includes/compat/normal/UtfNormal.php

   1 <?php
   2 /**
   3  * Unicode normalization routines
   4  *
   5  * Copyright © 2004 Brion Vibber <brion@pobox.com>
   6  * https://www.mediawiki.org/
   7  *
   8  * This program is free software; you can redistribute it and/or modify
   9  * it under the terms of the GNU General Public License as published by
  10  * the Free Software Foundation; either version 2 of the License, or
  11  * (at your option) any later version.
  12  *
  13  * This program is distributed in the hope that it will be useful,
  14  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  15  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  16  * GNU General Public License for more details.
  17  *
  18  * You should have received a copy of the GNU General Public License along
  19  * with this program; if not, write to the Free Software Foundation, Inc.,
  20  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  21  * http://www.gnu.org/copyleft/gpl.html
  22  *
  23  * @file
  24  * @ingroup UtfNormal
  25  */
  26
  27 /**
  28  * @defgroup UtfNormal UtfNormal
  29  */
  30
  31 use UtfNormal\Validator;
  32
  33 /**
  34  * Unicode normalization routines for working with UTF-8 strings.
  35  * Currently assumes that input strings are valid UTF-8!
  36  *
  37  * Not as fast as I'd like, but should be usable for most purposes.
  38  * UtfNormal::toNFC() will bail early if given ASCII text or text
  39  * it can quickly determine is already normalized.
  40  *
  41  * All functions can be called static.
  42  *
  43  * See description of forms at http://www.unicode.org/reports/tr15/
  44  *
  45  * @deprecated since 1.25, use UtfNormal\Validator directly
  46  * @ingroup UtfNormal
  47  */
  48 class UtfNormal {
  49         /**
  50          * The ultimate convenience function! Clean up invalid UTF-8 sequences,
  51          * and convert to normal form C, canonical composition.
  52          *
  53          * Fast return for pure ASCII strings; some lesser optimizations for
  54          * strings containing only known-good characters. Not as fast as toNFC().
  55          *
  56          * @param string $string a UTF-8 string
  57          * @return string a clean, shiny, normalized UTF-8 string
  58          */
  59         static function cleanUp( $string ) {
  60                 return Validator::cleanUp( $string );
  61         }
  62
  63         /**
  64          * Convert a UTF-8 string to normal form C, canonical composition.
  65          * Fast return for pure ASCII strings; some lesser optimizations for
  66          * strings containing only known-good characters.
  67          *
  68          * @param string $string a valid UTF-8 string. Input is not validated.
  69          * @return string a UTF-8 string in normal form C
  70          */
  71         static function toNFC( $string ) {
  72                 return Validator::toNFC( $string );
  73         }
  74
  75         /**
  76          * Convert a UTF-8 string to normal form D, canonical decomposition.
  77          * Fast return for pure ASCII strings.
  78          *
  79          * @param string $string a valid UTF-8 string. Input is not validated.
  80          * @return string a UTF-8 string in normal form D
  81          */
  82         static function toNFD( $string ) {
  83                 return Validator::toNFD( $string );
  84         }
  85
  86         /**
  87          * Convert a UTF-8 string to normal form KC, compatibility composition.
  88          * This may cause irreversible information loss, use judiciously.
  89          * Fast return for pure ASCII strings.
  90          *
  91          * @param string $string a valid UTF-8 string. Input is not validated.
  92          * @return string a UTF-8 string in normal form KC
  93          */
  94         static function toNFKC( $string ) {
  95                 return Validator::toNFKC( $string );
  96         }
  97
  98         /**
  99          * Convert a UTF-8 string to normal form KD, compatibility decomposition.
 100          * This may cause irreversible information loss, use judiciously.
 101          * Fast return for pure ASCII strings.
 102          *
 103          * @param string $string a valid UTF-8 string. Input is not validated.
 104          * @return string a UTF-8 string in normal form KD
 105          */
 106         static function toNFKD( $string ) {
 107                 return Validator::toNFKD( $string );
 108         }
 109
 110         /**
 111          * Returns true if the string is _definitely_ in NFC.
 112          * Returns false if not or uncertain.
 113          * @param string $string a valid UTF-8 string. Input is not validated.
 114          * @return bool
 115          */
 116         static function quickIsNFC( $string ) {
 117                 return Validator::quickIsNFC( $string );
 118         }
 119
 120         /**
 121          * Returns true if the string is _definitely_ in NFC.
 122          * Returns false if not or uncertain.
 123          * @param string $string a UTF-8 string, altered on output to be valid UTF-8 safe for XML.
 124          * @return bool
 125          */
 126         static function quickIsNFCVerify( &$string ) {
 127                 return Validator::quickIsNFCVerify( $string );
 128         }
 129 }