ILIAS  release_5-2 Revision v5.2.25-18-g3f80b82851
UtfNormalUtil.php
Go to the documentation of this file.
1 <?php
2 # Copyright (C) 2004 Brion Vibber <brion@pobox.com>
3 # http://www.mediawiki.org/
4 #
5 # This program is free software; you can redistribute it and/or modify
6 # it under the terms of the GNU General Public License as published by
7 # the Free Software Foundation; either version 2 of the License, or
8 # (at your option) any later version.
9 #
10 # This program is distributed in the hope that it will be useful,
11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 # GNU General Public License for more details.
14 #
15 # You should have received a copy of the GNU General Public License along
16 # with this program; if not, write to the Free Software Foundation, Inc.,
17 # 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 # http://www.gnu.org/copyleft/gpl.html
19 
38 function codepointToUtf8( $codepoint ) {
39  if($codepoint < 0x80) return chr($codepoint);
40  if($codepoint < 0x800) return chr($codepoint >> 6 & 0x3f | 0xc0) .
41  chr($codepoint & 0x3f | 0x80);
42  if($codepoint < 0x10000) return chr($codepoint >> 12 & 0x0f | 0xe0) .
43  chr($codepoint >> 6 & 0x3f | 0x80) .
44  chr($codepoint & 0x3f | 0x80);
45  if($codepoint < 0x110000) return chr($codepoint >> 18 & 0x07 | 0xf0) .
46  chr($codepoint >> 12 & 0x3f | 0x80) .
47  chr($codepoint >> 6 & 0x3f | 0x80) .
48  chr($codepoint & 0x3f | 0x80);
49 
50  echo "Asked for code outside of range ($codepoint)\n";
51  die( -1 );
52 }
53 
63 function hexSequenceToUtf8( $sequence ) {
64  $utf = '';
65  foreach( explode( ' ', $sequence ) as $hex ) {
66  $n = hexdec( $hex );
67  $utf .= codepointToUtf8( $n );
68  }
69  return $utf;
70 }
71 
80 function utf8ToHexSequence( $str ) {
81  return rtrim( preg_replace_callback(
82  '/(.)/uS',
83  function($hit) {
84  return sprintf("%04x ", utf8ToCodepoint($hit[1]));
85  },
86  $str
87  ));
88 }
89 
98 function utf8ToCodepoint( $char ) {
99  # Find the length
100  $z = ord( $char{0} );
101  if ( $z & 0x80 ) {
102  $length = 0;
103  while ( $z & 0x80 ) {
104  $length++;
105  $z <<= 1;
106  }
107  } else {
108  $length = 1;
109  }
110 
111  if ( $length != strlen( $char ) ) {
112  return false;
113  }
114  if ( $length == 1 ) {
115  return ord( $char );
116  }
117 
118  # Mask off the length-determining bits and shift back to the original location
119  $z &= 0xff;
120  $z >>= $length;
121 
122  # Add in the free bits from subsequent bytes
123  for ( $i=1; $i<$length; $i++ ) {
124  $z <<= 6;
125  $z |= ord( $char{$i} ) & 0x3f;
126  }
127 
128  return $z;
129 }
130 
138 function escapeSingleString( $string ) {
139  return strtr( $string,
140  array(
141  '\\' => '\\\\',
142  '\'' => '\\\''
143  ));
144 }
145 
146 ?>
codepointToUtf8( $codepoint)
Return UTF-8 sequence for a given Unicode code point.
hexSequenceToUtf8( $sequence)
Take a series of space-separated hexadecimal numbers representing Unicode code points and return a UT...
utf8ToCodepoint( $char)
Determine the Unicode codepoint of a single-character UTF-8 sequence.
utf8ToHexSequence( $str)
Take a UTF-8 string and return a space-separated series of hex numbers representing Unicode code poin...
$n
Definition: RandomTest.php:80
escapeSingleString( $string)
Escape a string for inclusion in a PHP single-quoted string literal.
Create styles array
The data for the language used.