415. 	 * @param	string		Charset for which to initialize conversion.
    1416. 

  1416. 	 * @return	integer		Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached).
    1417. 

  1417. 	 * @access private
    1418. 

  1418. 	 */
    1419. 

  1419. 	function initToASCII($charset) {
    1420. 

  1420. 			// Only process if the case table is not yet loaded:
    1421. 

  1421. 		if (is_array($this->toASCII[$charset])) {
    1422. 

  1422. 			return 1;
    1423. 

  1423. 		}
    1424. 

  1424. 

  1425. 			// Use cached version if possible
    1426. 

  1426. 		$cacheFile = t3lib_div::getFileAbsFileName('typo3temp/cs/csascii_' . $charset . '.tbl');
    1427. 

  1427. 		if ($cacheFile && @is_file($cacheFile)) {
    1428. 

  1428. 			$this->toASCII[$charset] = unserialize(t3lib_div::getUrl($cacheFile));
    1429. 

  1429. 			return 2;
    1430. 

  1430. 		}
    1431. 

  1431. 

  1432. 			// init UTF-8 conversion for this charset
    1433. 

  1433. 		if (!$this->initCharset($charset)) {
    1434. 

  1434. 			return FALSE;
    1435. 

  1435. 		}
    1436. 

  1436. 

  1437. 			// UTF-8/ASCII transliteration is used as the base conversion table
    1438. 

  1438. 		if (!$this->initUnicodeData('ascii')) {
    1439. 

  1439. 			return FALSE;
    1440. 

  1440. 		}
    1441. 

  1441. 

  1442. 		$nochar = chr($this->noCharByteVal);
    1443. 

  1443. 		foreach ($this->parsedCharsets[$charset]['local'] as $ci => $utf8) {
    1444. 

  1444. 				// reconvert to charset (don't use chr() of numeric value, might be muli-byte)
    1445. 

  1445. 			$c = $this->utf8_decode($utf8, $charset);
    1446. 

  1446. 

  1447. 			if (isset($this->toASCII['utf-8'][$utf8])) {
    1448. 

  1448. 				$this->toASCII[$charset][$c] = $this->toASCII['utf-8'][$utf8];
    1449. 

  1449. 			}
    1450. 

  1450. 		}
    1451. 

  1451. 

  1452. 		if ($cacheFile) {
    1453. 

  1453. 			t3lib_div::writeFileToTypo3tempDir($cacheFile, serialize($this->toASCII[$charset]));
    1454. 

  1454. 		}
    1455. 

  1455. 

  1456. 		return 3;
    1457. 

  1457. 	}
    1458. 

  1458. 

  1459. 

  1460. 	/********************************************
    1461. 

  1461. 	 *
    1462. 

  1462. 	 * String operation functions
    1463. 

  1463. 	 *
    1464. 

  1464. 	 ********************************************/
    1465. 

  1465. 

  1466. 	/**
    1467. 

  1467. 	 * Returns a part of a string.
    1468. 

  1468. 	 * Unit-tested by Kasper (single byte charsets only)
    1469. 

  1469. 	 *
    1470. 

  1470. 	 * @param	string		The character set
    1471. 

  1471. 	 * @param	string		Character string
    1472. 

  1472. 	 * @param	integer		Start position (character position)
    1473. 

  1473. 	 * @param	integer		Length (in characters)
    1474. 

  1474. 	 * @return	string		The substring
    1475. 

  1475. 	 * @see substr(), mb_substr()
    1476. 

  1476. 	 * @author	Martin Kutschker <martin.t.kutschker@blackbox.net>
    1477. 

  1477. 	 */
    1478. 

  1478. 	function substr($charset, $string, $start, $len = NULL) {
    1479. 

  1479. 		if ($len === 0 || $string === '') {
    1480. 

  1480. 			return '';
    1481. 

  1481. 		}
    1482. 

  1482. 

  1483. 		if ($GLOBALS['TYPO3_CONF_VARS']['SYS']['t3lib_cs_utils'] == 'mbstring') {
    1484. 

  1484. 				// cannot omit $len, when specifying charset
    1485. 

  1485. 			if ($len == NULL) {
    1486. 

  1486. 				$enc = mb_internal_encoding(); // save internal encoding
    1487. 

  1487. 				mb_internal_encoding($charset);
    1488. 

  1488. 				$str = mb_substr($string, $start);
    1489. 

  1489. 				mb_internal_encoding($enc); // restore internal encoding
    1490. 

  1490. 

  1491. 				return $str;
    1492. 

  1492. 			}
    1493. 

  1493. 			else {
    1494. 

  1494. 				return mb_substr($string, $start, $len, $charset);
    1495. 

  1495. 			}
    1496. 

  1496. 		} elseif ($GLOBALS['TYPO3_CONF_VARS']['SYS']['t3lib_cs_utils'] == 'iconv') {
    1497. 

  1497. 				// cannot omit $len, when specifying charset
    1498. 

  1498. 			if ($len == NULL) {
    1499. 

  1499. 				$enc = iconv_get_encoding('internal_encoding'); // …
    1500.

Large files files are truncated, but you can click here to view the full file