TYPO3 /
TYPO3.CMS
| Total Complexity | 143 |
| Total Lines | 775 |
| Duplicated Lines | 0 % |
| Changes | 0 | ||
Complex classes like CharsetConverter often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use CharsetConverter, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 52 | class CharsetConverter implements SingletonInterface |
||
| 53 | { |
||
| 54 | /** |
||
| 55 | * ASCII Value for chars with no equivalent. |
||
| 56 | * |
||
| 57 | * @var int |
||
| 58 | */ |
||
| 59 | protected $noCharByteVal = 63; |
||
| 60 | |||
| 61 | /** |
||
| 62 | * This is the array where parsed conversion tables are stored (cached) |
||
| 63 | * |
||
| 64 | * @var array |
||
| 65 | */ |
||
| 66 | protected $parsedCharsets = []; |
||
| 67 | |||
| 68 | /** |
||
| 69 | * An array where charset-to-ASCII mappings are stored (cached) |
||
| 70 | * |
||
| 71 | * @var array |
||
| 72 | */ |
||
| 73 | protected $toASCII = []; |
||
| 74 | |||
| 75 | /** |
||
| 76 | * This tells the converter which charsets has two bytes per char: |
||
| 77 | * |
||
| 78 | * @var array |
||
| 79 | */ |
||
| 80 | protected $twoByteSets = [ |
||
| 81 | 'ucs-2' => 1 |
||
| 82 | ]; |
||
| 83 | |||
| 84 | /** |
||
| 85 | * This tells the converter which charsets use a scheme like the Extended Unix Code: |
||
| 86 | * |
||
| 87 | * @var array |
||
| 88 | */ |
||
| 89 | protected $eucBasedSets = [ |
||
| 90 | 'gb2312' => 1, // Chinese, simplified. |
||
| 91 | 'big5' => 1, // Chinese, traditional. |
||
| 92 | 'euc-kr' => 1, // Korean |
||
| 93 | 'shift_jis' => 1 |
||
| 94 | ]; |
||
| 95 | |||
| 96 | /******************************************** |
||
| 97 | * |
||
| 98 | * Charset Conversion functions |
||
| 99 | * |
||
| 100 | ********************************************/ |
||
| 101 | /** |
||
| 102 | * Convert from one charset to another charset. |
||
| 103 | * |
||
| 104 | * @param string $inputString Input string |
||
| 105 | * @param string $fromCharset From charset (the current charset of the string) |
||
| 106 | * @param string $toCharset To charset (the output charset wanted) |
||
| 107 | * @return string Converted string |
||
| 108 | */ |
||
| 109 | public function conv($inputString, $fromCharset, $toCharset) |
||
| 110 | { |
||
| 111 | if ($fromCharset === $toCharset) { |
||
| 112 | return $inputString; |
||
| 113 | } |
||
| 114 | // PHP-libs don't support fallback to SGML entities, but UTF-8 handles everything |
||
| 115 | if ($toCharset === 'utf-8') { |
||
| 116 | // Returns FALSE for unsupported charsets |
||
| 117 | $convertedString = mb_convert_encoding($inputString, $toCharset, $fromCharset); |
||
| 118 | if (false !== $convertedString) { |
||
| 119 | return $convertedString; |
||
| 120 | } |
||
| 121 | } |
||
| 122 | if ($fromCharset !== 'utf-8') { |
||
| 123 | $inputString = $this->utf8_encode($inputString, $fromCharset); |
||
| 124 | } |
||
| 125 | if ($toCharset !== 'utf-8') { |
||
| 126 | $inputString = $this->utf8_decode($inputString, $toCharset, true); |
||
| 127 | } |
||
| 128 | return $inputString; |
||
| 129 | } |
||
| 130 | |||
| 131 | /** |
||
| 132 | * Converts $str from $charset to UTF-8 |
||
| 133 | * |
||
| 134 | * @param string $str String in local charset to convert to UTF-8 |
||
| 135 | * @param string $charset Charset, lowercase. Must be found in csconvtbl/ folder. |
||
| 136 | * @return string Output string, converted to UTF-8 |
||
| 137 | */ |
||
| 138 | public function utf8_encode($str, $charset) |
||
|
Coding Style
introduced
by
|
|||
| 139 | { |
||
| 140 | if ($charset === 'utf-8') { |
||
| 141 | return $str; |
||
| 142 | } |
||
| 143 | // Charset is case-insensitive |
||
| 144 | // Parse conv. table if not already |
||
| 145 | if ($this->initCharset($charset)) { |
||
| 146 | $strLen = strlen($str); |
||
| 147 | $outStr = ''; |
||
| 148 | // Traverse each char in string |
||
| 149 | for ($a = 0; $a < $strLen; $a++) { |
||
| 150 | $chr = substr($str, $a, 1); |
||
| 151 | $ord = ord($chr); |
||
| 152 | // If the charset has two bytes per char |
||
| 153 | if (isset($this->twoByteSets[$charset])) { |
||
| 154 | $ord2 = ord($str[$a + 1]); |
||
| 155 | // Assume big endian |
||
| 156 | $ord = $ord << 8 | $ord2; |
||
| 157 | // If the local char-number was found in parsed conv. table then we use that, otherwise 127 (no char?) |
||
| 158 | if (isset($this->parsedCharsets[$charset]['local'][$ord])) { |
||
| 159 | $outStr .= $this->parsedCharsets[$charset]['local'][$ord]; |
||
| 160 | } else { |
||
| 161 | $outStr .= chr($this->noCharByteVal); |
||
| 162 | } |
||
| 163 | // No char exists |
||
| 164 | $a++; |
||
| 165 | } elseif ($ord > 127) { |
||
| 166 | // If char has value over 127 it's a multibyte char in UTF-8 |
||
| 167 | // EUC uses two-bytes above 127; we get both and advance pointer and make $ord a 16bit int. |
||
| 168 | if (isset($this->eucBasedSets[$charset])) { |
||
| 169 | // Shift-JIS: chars between 160 and 223 are single byte |
||
| 170 | if ($charset !== 'shift_jis' || ($ord < 160 || $ord > 223)) { |
||
| 171 | $a++; |
||
| 172 | $ord2 = ord(substr($str, $a, 1)); |
||
| 173 | $ord = $ord * 256 + $ord2; |
||
| 174 | } |
||
| 175 | } |
||
| 176 | if (isset($this->parsedCharsets[$charset]['local'][$ord])) { |
||
| 177 | // If the local char-number was found in parsed conv. table then we use that, otherwise 127 (no char?) |
||
| 178 | $outStr .= $this->parsedCharsets[$charset]['local'][$ord]; |
||
| 179 | } else { |
||
| 180 | $outStr .= chr($this->noCharByteVal); |
||
| 181 | } |
||
| 182 | } else { |
||
| 183 | $outStr .= $chr; |
||
| 184 | } |
||
| 185 | } |
||
| 186 | return $outStr; |
||
| 187 | } |
||
| 188 | return ''; |
||
| 189 | } |
||
| 190 | |||
| 191 | /** |
||
| 192 | * Converts $str from UTF-8 to $charset |
||
| 193 | * |
||
| 194 | * @param string $str String in UTF-8 to convert to local charset |
||
| 195 | * @param string $charset Charset, lowercase. Must be found in csconvtbl/ folder. |
||
| 196 | * @param bool $useEntityForNoChar If set, then characters that are not available in the destination character set will be encoded as numeric entities |
||
| 197 | * @return string Output string, converted to local charset |
||
| 198 | */ |
||
| 199 | public function utf8_decode($str, $charset, $useEntityForNoChar = false) |
||
| 200 | { |
||
| 201 | if ($charset === 'utf-8') { |
||
| 202 | return $str; |
||
| 203 | } |
||
| 204 | // Charset is case-insensitive. |
||
| 205 | // Parse conv. table if not already |
||
| 206 | if ($this->initCharset($charset)) { |
||
| 207 | $strLen = strlen($str); |
||
| 208 | $outStr = ''; |
||
| 209 | // Traverse each char in UTF-8 string |
||
| 210 | for ($a = 0, $i = 0; $a < $strLen; $a++, $i++) { |
||
| 211 | $chr = substr($str, $a, 1); |
||
| 212 | $ord = ord($chr); |
||
| 213 | // This means multibyte! (first byte!) |
||
| 214 | if ($ord > 127) { |
||
| 215 | // Since the first byte must have the 7th bit set we check that. Otherwise we might be in the middle of a byte sequence. |
||
| 216 | if ($ord & 64) { |
||
| 217 | // Add first byte |
||
| 218 | $buf = $chr; |
||
| 219 | // For each byte in multibyte string |
||
| 220 | for ($b = 0; $b < 8; $b++) { |
||
| 221 | // Shift it left and |
||
| 222 | $ord = $ord << 1; |
||
| 223 | // ... and with 8th bit - if that is set, then there are still bytes in sequence. |
||
| 224 | if ($ord & 128) { |
||
| 225 | $a++; |
||
| 226 | // ... and add the next char. |
||
| 227 | $buf .= substr($str, $a, 1); |
||
| 228 | } else { |
||
| 229 | break; |
||
| 230 | } |
||
| 231 | } |
||
| 232 | // If the UTF-8 char-sequence is found then... |
||
| 233 | if (isset($this->parsedCharsets[$charset]['utf8'][$buf])) { |
||
| 234 | // The local number |
||
| 235 | $mByte = $this->parsedCharsets[$charset]['utf8'][$buf]; |
||
| 236 | // If the local number is greater than 255 we will need to split the byte (16bit word assumed) in two chars. |
||
| 237 | if ($mByte > 255) { |
||
| 238 | $outStr .= chr($mByte >> 8 & 255) . chr($mByte & 255); |
||
| 239 | } else { |
||
| 240 | $outStr .= chr($mByte); |
||
| 241 | } |
||
| 242 | } elseif ($useEntityForNoChar) { |
||
| 243 | // Create num entity: |
||
| 244 | $outStr .= '&#' . $this->utf8CharToUnumber($buf, true) . ';'; |
||
| 245 | } else { |
||
| 246 | $outStr .= chr($this->noCharByteVal); |
||
| 247 | } |
||
| 248 | } else { |
||
| 249 | $outStr .= chr($this->noCharByteVal); |
||
| 250 | } |
||
| 251 | } else { |
||
| 252 | $outStr .= $chr; |
||
| 253 | } |
||
| 254 | } |
||
| 255 | return $outStr; |
||
| 256 | } |
||
| 257 | return ''; |
||
| 258 | } |
||
| 259 | |||
| 260 | /** |
||
| 261 | * Converts all chars in the input UTF-8 string into integer numbers returned in an array. |
||
| 262 | * All HTML entities (like & or £ or { or 㽝) will be detected as characters. |
||
| 263 | * Also, instead of integer numbers the real UTF-8 char is returned. |
||
| 264 | * |
||
| 265 | * @param string $str Input string, UTF-8 |
||
| 266 | * @return array Output array with the char numbers |
||
| 267 | */ |
||
| 268 | public function utf8_to_numberarray($str) |
||
| 269 | { |
||
| 270 | // Entities must be registered as well |
||
| 271 | $str = html_entity_decode($str, ENT_COMPAT, 'utf-8'); |
||
| 272 | |||
| 273 | // Do conversion: |
||
| 274 | $strLen = strlen($str); |
||
| 275 | $outArr = []; |
||
| 276 | // Traverse each char in UTF-8 string. |
||
| 277 | for ($a = 0; $a < $strLen; $a++) { |
||
| 278 | $chr = substr($str, $a, 1); |
||
| 279 | $ord = ord($chr); |
||
| 280 | // This means multibyte! (first byte!) |
||
| 281 | if ($ord > 127) { |
||
| 282 | // Since the first byte must have the 7th bit set we check that. Otherwise we might be in the middle of a byte sequence. |
||
| 283 | if ($ord & 64) { |
||
| 284 | // Add first byte |
||
| 285 | $buf = $chr; |
||
| 286 | // For each byte in multibyte string... |
||
| 287 | for ($b = 0; $b < 8; $b++) { |
||
| 288 | // Shift it left and ... |
||
| 289 | $ord = $ord << 1; |
||
| 290 | // ... and with 8th bit - if that is set, then there are still bytes in sequence. |
||
| 291 | if ($ord & 128) { |
||
| 292 | $a++; |
||
| 293 | // ... and add the next char. |
||
| 294 | $buf .= substr($str, $a, 1); |
||
| 295 | } else { |
||
| 296 | break; |
||
| 297 | } |
||
| 298 | } |
||
| 299 | $outArr[] = $buf; |
||
| 300 | } else { |
||
| 301 | $outArr[] = chr($this->noCharByteVal); |
||
| 302 | } |
||
| 303 | } else { |
||
| 304 | $outArr[] = chr($ord); |
||
| 305 | } |
||
| 306 | } |
||
| 307 | return $outArr; |
||
| 308 | } |
||
| 309 | |||
| 310 | /** |
||
| 311 | * Converts a UNICODE number to a UTF-8 multibyte character |
||
| 312 | * Algorithm based on script found at From: http://czyborra.com/utf/ |
||
| 313 | * Unit-tested by Kasper |
||
| 314 | * |
||
| 315 | * The binary representation of the character's integer value is thus simply spread across the bytes |
||
| 316 | * and the number of high bits set in the lead byte announces the number of bytes in the multibyte sequence: |
||
| 317 | * |
||
| 318 | * bytes | bits | representation |
||
| 319 | * 1 | 7 | 0vvvvvvv |
||
| 320 | * 2 | 11 | 110vvvvv 10vvvvvv |
||
| 321 | * 3 | 16 | 1110vvvv 10vvvvvv 10vvvvvv |
||
| 322 | * 4 | 21 | 11110vvv 10vvvvvv 10vvvvvv 10vvvvvv |
||
| 323 | * 5 | 26 | 111110vv 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv |
||
| 324 | * 6 | 31 | 1111110v 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv |
||
| 325 | * |
||
| 326 | * @param int $unicodeInteger UNICODE integer |
||
| 327 | * @return string UTF-8 multibyte character string |
||
| 328 | * @see utf8CharToUnumber() |
||
| 329 | */ |
||
| 330 | public function UnumberToChar($unicodeInteger) |
||
| 331 | { |
||
| 332 | $str = ''; |
||
| 333 | if ($unicodeInteger < 128) { |
||
| 334 | $str .= chr($unicodeInteger); |
||
| 335 | } elseif ($unicodeInteger < 2048) { |
||
| 336 | $str .= chr(192 | $unicodeInteger >> 6); |
||
| 337 | $str .= chr(128 | $unicodeInteger & 63); |
||
| 338 | } elseif ($unicodeInteger < 65536) { |
||
| 339 | $str .= chr(224 | $unicodeInteger >> 12); |
||
| 340 | $str .= chr(128 | $unicodeInteger >> 6 & 63); |
||
| 341 | $str .= chr(128 | $unicodeInteger & 63); |
||
| 342 | } elseif ($unicodeInteger < 2097152) { |
||
| 343 | $str .= chr(240 | $unicodeInteger >> 18); |
||
| 344 | $str .= chr(128 | $unicodeInteger >> 12 & 63); |
||
| 345 | $str .= chr(128 | $unicodeInteger >> 6 & 63); |
||
| 346 | $str .= chr(128 | $unicodeInteger & 63); |
||
| 347 | } elseif ($unicodeInteger < 67108864) { |
||
| 348 | $str .= chr(248 | $unicodeInteger >> 24); |
||
| 349 | $str .= chr(128 | $unicodeInteger >> 18 & 63); |
||
| 350 | $str .= chr(128 | $unicodeInteger >> 12 & 63); |
||
| 351 | $str .= chr(128 | $unicodeInteger >> 6 & 63); |
||
| 352 | $str .= chr(128 | $unicodeInteger & 63); |
||
| 353 | } elseif ($unicodeInteger < 2147483648) { |
||
| 354 | $str .= chr(252 | $unicodeInteger >> 30); |
||
| 355 | $str .= chr(128 | $unicodeInteger >> 24 & 63); |
||
| 356 | $str .= chr(128 | $unicodeInteger >> 18 & 63); |
||
| 357 | $str .= chr(128 | $unicodeInteger >> 12 & 63); |
||
| 358 | $str .= chr(128 | $unicodeInteger >> 6 & 63); |
||
| 359 | $str .= chr(128 | $unicodeInteger & 63); |
||
| 360 | } else { |
||
| 361 | // Cannot express a 32-bit character in UTF-8 |
||
| 362 | $str .= chr($this->noCharByteVal); |
||
| 363 | } |
||
| 364 | return $str; |
||
| 365 | } |
||
| 366 | |||
| 367 | /** |
||
| 368 | * Converts a UTF-8 Multibyte character to a UNICODE number |
||
| 369 | * Unit-tested by Kasper |
||
| 370 | * |
||
| 371 | * @param string $str UTF-8 multibyte character string |
||
| 372 | * @param bool $hex If set, then a hex. number is returned. |
||
| 373 | * @return int UNICODE integer |
||
| 374 | * @see UnumberToChar() |
||
| 375 | */ |
||
| 376 | public function utf8CharToUnumber($str, $hex = false) |
||
| 377 | { |
||
| 378 | // First char |
||
| 379 | $ord = ord($str[0]); |
||
| 380 | // This verifies that it IS a multi byte string |
||
| 381 | if (($ord & 192) === 192) { |
||
| 382 | $binBuf = ''; |
||
| 383 | $b = 0; |
||
| 384 | // For each byte in multibyte string... |
||
| 385 | for (; $b < 8; $b++) { |
||
| 386 | // Shift it left and ... |
||
| 387 | $ord = $ord << 1; |
||
| 388 | // ... and with 8th bit - if that is set, then there are still bytes in sequence. |
||
| 389 | if ($ord & 128) { |
||
| 390 | $binBuf .= substr('00000000' . decbin(ord(substr($str, $b + 1, 1))), -6); |
||
| 391 | } else { |
||
| 392 | break; |
||
| 393 | } |
||
| 394 | } |
||
| 395 | $binBuf = substr('00000000' . decbin(ord($str[0])), -(6 - $b)) . $binBuf; |
||
| 396 | $int = bindec($binBuf); |
||
| 397 | } else { |
||
| 398 | $int = $ord; |
||
| 399 | } |
||
| 400 | return $hex ? 'x' . dechex($int) : $int; |
||
| 401 | } |
||
| 402 | |||
| 403 | /******************************************** |
||
| 404 | * |
||
| 405 | * Init functions |
||
| 406 | * |
||
| 407 | ********************************************/ |
||
| 408 | /** |
||
| 409 | * This will initialize a charset for use if it's defined in the 'typo3/sysext/core/Resources/Private/Charsets/csconvtbl/' folder |
||
| 410 | * This function is automatically called by the conversion functions |
||
| 411 | * |
||
| 412 | * PLEASE SEE: http://www.unicode.org/Public/MAPPINGS/ |
||
| 413 | * |
||
| 414 | * @param string $charset The charset to be initialized. Use lowercase charset always (the charset must match exactly with a filename in csconvtbl/ folder ([charset].tbl) |
||
| 415 | * @return int Returns '1' if already loaded, '2' if the charset conversion table was found and parsed. |
||
| 416 | * @throws UnknownCharsetException if no charset table was found |
||
| 417 | */ |
||
| 418 | protected function initCharset($charset) |
||
| 419 | { |
||
| 420 | // Only process if the charset is not yet loaded: |
||
| 421 | if (empty($this->parsedCharsets[$charset])) { |
||
| 422 | // Conversion table filename: |
||
| 423 | $charsetConvTableFile = ExtensionManagementUtility::extPath('core') . 'Resources/Private/Charsets/csconvtbl/' . $charset . '.tbl'; |
||
| 424 | // If the conversion table is found: |
||
| 425 | if ($charset && GeneralUtility::validPathStr($charsetConvTableFile) && @is_file($charsetConvTableFile)) { |
||
| 426 | // Cache file for charsets: |
||
| 427 | // Caching brought parsing time for gb2312 down from 2400 ms to 150 ms. For other charsets we are talking 11 ms down to zero. |
||
| 428 | $cacheFile = Environment::getVarPath() . '/charset/charset_' . $charset . '.tbl'; |
||
| 429 | if ($cacheFile && @is_file($cacheFile)) { |
||
| 430 | $this->parsedCharsets[$charset] = unserialize(file_get_contents($cacheFile)); |
||
| 431 | } else { |
||
| 432 | // Parse conversion table into lines: |
||
| 433 | $lines = GeneralUtility::trimExplode(LF, file_get_contents($charsetConvTableFile), true); |
||
| 434 | // Initialize the internal variable holding the conv. table: |
||
| 435 | $this->parsedCharsets[$charset] = ['local' => [], 'utf8' => []]; |
||
| 436 | // traverse the lines: |
||
| 437 | $detectedType = ''; |
||
| 438 | foreach ($lines as $value) { |
||
| 439 | // Comment line or blanks are ignored. |
||
| 440 | if (trim($value) && $value[0] !== '#') { |
||
| 441 | // Detect type if not done yet: (Done on first real line) |
||
| 442 | // The "whitespaced" type is on the syntax "0x0A 0x000A #LINE FEED" while "ms-token" is like "B9 = U+00B9 : SUPERSCRIPT ONE" |
||
| 443 | if (!$detectedType) { |
||
| 444 | $detectedType = preg_match('/[[:space:]]*0x([[:alnum:]]*)[[:space:]]+0x([[:alnum:]]*)[[:space:]]+/', $value) ? 'whitespaced' : 'ms-token'; |
||
| 445 | } |
||
| 446 | $hexbyte = ''; |
||
| 447 | $utf8 = ''; |
||
| 448 | if ($detectedType === 'ms-token') { |
||
| 449 | list($hexbyte, $utf8) = preg_split('/[=:]/', $value, 3); |
||
| 450 | } elseif ($detectedType === 'whitespaced') { |
||
| 451 | $regA = []; |
||
| 452 | preg_match('/[[:space:]]*0x([[:alnum:]]*)[[:space:]]+0x([[:alnum:]]*)[[:space:]]+/', $value, $regA); |
||
| 453 | $hexbyte = $regA[1]; |
||
| 454 | $utf8 = 'U+' . $regA[2]; |
||
| 455 | } |
||
| 456 | $decval = hexdec(trim($hexbyte)); |
||
| 457 | if ($decval > 127) { |
||
| 458 | $utf8decval = hexdec(substr(trim($utf8), 2)); |
||
| 459 | $this->parsedCharsets[$charset]['local'][$decval] = $this->UnumberToChar($utf8decval); |
||
| 460 | $this->parsedCharsets[$charset]['utf8'][$this->parsedCharsets[$charset]['local'][$decval]] = $decval; |
||
| 461 | } |
||
| 462 | } |
||
| 463 | } |
||
| 464 | if ($cacheFile) { |
||
| 465 | GeneralUtility::writeFileToTypo3tempDir($cacheFile, serialize($this->parsedCharsets[$charset])); |
||
| 466 | } |
||
| 467 | } |
||
| 468 | return 2; |
||
| 469 | } |
||
| 470 | throw new UnknownCharsetException(sprintf('Unknown charset "%s"', $charset), 1508916031); |
||
| 471 | } |
||
| 472 | return 1; |
||
| 473 | } |
||
| 474 | |||
| 475 | /** |
||
| 476 | * This function initializes all UTF-8 character data tables. |
||
| 477 | * |
||
| 478 | * PLEASE SEE: http://www.unicode.org/Public/UNIDATA/ |
||
| 479 | * |
||
| 480 | * @return int Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached). |
||
| 481 | */ |
||
| 482 | protected function initUnicodeData() |
||
| 633 | } |
||
| 634 | |||
| 635 | /** |
||
| 636 | * This function initializes the to-ASCII conversion table for a charset other than UTF-8. |
||
| 637 | * This function is automatically called by the ASCII transliteration functions. |
||
| 638 | * |
||
| 639 | * @param string $charset Charset for which to initialize conversion. |
||
| 640 | * @return int Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached). |
||
| 641 | */ |
||
| 642 | protected function initToASCII($charset) |
||
| 643 | { |
||
| 644 | // Only process if the case table is not yet loaded: |
||
| 645 | if (isset($this->toASCII[$charset]) && is_array($this->toASCII[$charset])) { |
||
| 646 | return 1; |
||
| 647 | } |
||
| 648 | // Use cached version if possible |
||
| 649 | $cacheFile = Environment::getVarPath() . '/charset/csascii_' . $charset . '.tbl'; |
||
| 650 | if ($cacheFile && @is_file($cacheFile)) { |
||
| 651 | $this->toASCII[$charset] = unserialize(file_get_contents($cacheFile)); |
||
| 652 | return 2; |
||
| 653 | } |
||
| 654 | // Init UTF-8 conversion for this charset |
||
| 655 | if (!$this->initCharset($charset)) { |
||
| 656 | return false; |
||
| 657 | } |
||
| 658 | // UTF-8/ASCII transliteration is used as the base conversion table |
||
| 659 | if (!$this->initUnicodeData()) { |
||
| 660 | return false; |
||
| 661 | } |
||
| 662 | foreach ($this->parsedCharsets[$charset]['local'] as $ci => $utf8) { |
||
| 663 | // Reconvert to charset (don't use chr() of numeric value, might be muli-byte) |
||
| 664 | $c = $this->utf8_decode($utf8, $charset); |
||
| 665 | if (isset($this->toASCII['utf-8'][$utf8])) { |
||
| 666 | $this->toASCII[$charset][$c] = $this->toASCII['utf-8'][$utf8]; |
||
| 667 | } |
||
| 668 | } |
||
| 669 | if ($cacheFile) { |
||
| 670 | GeneralUtility::writeFileToTypo3tempDir($cacheFile, serialize($this->toASCII[$charset])); |
||
| 671 | } |
||
| 672 | return 3; |
||
| 673 | } |
||
| 674 | |||
| 675 | /******************************************** |
||
| 676 | * |
||
| 677 | * String operation functions |
||
| 678 | * |
||
| 679 | ********************************************/ |
||
| 680 | |||
| 681 | /** |
||
| 682 | * Converts special chars (like æøåÆØÅ, umlauts etc) to ascii equivalents (usually double-bytes, like æ => ae etc.) |
||
| 683 | * |
||
| 684 | * @param string $charset Character set of string |
||
| 685 | * @param string $string Input string to convert |
||
| 686 | * @return string The converted string |
||
| 687 | */ |
||
| 688 | public function specCharsToASCII($charset, $string) |
||
| 699 | } |
||
| 700 | |||
| 701 | /******************************************** |
||
| 702 | * |
||
| 703 | * Internal string operation functions |
||
| 704 | * |
||
| 705 | ********************************************/ |
||
| 706 | /** |
||
| 707 | * Maps all characters of a string in a single byte charset. |
||
| 708 | * |
||
| 709 | * @param string $str The string |
||
| 710 | * @param string $charset The charset |
||
| 711 | * @return string The converted string |
||
| 712 | */ |
||
| 713 | public function sb_char_mapping($str, $charset) |
||
| 714 | { |
||
| 715 | if (!$this->initToASCII($charset)) { |
||
| 716 | return $str; |
||
| 717 | } |
||
| 718 | // Do nothing |
||
| 719 | $map = &$this->toASCII[$charset]; |
||
| 720 | $out = ''; |
||
| 721 | for ($i = 0; isset($str[$i]); $i++) { |
||
| 722 | $c = $str[$i]; |
||
| 723 | if (isset($map[$c])) { |
||
| 724 | $out .= $map[$c]; |
||
| 725 | } else { |
||
| 726 | $out .= $c; |
||
| 727 | } |
||
| 728 | } |
||
| 729 | return $out; |
||
| 730 | } |
||
| 731 | |||
| 732 | /******************************************** |
||
| 733 | * |
||
| 734 | * Internal UTF-8 string operation functions |
||
| 735 | * |
||
| 736 | ********************************************/ |
||
| 737 | |||
| 738 | /** |
||
| 739 | * Maps all characters of an UTF-8 string. |
||
| 740 | * |
||
| 741 | * @param string $str UTF-8 string |
||
| 742 | * @return string The converted string |
||
| 743 | */ |
||
| 744 | public function utf8_char_mapping($str) |
||
| 745 | { |
||
| 746 | if (!$this->initUnicodeData()) { |
||
| 747 | // Do nothing |
||
| 748 | return $str; |
||
| 749 | } |
||
| 750 | $out = ''; |
||
| 751 | $map = &$this->toASCII['utf-8']; |
||
| 752 | for ($i = 0; isset($str[$i]); $i++) { |
||
| 753 | $c = ord($str[$i]); |
||
| 754 | $mbc = ''; |
||
| 755 | // single-byte (0xxxxxx) |
||
| 756 | if (!($c & 128)) { |
||
| 757 | $mbc = $str[$i]; |
||
| 758 | } elseif (($c & 192) === 192) { |
||
| 759 | $bc = 0; |
||
| 760 | // multi-byte starting byte (11xxxxxx) |
||
| 761 | for (; $c & 128; $c = $c << 1) { |
||
| 762 | $bc++; |
||
| 763 | } |
||
| 764 | // calculate number of bytes |
||
| 765 | $mbc = substr($str, $i, $bc); |
||
| 766 | $i += $bc - 1; |
||
| 767 | } |
||
| 768 | if (isset($map[$mbc])) { |
||
| 769 | $out .= $map[$mbc]; |
||
| 770 | } else { |
||
| 771 | $out .= $mbc; |
||
| 772 | } |
||
| 773 | } |
||
| 774 | return $out; |
||
| 775 | } |
||
| 776 | |||
| 777 | /******************************************** |
||
| 778 | * |
||
| 779 | * Internal EUC string operation functions |
||
| 780 | * |
||
| 781 | * Extended Unix Code: |
||
| 782 | * ASCII compatible 7bit single bytes chars |
||
| 783 | * 8bit two byte chars |
||
| 784 | * |
||
| 785 | * Shift-JIS is treated as a special case. |
||
| 786 | * |
||
| 787 | ********************************************/ |
||
| 788 | |||
| 789 | /** |
||
| 790 | * Maps all characters of a string in the EUC charset family. |
||
| 791 | * |
||
| 792 | * @param string $str EUC multibyte character string |
||
| 793 | * @param string $charset The charset |
||
| 794 | * @return string The converted string |
||
| 795 | */ |
||
| 796 | public function euc_char_mapping($str, $charset) |
||
| 827 | } |
||
| 828 | } |
||
| 829 |
Loading history...
