| blob | d22acf6617b2514b1aa4d68c8ef7cb0a40a4f488 |
1 <?php
2 /***************************************************************
3 * Copyright notice
4 *
5 * (c) 2003-2011 Kasper Skårhøj (kasperYYYY@typo3.com)
6 * All rights reserved
7 *
8 * This script is part of the Typo3 project. The Typo3 project is
9 * free software; you can redistribute it and/or modify
10 * it under the terms of the GNU General Public License as published by
11 * the Free Software Foundation; either version 2 of the License, or
12 * (at your option) any later version.
13 *
14 * The GNU General Public License can be found at
15 * http://www.gnu.org/copyleft/gpl.html.
16 *
17 * This script is distributed in the hope that it will be useful,
18 * but WITHOUT ANY WARRANTY; without even the implied warranty of
19 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
20 * GNU General Public License for more details.
21 *
22 * This copyright notice MUST APPEAR in all copies of the script!
23 ***************************************************************/
24 /**
25 * Class for conversion between charsets.
26 *
27 * @author Kasper Skårhøj <kasperYYYY@typo3.com>
28 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
29 */
32 /**
33 * Notes on UTF-8
34 *
35 * Functions working on UTF-8 strings:
36 *
37 * - strchr/strstr
38 * - strrchr
39 * - substr_count
40 * - implode/explode/join
41 *
42 * Functions nearly working on UTF-8 strings:
43 *
44 * - strlen: returns the length in BYTES, if you need the length in CHARACTERS use utf8_strlen
45 * - trim/ltrim/rtrim: the second parameter 'charlist' won't work for characters not contained in 7-bit ASCII
46 * - strpos/strrpos: they return the BYTE position, if you need the CHARACTER position use utf8_strpos/utf8_strrpos
47 * - htmlentities: charset support for UTF-8 only since PHP 4.3.0
48 * - preg_*: Support compiled into PHP by default nowadays, but could be unavailable, need to use modifier
49 *
50 * Functions NOT working on UTF-8 strings:
51 *
52 * - str*cmp
53 * - stristr
54 * - stripos
55 * - substr
56 * - strrev
57 * - split/spliti
58 * - ...
59 *
60 */
61 /**
62 * Class for conversion between charsets
63 *
64 * @author Kasper Skårhøj <kasperYYYY@typo3.com>
65 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
66 * @package TYPO3
67 * @subpackage t3lib
68 */
71 /**
72 * @var t3lib_l10n_Locales
73 */
78 // This is the array where parsed conversion tables are stored (cached)
81 // An array where case folding data will be stored (cached)
84 // An array where charset-to-ASCII mappings are stored (cached)
87 // This tells the converter which charsets has two bytes per char:
90 );
92 // This tells the converter which charsets has four bytes per char:
96 );
98 // This tells the converter which charsets use a scheme like the Extended Unix Code:
103 'shift_jis' => 1, // Japanese - WARNING: Shift-JIS includes half-width katakana single-bytes characters above 0x80!
104 );
106 // see http://developer.apple.com/documentation/macos8/TextIntlSvcs/TextEncodingConversionManager/TEC1.5/TEC.b0.html
107 // http://czyborra.com/charsets/iso8859.html
192 );
194 // mapping of iso-639-1 language codes to script names
196 // iso-639-1 language codes, see http://www.loc.gov/standards/iso639-2/php/code_list.php
246 // MS language codes, see http://msdn.microsoft.com/library/default.asp?url=/library/en-us/vclib/html/_crt_language_strings.asp
247 // http://msdn.microsoft.com/library/default.asp?url=/library/en-us/wceinternational5/html/wce50conLanguageIdentifiersandLocales.asp
308 // English language names
356 );
358 // mapping of language (family) names to charsets on Unix
379 );
381 // mapping of language (family) names to charsets on Windows
402 );
404 // mapping of locale names to charsets
413 );
415 // TYPO3 specific: Array with the system charsets used for each system language in TYPO3:
416 // Empty values means "iso-8859-1"
484 );
486 // TYPO3 specific: Array with the iso names used for each system language in TYPO3:
487 // Missing keys means: same as TYPO3
488 // @deprecated since TYPO3 4.6, will be removed in TYPO3 6.0 - use t3lib_l10n_Locales::getIsoMapping()
507 );
509 /**
510 * Default constructor.
511 */
514 }
516 /**
517 * Normalize - changes input character set to lowercase letters.
518 *
519 * @param string Input charset
520 * @return string Normalized charset
521 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
522 */
527 }
530 }
532 /**
533 * Get the charset of a locale.
534 *
535 * ln language
536 * ln_CN language / country
537 * ln_CN.cs language / country / charset
538 * ln_CN.cs@mod language / country / charset / modifier
539 *
540 * @param string Locale string
541 * @return string Charset resolved for locale string
542 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
543 */
547 // exact locale specific charset?
550 }
552 // get modifier
555 // locale contains charset: use it
559 }
561 // modifier is 'euro' (after charset check, because of xx.utf-8@euro)
564 }
566 // get language
570 }
573 $cs = $this->script_to_charset_windows[$script] ? $this->script_to_charset_windows[$script] : 'windows-1252';
575 $cs = $this->script_to_charset_unix[$script] ? $this->script_to_charset_unix[$script] : 'utf-8';
576 }
579 }
582 /********************************************
583 *
584 * Charset Conversion functions
585 *
586 ********************************************/
588 /**
589 * Convert from one charset to another charset.
590 *
591 * @param string Input string
592 * @param string From charset (the current charset of the string)
593 * @param string To charset (the output charset wanted)
594 * @param boolean If set, then characters that are not available in the destination character set will be encoded as numeric entities
595 * @return string Converted string
596 * @see convArray()
597 */
601 }
603 // PHP-libs don't support fallback to SGML entities, but UTF-8 handles everything
617 }
624 }
626 }
627 // fallback to TYPO3 conversion
628 }
632 }
635 }
637 }
639 /**
640 * Convert all elements in ARRAY with type string from one charset to another charset.
641 * NOTICE: Array is passed by reference!
642 *
643 * @param string Input array, possibly multidimensional
644 * @param string From charset (the current charset of the string)
645 * @param string To charset (the output charset wanted)
646 * @param boolean If set, then characters that are not available in the destination character set will be encoded as numeric entities
647 * @return void
648 * @see conv()
649 */
656 }
657 }
658 }
660 /**
661 * Converts $str from $charset to UTF-8
662 *
663 * @param string String in local charset to convert to UTF-8
664 * @param string Charset, lowercase. Must be found in csconvtbl/ folder.
665 * @return string Output string, converted to UTF-8
666 */
671 }
673 // Charset is case-insensitive.
685 if (isset($this->parsedCharsets[$charset]['local'][$ord])) { // If the local char-number was found in parsed conv. table then we use that, otherwise 127 (no char?)
692 if (isset($this->eucBasedSets[$charset])) { // EUC uses two-bytes above 127; we get both and advance pointer and make $ord a 16bit int.
693 if ($charset != 'shift_jis' || ($ord < 0xA0 || $ord > 0xDF)) { // Shift-JIS: chars between 160 and 223 are single byte
697 }
698 }
700 if (isset($this->parsedCharsets[$charset]['local'][$ord])) { // If the local char-number was found in parsed conv. table then we use that, otherwise 127 (no char?)
708 }
710 }
711 }
713 /**
714 * Converts $str from UTF-8 to $charset
715 *
716 * @param string String in UTF-8 to convert to local charset
717 * @param string Charset, lowercase. Must be found in csconvtbl/ folder.
718 * @param boolean If set, then characters that are not available in the destination character set will be encoded as numeric entities
719 * @return string Output string, converted to local charset
720 */
725 }
727 // Charset is case-insensitive.
736 if ($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.
741 if ($ord & 128) { // ... and with 8th bit - if that is set, then there are still bytes in sequence.
746 }
747 }
749 if (isset($this->parsedCharsets[$charset]['utf8'][$buf])) { // If the UTF-8 char-sequence is found then...
751 if ($mByte > 255) { // If the local number is greater than 255 we will need to split the byte (16bit word assumed) in two chars.
755 }
767 }
769 }
770 }
772 /**
773 * Converts all chars > 127 to numeric entities.
774 *
775 * @param string Input string
776 * @return string Output string
777 */
786 if ($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.
790 if ($ord & 128) { // ... and with 8th bit - if that is set, then there are still bytes in sequence.
795 }
796 }
805 }
808 }
810 /**
811 * Converts numeric entities (UNICODE, eg. decimal (Ӓ) or hexadecimal ()) to UTF-8 multibyte chars
812 *
813 * @param string Input string, UTF-8
814 * @param boolean If set, then all string-HTML entities (like & or £ will be converted as well)
815 * @return string Output string
816 */
818 // Workaround for #39287: 3rd parameter for get_html_translation_table() was only added in PHP 5.3.4 and later
819 // see http://php.net/manual/en/function.get-html-translation-table.php
827 }
828 }
831 $parts = explode($token, preg_replace('/(&([#[:alnum:]]*);)/', $token . '${2}' . $token, $str));
833 // only take every second element
836 }
845 }
851 }
855 }
856 }
859 }
861 /**
862 * Converts all chars in the input UTF-8 string into integer numbers returned in an array
863 *
864 * @param string Input string, UTF-8
865 * @param boolean If set, then all HTML entities (like & or £ or { or 㽝) will be detected as characters.
866 * @param boolean If set, then instead of integer numbers the real UTF-8 char is returned.
867 * @return array Output array with the char numbers
868 */
870 // If entities must be registered as well...:
873 }
874 // Do conversion:
882 if ($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.
886 if ($ord & 128) { // ... and with 8th bit - if that is set, then there are still bytes in sequence.
891 }
892 }
901 }
904 }
906 /**
907 * Converts a UNICODE number to a UTF-8 multibyte character
908 * Algorithm based on script found at From: http://czyborra.com/utf/
909 * Unit-tested by Kasper
910 *
911 * The binary representation of the character's integer value is thus simply spread across the bytes and the number of high bits set in the lead byte announces the number of bytes in the multibyte sequence:
912 *
913 * bytes | bits | representation
914 * 1 | 7 | 0vvvvvvv
915 * 2 | 11 | 110vvvvv 10vvvvvv
916 * 3 | 16 | 1110vvvv 10vvvvvv 10vvvvvv
917 * 4 | 21 | 11110vvv 10vvvvvv 10vvvvvv 10vvvvvv
918 * 5 | 26 | 111110vv 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv
919 * 6 | 31 | 1111110v 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv 10vvvvvv
920 *
921 * @param integer UNICODE integer
922 * @return string UTF-8 multibyte character string
923 * @see utf8CharToUnumber()
924 */
962 }
963 }
964 }
965 }
966 }
967 }
969 }
971 /**
972 * Converts a UTF-8 Multibyte character to a UNICODE number
973 * Unit-tested by Kasper
974 *
975 * @param string UTF-8 multibyte character string
976 * @param boolean If set, then a hex. number is returned.
977 * @return integer UNICODE integer
978 * @see UnumberToChar()
979 */
987 if ($ord & 128) { // ... and with 8th bit - if that is set, then there are still bytes in sequence.
991 }
992 }
998 }
1001 }
1004 /********************************************
1005 *
1006 * Init functions
1007 *
1008 ********************************************/
1010 /**
1011 * This will initialize a charset for use if it's defined in the PATH_t3lib.'csconvtbl/' folder
1012 * This function is automatically called by the conversion functions
1013 *
1014 * PLEASE SEE: http://www.unicode.org/Public/MAPPINGS/
1015 *
1016 * @param string The charset to be initialized. Use lowercase charset always (the charset must match exactly with a filename in csconvtbl/ folder ([charset].tbl)
1017 * @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.
1018 * @access private
1019 */
1021 // Only process if the charset is not yet loaded:
1024 // Conversion table filename:
1027 // If the conversion table is found:
1028 if ($charset && t3lib_div::validPathStr($charsetConvTableFile) && @is_file($charsetConvTableFile)) {
1029 // Cache file for charsets:
1030 // Caching brought parsing time for gb2312 down from 2400 ms to 150 ms. For other charsets we are talking 11 ms down to zero.
1035 // Parse conversion table into lines:
1037 // Initialize the internal variable holding the conv. table:
1039 // traverse the lines:
1044 // Detect type if not done yet: (Done on first real line)
1045 // The "whitespaced" type is on the syntax "0x0A 0x000A #LINE FEED" while "ms-token" is like "B9 = U+00B9 : SUPERSCRIPT ONE"
1047 $detectedType = preg_match('/[[:space:]]*0x([[:alnum:]]*)[[:space:]]+0x([[:alnum:]]*)[[:space:]]+/', $value) ? 'whitespaced' : 'ms-token';
1048 }
1054 preg_match('/[[:space:]]*0x([[:alnum:]]*)[[:space:]]+0x([[:alnum:]]*)[[:space:]]+/', $value, $regA);
1057 }
1062 $this->parsedCharsets[$charset]['utf8'][$this->parsedCharsets[$charset]['local'][$decval]] = $decval;
1063 }
1064 }
1065 }
1068 }
1069 }
1073 }
1076 }
1077 }
1079 /**
1080 * This function initializes all UTF-8 character data tables.
1081 *
1082 * PLEASE SEE: http://www.unicode.org/Public/UNIDATA/
1083 *
1084 * @param string Mode ("case", "ascii", ...)
1085 * @return integer Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached).
1086 * @access private
1087 */
1089 // cache files
1093 // Only process if the tables are not yet loaded
1098 }
1100 // Use cached version if possible
1104 }
1110 }
1112 // Use cached version if possible
1116 }
1118 }
1120 // process main Unicode data file
1124 }
1129 }
1131 // key = utf8 char (single codepoint), value = utf8 string (codepoint sequence)
1132 // note: we use the UTF-8 characters here and not the Unicode numbers to avoid conversion roundtrip in utf8_strtolower/-upper)
1146 // has a lot of info
1147 list($char, $name, $cat, , , $decomp, , , $num, , , , $upper, $lower, $title,) = explode(';', rtrim($line));
1158 }
1161 }
1162 // store "title" only when different from "upper" (only a few)
1165 }
1175 }
1176 }
1178 // accented Latin letters without "official" decomposition
1184 }
1188 }
1204 }
1207 // ignore Arabic and vertical layout presentation decomposition
1214 }
1216 }
1217 }
1220 // process additional Unicode data for casing (allow folded characters to expand into a sequence)
1236 }
1238 }
1243 }
1245 }
1250 }
1252 }
1253 }
1254 }
1255 }
1257 }
1258 }
1260 // process custom decompositions
1271 }
1274 }
1275 }
1277 }
1278 }
1280 // decompose and remove marks; inspired by unac (Loic Dachary <loic@senga.org>)
1288 }
1291 }
1292 }
1297 }
1298 }
1300 // create ascii only mapping
1311 else
1312 {
1314 }
1315 }
1317 }
1319 // add numeric decompositions
1324 }
1325 }
1329 }
1333 }
1336 }
1338 /**
1339 * This function initializes the folding table for a charset other than UTF-8.
1340 * This function is automatically called by the case folding functions.
1341 *
1342 * @param string Charset for which to initialize case folding.
1343 * @return integer Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached).
1344 * @access private
1345 */
1347 // Only process if the case table is not yet loaded:
1350 }
1352 // Use cached version if possible
1357 }
1359 // init UTF-8 conversion for this charset
1362 }
1364 // UTF-8 case folding is used as the base conversion table
1367 }
1371 // reconvert to charset (don't use chr() of numeric value, might be muli-byte)
1374 // $cc = $this->conv($this->caseFolding['utf-8']['toUpper'][$utf8], 'utf-8', $charset);
1378 }
1380 // $cc = $this->conv($this->caseFolding['utf-8']['toLower'][$utf8], 'utf-8', $charset);
1384 }
1386 // $cc = $this->conv($this->caseFolding['utf-8']['toTitle'][$utf8], 'utf-8', $charset);
1390 }
1391 }
1393 // add the ASCII case table
1396 }
1399 }
1403 }
1406 }
1408 /**
1409 * This function initializes the to-ASCII conversion table for a charset other than UTF-8.
1410 * This function is automatically called by the ASCII transliteration functions.
1411 *
1412 * @param string Charset for which to initialize conversion.
1413 * @return integer Returns FALSE on error, a TRUE value on success: 1 table already loaded, 2, cached version, 3 table parsed (and cached).
1414 * @access private
1415 */
1417 // Only process if the case table is not yet loaded:
1420 }
1422 // Use cached version if possible
1427 }
1429 // init UTF-8 conversion for this charset
1432 }
1434 // UTF-8/ASCII transliteration is used as the base conversion table
1437 }
1441 // reconvert to charset (don't use chr() of numeric value, might be muli-byte)
1446 }
1447 }
1451 }
1454 }
1457 /********************************************
1458 *
1459 * String operation functions
1460 *
1461 ********************************************/
1463 /**
1464 * Returns a part of a string.
1465 * Unit-tested by Kasper (single byte charsets only)
1466 *
1467 * @param string The character set
1468 * @param string Character string
1469 * @param integer Start position (character position)
1470 * @param integer Length (in characters)
1471 * @return string The substring
1472 * @see substr(), mb_substr()
1473 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1474 */
1478 }
1481 // cannot omit $len, when specifying charset
1489 }
1492 }
1494 // cannot omit $len, when specifying charset
1502 }
1505 }
1514 }
1516 // treat everything else as single-byte encoding
1518 }
1520 /**
1521 * Counts the number of characters.
1522 * Unit-tested by Kasper (single byte charsets only)
1523 *
1524 * @param string The character set
1525 * @param string Character string
1526 * @return integer The number of characters
1527 * @see strlen()
1528 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1529 */
1543 }
1544 // treat everything else as single-byte encoding
1546 }
1548 /**
1549 * Method to crop strings using the mb_substr function.
1550 *
1551 * @param string The character set
1552 * @param string String to be cropped
1553 * @param integer Crop length (in characters)
1554 * @param string Crop signifier
1555 * @return string The shortened string
1556 * @see mb_strlen(), mb_substr()
1557 */
1561 }
1567 }
1570 }
1572 /**
1573 * Truncates a string and pre-/appends a string.
1574 * Unit tested by Kasper
1575 *
1576 * @param string The character set
1577 * @param string Character string
1578 * @param integer Length (in characters)
1579 * @param string Crop signifier
1580 * @return string The shortened string
1581 * @see substr(), mb_strimwidth()
1582 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1583 */
1587 }
1591 }
1604 }
1605 }
1606 }
1615 }
1619 }
1620 }
1622 /*
1623 if (abs($len)<$this->strlen($charset,$string)) { // Has to use ->strlen() - otherwise multibyte strings ending with a multibyte char will return TRUE here (which is not a catastrophe, but...)
1624 if ($len > 0) {
1625 return substr($string,0,$i).$crop;
1626 } else {
1627 return $crop.substr($string,$i);
1628 }
1629 }
1630 */
1631 }
1633 }
1635 /**
1636 * Cuts a string short at a given byte length.
1637 *
1638 * @param string The character set
1639 * @param string Character string
1640 * @param integer The byte length
1641 * @return string The shortened string
1642 * @see mb_strcut()
1643 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1644 */
1648 }
1663 }
1664 // treat everything else as single-byte encoding
1666 }
1668 /**
1669 * Translates all characters of a string into their respective case values.
1670 * Unlike strtolower() and strtoupper() this method is locale independent.
1671 * Note that the string length may change!
1672 * eg. lower case German "ß" (sharp S) becomes upper case "SS"
1673 * Unit-tested by Kasper
1674 * Real case folding is language dependent, this method ignores this fact.
1675 *
1676 * @param string Character set of string
1677 * @param string Input string to convert case for
1678 * @param string Case keyword: "toLower" means lowercase conversion, anything else is uppercase (use "toUpper" )
1679 * @return string The converted string
1680 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1681 * @see strtolower(), strtoupper()
1682 */
1689 }
1695 // treat everything else as single-byte encoding
1697 }
1700 }
1702 /**
1703 * Equivalent of lcfirst/ucfirst but using character set.
1704 *
1705 * @param string $charset
1706 * @param string $string
1707 * @param string $case
1708 * @return string
1709 * @see t3lib_cs::conv_case()
1710 */
1716 }
1718 /**
1719 * Converts special chars (like æøåÆØÅ, umlauts etc) to ascii equivalents (usually double-bytes, like æ => ae etc.)
1720 *
1721 * @param string $charset Character set of string
1722 * @param string $string Input string to convert
1723 * @return string The converted string
1724 */
1731 // treat everything else as single-byte encoding
1733 }
1736 }
1739 /**
1740 * converts the language codes that we get from the client (usually HTTP_ACCEPT_LANGUAGE)
1741 * into a TYPO3-readable language code
1742 * @param $languageCodesList list of language codes. something like 'de,en-us;q=0.9,de-de;q=0.7,es-cl;q=0.6,en;q=0.4,es;q=0.3,zh;q=0.1'
1743 * see http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.4
1744 * @return string a preferred language that TYPO3 supports, or "default" if none found
1745 * @author Benjamin Mack (benni.typo3.org)
1746 */
1751 // get all languages where TYPO3 code is the same as the ISO code
1754 }
1756 // get all languages where TYPO3 code differs from ISO code
1757 // or needs the country part
1758 // the iso codes will here overwrite the default typo3 language in the key
1762 }
1764 // move the iso codes to the (because we're comparing the keys with "isset" later on)
1769 // order the preferred languages after they key
1775 }
1777 }
1779 // loop through the languages, with the highest priority first
1785 }
1787 // strip the country code from the end
1792 }
1793 }
1796 }
1798 }
1801 /********************************************
1802 *
1803 * Internal string operation functions
1804 *
1805 ********************************************/
1807 /**
1808 * Maps all characters of a string in a single byte charset.
1809 *
1810 * @param string the string
1811 * @param string the charset
1812 * @param string mode: 'case' (case folding) or 'ascii' (ASCII transliteration)
1813 * @param string 'case': conversion 'toLower' or 'toUpper'
1814 * @return string the converted string
1815 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1816 */
1835 }
1844 }
1845 }
1848 }
1851 /********************************************
1852 *
1853 * Internal UTF-8 string operation functions
1854 *
1855 ********************************************/
1857 /**
1858 * Returns a part of a UTF-8 string.
1859 * Unit-tested by Kasper and works 100% like substr() / mb_substr() for full range of $start/$len
1860 *
1861 * @param string UTF-8 string
1862 * @param integer Start position (character position)
1863 * @param integer Length (in characters)
1864 * @return string The substring
1865 * @see substr()
1866 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1867 */
1871 }
1879 }
1880 }
1887 {
1890 else
1891 {
1893 }
1894 }
1897 }
1898 }
1900 /**
1901 * Counts the number of characters of a string in UTF-8.
1902 * Unit-tested by Kasper and works 100% like strlen() / mb_strlen()
1903 *
1904 * @param string UTF-8 multibyte character string
1905 * @return integer The number of characters
1906 * @see strlen()
1907 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1908 */
1914 {
1916 }
1918 {
1920 }
1921 }
1923 }
1925 /**
1926 * Truncates a string in UTF-8 short at a given byte length.
1927 *
1928 * @param string UTF-8 multibyte character string
1929 * @param integer the byte length
1930 * @return string the shortened string
1931 * @see mb_strcut()
1932 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1933 */
1938 // find the first byte
1939 ;
1940 }
1945 // calculate number of bytes
1947 }
1950 }
1951 // fallthru: multibyte char fits into length
1952 }
1954 }
1956 /**
1957 * Find position of first occurrence of a string, both arguments are in UTF-8.
1958 *
1959 * @param string UTF-8 string to search in
1960 * @param string UTF-8 string to search for
1961 * @param integer Positition to start the search
1962 * @return integer The character position
1963 * @see strpos()
1964 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1965 */
1971 }
1984 }
1986 /**
1987 * Find position of last occurrence of a char in a string, both arguments are in UTF-8.
1988 *
1989 * @param string UTF-8 string to search in
1990 * @param string UTF-8 character to search for (single character)
1991 * @return integer The character position
1992 * @see strrpos()
1993 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
1994 */
2000 }
2008 }
2010 /**
2011 * Translates a character position into an 'absolute' byte position.
2012 * Unit tested by Kasper.
2013 *
2014 * @param string UTF-8 string
2015 * @param integer Character position (negative values start from the end)
2016 * @return integer Byte position
2017 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
2018 */
2029 }
2034 {
2036 }
2038 {
2040 }
2041 }
2047 // skip trailing multi-byte data bytes
2050 }
2052 // correct offset
2054 }
2057 }
2059 /**
2060 * Translates an 'absolute' byte position into a character position.
2061 * Unit tested by Kasper.
2062 *
2063 * @param string UTF-8 string
2064 * @param integer byte position
2065 * @return integer character position
2066 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
2067 */
2073 {
2075 }
2077 {
2079 }
2080 }
2086 }
2088 /**
2089 * Maps all characters of an UTF-8 string.
2090 *
2091 * @param string UTF-8 string
2092 * @param string mode: 'case' (case folding) or 'ascii' (ASCII transliteration)
2093 * @param string 'case': conversion 'toLower' or 'toUpper'
2094 * @return string the converted string
2095 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
2096 */
2114 }
2119 {
2121 }
2128 }
2134 }
2135 }
2138 }
2141 /********************************************
2142 *
2143 * Internal EUC string operation functions
2144 *
2145 * Extended Unix Code:
2146 * ASCII compatible 7bit single bytes chars
2147 * 8bit two byte chars
2148 *
2149 * Shift-JIS is treated as a special case.
2150 *
2151 ********************************************/
2153 /**
2154 * Cuts a string in the EUC charset family short at a given byte length.
2155 *
2156 * @param string EUC multibyte character string
2157 * @param integer the byte length
2158 * @param string the charset
2159 * @return string the shortened string
2160 * @see mb_strcut()
2161 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
2162 */
2171 }
2176 }
2177 }
2186 }
2187 }
2189 /**
2190 * Returns a part of a string in the EUC charset family.
2191 *
2192 * @param string EUC multibyte character string
2193 * @param integer start position (character position)
2194 * @param string the charset
2195 * @param integer length (in characters)
2196 * @return string the substring
2197 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
2198 */
2210 {
2212 }
2213 else
2214 {
2216 }
2217 }
2220 }
2221 }
2223 /**
2224 * Counts the number of characters of a string in the EUC charset family.
2225 *
2226 * @param string EUC multibyte character string
2227 * @param string the charset
2228 * @return integer the number of characters
2229 * @see strlen()
2230 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
2231 */
2241 }
2246 }
2249 }
2252 }
2254 /**
2255 * Translates a character position into an 'absolute' byte position.
2256 *
2257 * @param string EUC multibyte character string
2258 * @param integer character position (negative values start from the end)
2259 * @param string the charset
2260 * @return integer byte position
2261 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
2262 */
2274 }
2282 }
2287 }
2290 }
2300 }
2302 /**
2303 * Maps all characters of a string in the EUC charset family.
2304 *
2305 * @param string EUC multibyte character string
2306 * @param string the charset
2307 * @param string mode: 'case' (case folding) or 'ascii' (ASCII transliteration)
2308 * @param string 'case': conversion 'toLower' or 'toUpper'
2309 * @return string the converted string
2310 * @author Martin Kutschker <martin.t.kutschker@blackbox.net>
2311 */
2330 }
2342 }
2343 }
2348 }
2349 }
2355 }
2356 }
2359 }
2361 }
2363 if (defined('TYPO3_MODE') && isset($GLOBALS['TYPO3_CONF_VARS'][TYPO3_MODE]['XCLASS']['t3lib/class.t3lib_cs.php'])) {
2365 }
2367 ?>