includes/normal/UtfNormalUtil.php

   1 <?php
   2 /**
   3  * Some of these functions are adapted from places in MediaWiki.
   4  * Should probably merge them for consistency.
   5  *
   6  * Copyright © 2004 Brion Vibber <brion@pobox.com>
   7  * https://www.mediawiki.org/
   8  *
   9  * This program is 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  * This program is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  17  * GNU General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU General Public License along
  20  * with this program; if not, write to the Free Software Foundation, Inc.,
  21  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  22  * http://www.gnu.org/copyleft/gpl.html
  23  *
  24  * @file
  25  * @ingroup UtfNormal
  26  */
  27
  28 /**
  29  * Return UTF-8 sequence for a given Unicode code point.
  30  * May die if fed out of range data.
  31  *
  32  * @param $codepoint Integer:
  33  * @return String
  34  * @public
  35  */
  36 function codepointToUtf8( $codepoint ) {
  37         if ( $codepoint < 0x80 ) {
  38                 return chr( $codepoint );
  39         }
  40
  41         if ( $codepoint < 0x800 ) {
  42                 return chr( $codepoint >> 6 & 0x3f | 0xc0 ) .
  43                         chr( $codepoint & 0x3f | 0x80 );
  44         }
  45
  46         if ( $codepoint < 0x10000 ) {
  47                 return chr( $codepoint >> 12 & 0x0f | 0xe0 ) .
  48                         chr( $codepoint >> 6 & 0x3f | 0x80 ) .
  49                         chr( $codepoint & 0x3f | 0x80 );
  50         }
  51
  52         if ( $codepoint < 0x110000 ) {
  53                 return chr( $codepoint >> 18 & 0x07 | 0xf0 ) .
  54                         chr( $codepoint >> 12 & 0x3f | 0x80 ) .
  55                         chr( $codepoint >> 6 & 0x3f | 0x80 ) .
  56                         chr( $codepoint & 0x3f | 0x80 );
  57         }
  58
  59         echo "Asked for code outside of range ($codepoint)\n";
  60         die( -1 );
  61 }
  62
  63 /**
  64  * Take a series of space-separated hexadecimal numbers representing
  65  * Unicode code points and return a UTF-8 string composed of those
  66  * characters. Used by UTF-8 data generation and testing routines.
  67  *
  68  * @param $sequence String
  69  * @return String
  70  * @private
  71  */
  72 function hexSequenceToUtf8( $sequence ) {
  73         $utf = '';
  74         foreach ( explode( ' ', $sequence ) as $hex ) {
  75                 $n = hexdec( $hex );
  76                 $utf .= codepointToUtf8( $n );
  77         }
  78
  79         return $utf;
  80 }
  81
  82 /**
  83  * Take a UTF-8 string and return a space-separated series of hex
  84  * numbers representing Unicode code points. For debugging.
  85  *
  86  * @param string $str UTF-8 string.
  87  * @return string
  88  * @private
  89  */
  90 function utf8ToHexSequence( $str ) {
  91         $buf = '';
  92         foreach ( preg_split( '//u', $str, -1, PREG_SPLIT_NO_EMPTY ) as $cp ) {
  93                 $buf .= sprintf( '%04x ', utf8ToCodepoint( $cp ) );
  94         }
  95
  96         return rtrim( $buf );
  97 }
  98
  99 /**
 100  * Determine the Unicode codepoint of a single-character UTF-8 sequence.
 101  * Does not check for invalid input data.
 102  *
 103  * @param $char String
 104  * @return Integer
 105  * @public
 106  */
 107 function utf8ToCodepoint( $char ) {
 108         # Find the length
 109         $z = ord( $char[0] );
 110         if ( $z & 0x80 ) {
 111                 $length = 0;
 112                 while ( $z & 0x80 ) {
 113                         $length++;
 114                         $z <<= 1;
 115                 }
 116         } else {
 117                 $length = 1;
 118         }
 119
 120         if ( $length != strlen( $char ) ) {
 121                 return false;
 122         }
 123
 124         if ( $length == 1 ) {
 125                 return ord( $char );
 126         }
 127
 128         # Mask off the length-determining bits and shift back to the original location
 129         $z &= 0xff;
 130         $z >>= $length;
 131
 132         # Add in the free bits from subsequent bytes
 133         for ( $i = 1; $i < $length; $i++ ) {
 134                 $z <<= 6;
 135                 $z |= ord( $char[$i] ) & 0x3f;
 136         }
 137
 138         return $z;
 139 }
 140
 141 /**
 142  * Escape a string for inclusion in a PHP single-quoted string literal.
 143  *
 144  * @param string $string string to be escaped.
 145  * @return String: escaped string.
 146  * @public
 147  */
 148 function escapeSingleString( $string ) {
 149         return strtr( $string,
 150                 array(
 151                         '\\' => '\\\\',
 152                         '\'' => '\\\''
 153                 ) );
 154 }