summaryrefslogtreecommitdiff
path: root/includes/normal/UtfNormal.php
diff options
context:
space:
mode:
Diffstat (limited to 'includes/normal/UtfNormal.php')
-rw-r--r--includes/normal/UtfNormal.php291
1 files changed, 152 insertions, 139 deletions
diff --git a/includes/normal/UtfNormal.php b/includes/normal/UtfNormal.php
index 5a091afc..8204f974 100644
--- a/includes/normal/UtfNormal.php
+++ b/includes/normal/UtfNormal.php
@@ -3,7 +3,7 @@
* Unicode normalization routines
*
* Copyright © 2004 Brion Vibber <brion@pobox.com>
- * http://www.mediawiki.org/
+ * https://www.mediawiki.org/
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -50,21 +50,20 @@ class UtfNormal {
* For using the ICU wrapper
*/
const UNORM_NONE = 1;
- const UNORM_NFD = 2;
+ const UNORM_NFD = 2;
const UNORM_NFKD = 3;
- const UNORM_NFC = 4;
+ const UNORM_NFC = 4;
const UNORM_NFKC = 5;
- const UNORM_FCD = 6;
+ const UNORM_FCD = 6;
const UNORM_DEFAULT = self::UNORM_NFC;
- static $utfCombiningClass = null;
- static $utfCanonicalComp = null;
- static $utfCanonicalDecomp = null;
+ public static $utfCombiningClass = null;
+ public static $utfCanonicalComp = null;
+ public static $utfCanonicalDecomp = null;
# Load compatibility decompositions on demand if they are needed.
- static $utfCompatibilityDecomp = null;
-
- static $utfCheckNFC;
+ public static $utfCompatibilityDecomp = null;
+ public static $utfCheckNFC;
/**
* The ultimate convenience function! Clean up invalid UTF-8 sequences,
@@ -77,21 +76,21 @@ class UtfNormal {
* @return string a clean, shiny, normalized UTF-8 string
*/
static function cleanUp( $string ) {
- if( NORMALIZE_ICU ) {
+ if ( NORMALIZE_ICU ) {
$string = self::replaceForNativeNormalize( $string );
# UnicodeString constructor fails if the string ends with a
# head byte. Add a junk char at the end, we'll strip it off.
return rtrim( utf8_normalize( $string . "\x01", self::UNORM_NFC ), "\x01" );
- } elseif( NORMALIZE_INTL ) {
+ } elseif ( NORMALIZE_INTL ) {
$string = self::replaceForNativeNormalize( $string );
$norm = normalizer_normalize( $string, Normalizer::FORM_C );
- if( $norm === null || $norm === false ) {
+ if ( $norm === null || $norm === false ) {
# normalizer_normalize will either return false or null
# (depending on which doc you read) if invalid utf8 string.
# quickIsNFCVerify cleans up invalid sequences.
- if( UtfNormal::quickIsNFCVerify( $string ) ) {
+ if ( UtfNormal::quickIsNFCVerify( $string ) ) {
# if that's true, the string is actually already normal.
return $string;
} else {
@@ -101,7 +100,7 @@ class UtfNormal {
} else {
return $norm;
}
- } elseif( UtfNormal::quickIsNFCVerify( $string ) ) {
+ } elseif ( UtfNormal::quickIsNFCVerify( $string ) ) {
# Side effect -- $string has had UTF-8 errors cleaned up.
return $string;
} else {
@@ -118,11 +117,11 @@ class UtfNormal {
* @return string a UTF-8 string in normal form C
*/
static function toNFC( $string ) {
- if( NORMALIZE_INTL )
+ if ( NORMALIZE_INTL )
return normalizer_normalize( $string, Normalizer::FORM_C );
- elseif( NORMALIZE_ICU )
+ elseif ( NORMALIZE_ICU )
return utf8_normalize( $string, self::UNORM_NFC );
- elseif( UtfNormal::quickIsNFC( $string ) )
+ elseif ( UtfNormal::quickIsNFC( $string ) )
return $string;
else
return UtfNormal::NFC( $string );
@@ -136,11 +135,11 @@ class UtfNormal {
* @return string a UTF-8 string in normal form D
*/
static function toNFD( $string ) {
- if( NORMALIZE_INTL )
+ if ( NORMALIZE_INTL )
return normalizer_normalize( $string, Normalizer::FORM_D );
- elseif( NORMALIZE_ICU )
+ elseif ( NORMALIZE_ICU )
return utf8_normalize( $string, self::UNORM_NFD );
- elseif( preg_match( '/[\x80-\xff]/', $string ) )
+ elseif ( preg_match( '/[\x80-\xff]/', $string ) )
return UtfNormal::NFD( $string );
else
return $string;
@@ -155,11 +154,11 @@ class UtfNormal {
* @return string a UTF-8 string in normal form KC
*/
static function toNFKC( $string ) {
- if( NORMALIZE_INTL )
+ if ( NORMALIZE_INTL )
return normalizer_normalize( $string, Normalizer::FORM_KC );
- elseif( NORMALIZE_ICU )
+ elseif ( NORMALIZE_ICU )
return utf8_normalize( $string, self::UNORM_NFKC );
- elseif( preg_match( '/[\x80-\xff]/', $string ) )
+ elseif ( preg_match( '/[\x80-\xff]/', $string ) )
return UtfNormal::NFKC( $string );
else
return $string;
@@ -174,11 +173,11 @@ class UtfNormal {
* @return string a UTF-8 string in normal form KD
*/
static function toNFKD( $string ) {
- if( NORMALIZE_INTL )
+ if ( NORMALIZE_INTL )
return normalizer_normalize( $string, Normalizer::FORM_KD );
- elseif( NORMALIZE_ICU )
+ elseif ( NORMALIZE_ICU )
return utf8_normalize( $string, self::UNORM_NFKD );
- elseif( preg_match( '/[\x80-\xff]/', $string ) )
+ elseif ( preg_match( '/[\x80-\xff]/', $string ) )
return UtfNormal::NFKD( $string );
else
return $string;
@@ -189,7 +188,7 @@ class UtfNormal {
* @private
*/
static function loadData() {
- if( !isset( self::$utfCombiningClass ) ) {
+ if ( !isset( self::$utfCombiningClass ) ) {
require_once __DIR__ . '/UtfNormalData.inc';
}
}
@@ -203,34 +202,35 @@ class UtfNormal {
static function quickIsNFC( $string ) {
# ASCII is always valid NFC!
# If it's pure ASCII, let it through.
- if( !preg_match( '/[\x80-\xff]/', $string ) ) return true;
+ if ( !preg_match( '/[\x80-\xff]/', $string ) ) return true;
UtfNormal::loadData();
$len = strlen( $string );
- for( $i = 0; $i < $len; $i++ ) {
+ for ( $i = 0; $i < $len; $i++ ) {
$c = $string[$i];
$n = ord( $c );
- if( $n < 0x80 ) {
+ if ( $n < 0x80 ) {
continue;
- } elseif( $n >= 0xf0 ) {
+ } elseif ( $n >= 0xf0 ) {
$c = substr( $string, $i, 4 );
$i += 3;
- } elseif( $n >= 0xe0 ) {
+ } elseif ( $n >= 0xe0 ) {
$c = substr( $string, $i, 3 );
$i += 2;
- } elseif( $n >= 0xc0 ) {
+ } elseif ( $n >= 0xc0 ) {
$c = substr( $string, $i, 2 );
$i++;
}
- if( isset( self::$utfCheckNFC[$c] ) ) {
+ if ( isset( self::$utfCheckNFC[$c] ) ) {
# If it's NO or MAYBE, bail and do the slow check.
return false;
}
- if( isset( self::$utfCombiningClass[$c] ) ) {
+ if ( isset( self::$utfCombiningClass[$c] ) ) {
# Combining character? We might have to do sorting, at least.
return false;
}
}
+
return true;
}
@@ -247,10 +247,10 @@ class UtfNormal {
# ASCII is always valid NFC!
# If we're only ever given plain ASCII, we can avoid the overhead
# of initializing the decomposition tables by skipping out early.
- if( !preg_match( '/[\x80-\xff]/', $string ) ) return true;
+ if ( !preg_match( '/[\x80-\xff]/', $string ) ) return true;
static $checkit = null, $tailBytes = null, $utfCheckOrCombining = null;
- if( !isset( $checkit ) ) {
+ if ( !isset( $checkit ) ) {
# Load/build some scary lookup tables...
UtfNormal::loadData();
@@ -258,30 +258,30 @@ class UtfNormal {
# Head bytes for sequences which we should do further validity checks
$checkit = array_flip( array_map( 'chr',
- array( 0xc0, 0xc1, 0xe0, 0xed, 0xef,
- 0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7,
- 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff ) ) );
+ array( 0xc0, 0xc1, 0xe0, 0xed, 0xef,
+ 0xf0, 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7,
+ 0xf8, 0xf9, 0xfa, 0xfb, 0xfc, 0xfd, 0xfe, 0xff ) ) );
# Each UTF-8 head byte is followed by a certain
# number of tail bytes.
$tailBytes = array();
- for( $n = 0; $n < 256; $n++ ) {
- if( $n < 0xc0 ) {
+ for ( $n = 0; $n < 256; $n++ ) {
+ if ( $n < 0xc0 ) {
$remaining = 0;
- } elseif( $n < 0xe0 ) {
+ } elseif ( $n < 0xe0 ) {
$remaining = 1;
- } elseif( $n < 0xf0 ) {
+ } elseif ( $n < 0xf0 ) {
$remaining = 2;
- } elseif( $n < 0xf8 ) {
+ } elseif ( $n < 0xf8 ) {
$remaining = 3;
- } elseif( $n < 0xfc ) {
+ } elseif ( $n < 0xfc ) {
$remaining = 4;
- } elseif( $n < 0xfe ) {
+ } elseif ( $n < 0xfe ) {
$remaining = 5;
} else {
$remaining = 0;
}
- $tailBytes[chr($n)] = $remaining;
+ $tailBytes[chr( $n )] = $remaining;
}
}
@@ -297,10 +297,10 @@ class UtfNormal {
$looksNormal = true;
$base = 0;
$replace = array();
- foreach( $matches[1] as $str ) {
+ foreach ( $matches[1] as $str ) {
$chunk = strlen( $str );
- if( $str[0] < "\x80" ) {
+ if ( $str[0] < "\x80" ) {
# ASCII chunk: guaranteed to be valid UTF-8
# and in normal form C, so skip over it.
$base += $chunk;
@@ -317,30 +317,30 @@ class UtfNormal {
$head = '';
$len = $chunk + 1; # Counting down is faster. I'm *so* sorry.
- for( $i = -1; --$len; ) {
+ for ( $i = -1; --$len; ) {
$remaining = $tailBytes[$c = $str[++$i]];
- if( $remaining ) {
+ if ( $remaining ) {
# UTF-8 head byte!
$sequence = $head = $c;
do {
# Look for the defined number of tail bytes...
- if( --$len && ( $c = $str[++$i] ) >= "\x80" && $c < "\xc0" ) {
+ if ( --$len && ( $c = $str[++$i] ) >= "\x80" && $c < "\xc0" ) {
# Legal tail bytes are nice.
$sequence .= $c;
} else {
- if( 0 == $len ) {
+ if ( 0 == $len ) {
# Premature end of string!
# Drop a replacement character into output to
# represent the invalid UTF-8 sequence.
$replace[] = array( UTF8_REPLACEMENT,
- $base + $i + 1 - strlen( $sequence ),
- strlen( $sequence ) );
+ $base + $i + 1 - strlen( $sequence ),
+ strlen( $sequence ) );
break 2;
} else {
# Illegal tail byte; abandon the sequence.
$replace[] = array( UTF8_REPLACEMENT,
- $base + $i - strlen( $sequence ),
- strlen( $sequence ) );
+ $base + $i - strlen( $sequence ),
+ strlen( $sequence ) );
# Back up and reprocess this byte; it may itself
# be a legal ASCII or UTF-8 sequence head.
--$i;
@@ -348,59 +348,60 @@ class UtfNormal {
continue 2;
}
}
- } while( --$remaining );
+ } while ( --$remaining );
- if( isset( $checkit[$head] ) ) {
+ if ( isset( $checkit[$head] ) ) {
# Do some more detailed validity checks, for
# invalid characters and illegal sequences.
- if( $head == "\xed" ) {
+ if ( $head == "\xed" ) {
# 0xed is relatively frequent in Korean, which
# abuts the surrogate area, so we're doing
# this check separately to speed things up.
- if( $sequence >= UTF8_SURROGATE_FIRST ) {
+ if ( $sequence >= UTF8_SURROGATE_FIRST ) {
# Surrogates are legal only in UTF-16 code.
# They are totally forbidden here in UTF-8
# utopia.
$replace[] = array( UTF8_REPLACEMENT,
- $base + $i + 1 - strlen( $sequence ),
- strlen( $sequence ) );
+ $base + $i + 1 - strlen( $sequence ),
+ strlen( $sequence ) );
$head = '';
continue;
}
} else {
# Slower, but rarer checks...
$n = ord( $head );
- if(
+ if (
# "Overlong sequences" are those that are syntactically
# correct but use more UTF-8 bytes than are necessary to
# encode a character. Naïve string comparisons can be
# tricked into failing to see a match for an ASCII
# character, for instance, which can be a security hole
# if blacklist checks are being used.
- ($n < 0xc2 && $sequence <= UTF8_OVERLONG_A)
- || ($n == 0xe0 && $sequence <= UTF8_OVERLONG_B)
- || ($n == 0xf0 && $sequence <= UTF8_OVERLONG_C)
+ ( $n < 0xc2 && $sequence <= UTF8_OVERLONG_A )
+ || ( $n == 0xe0 && $sequence <= UTF8_OVERLONG_B )
+ || ( $n == 0xf0 && $sequence <= UTF8_OVERLONG_C )
# U+FFFE and U+FFFF are explicitly forbidden in Unicode.
- || ($n == 0xef &&
- ($sequence == UTF8_FFFE)
- || ($sequence == UTF8_FFFF) )
+ || ( $n == 0xef &&
+ ( $sequence == UTF8_FFFE )
+ || ( $sequence == UTF8_FFFF ) )
# Unicode has been limited to 21 bits; longer
# sequences are not allowed.
- || ($n >= 0xf0 && $sequence > UTF8_MAX) ) {
+ || ( $n >= 0xf0 && $sequence > UTF8_MAX )
+ ) {
$replace[] = array( UTF8_REPLACEMENT,
- $base + $i + 1 - strlen( $sequence ),
- strlen( $sequence ) );
+ $base + $i + 1 - strlen( $sequence ),
+ strlen( $sequence ) );
$head = '';
continue;
}
}
}
- if( isset( $utfCheckOrCombining[$sequence] ) ) {
+ if ( isset( $utfCheckOrCombining[$sequence] ) ) {
# If it's NO or MAYBE, we'll have to rip
# the string apart and put it back together.
# That's going to be mighty slow.
@@ -409,12 +410,12 @@ class UtfNormal {
# The sequence is legal!
$head = '';
- } elseif( $c < "\x80" ) {
+ } elseif ( $c < "\x80" ) {
# ASCII byte.
$head = '';
- } elseif( $c < "\xc0" ) {
+ } elseif ( $c < "\xc0" ) {
# Illegal tail bytes
- if( $head == '' ) {
+ if ( $head == '' ) {
# Out of the blue!
$replace[] = array( UTF8_REPLACEMENT, $base + $i, 1 );
} else {
@@ -431,23 +432,24 @@ class UtfNormal {
}
$base += $chunk;
}
- if( count( $replace ) ) {
+ if ( count( $replace ) ) {
# There were illegal UTF-8 sequences we need to fix up.
$out = '';
$last = 0;
- foreach( $replace as $rep ) {
+ foreach ( $replace as $rep ) {
list( $replacement, $start, $length ) = $rep;
- if( $last < $start ) {
+ if ( $last < $start ) {
$out .= substr( $string, $last, $start - $last );
}
$out .= $replacement;
$last = $start + $length;
}
- if( $last < strlen( $string ) ) {
+ if ( $last < strlen( $string ) ) {
$out .= substr( $string, $last );
}
$string = $out;
}
+
return $looksNormal;
}
@@ -490,14 +492,14 @@ class UtfNormal {
* @private
*/
static function NFKD( $string ) {
- if( !isset( self::$utfCompatibilityDecomp ) ) {
+ if ( !isset( self::$utfCompatibilityDecomp ) ) {
require_once 'UtfNormalDataK.inc';
}
+
return self::fastCombiningSort(
self::fastDecompose( $string, self::$utfCompatibilityDecomp ) );
}
-
/**
* Perform decomposition of a UTF-8 string into either D or KD form
* (depending on which decomposition map is passed to us).
@@ -511,45 +513,45 @@ class UtfNormal {
UtfNormal::loadData();
$len = strlen( $string );
$out = '';
- for( $i = 0; $i < $len; $i++ ) {
+ for ( $i = 0; $i < $len; $i++ ) {
$c = $string[$i];
$n = ord( $c );
- if( $n < 0x80 ) {
+ if ( $n < 0x80 ) {
# ASCII chars never decompose
# THEY ARE IMMORTAL
$out .= $c;
continue;
- } elseif( $n >= 0xf0 ) {
+ } elseif ( $n >= 0xf0 ) {
$c = substr( $string, $i, 4 );
$i += 3;
- } elseif( $n >= 0xe0 ) {
+ } elseif ( $n >= 0xe0 ) {
$c = substr( $string, $i, 3 );
$i += 2;
- } elseif( $n >= 0xc0 ) {
+ } elseif ( $n >= 0xc0 ) {
$c = substr( $string, $i, 2 );
$i++;
}
- if( isset( $map[$c] ) ) {
+ if ( isset( $map[$c] ) ) {
$out .= $map[$c];
continue;
} else {
- if( $c >= UTF8_HANGUL_FIRST && $c <= UTF8_HANGUL_LAST ) {
+ if ( $c >= UTF8_HANGUL_FIRST && $c <= UTF8_HANGUL_LAST ) {
# Decompose a hangul syllable into jamo;
# hardcoded for three-byte UTF-8 sequence.
# A lookup table would be slightly faster,
# but adds a lot of memory & disk needs.
#
- $index = ( (ord( $c[0] ) & 0x0f) << 12
- | (ord( $c[1] ) & 0x3f) << 6
- | (ord( $c[2] ) & 0x3f) )
- - UNICODE_HANGUL_FIRST;
+ $index = ( ( ord( $c[0] ) & 0x0f ) << 12
+ | ( ord( $c[1] ) & 0x3f ) << 6
+ | ( ord( $c[2] ) & 0x3f ) )
+ - UNICODE_HANGUL_FIRST;
$l = intval( $index / UNICODE_HANGUL_NCOUNT );
- $v = intval( ($index % UNICODE_HANGUL_NCOUNT) / UNICODE_HANGUL_TCOUNT);
+ $v = intval( ( $index % UNICODE_HANGUL_NCOUNT ) / UNICODE_HANGUL_TCOUNT );
$t = $index % UNICODE_HANGUL_TCOUNT;
$out .= "\xe1\x84" . chr( 0x80 + $l ) . "\xe1\x85" . chr( 0xa1 + $v );
- if( $t >= 25 ) {
+ if ( $t >= 25 ) {
$out .= "\xe1\x87" . chr( 0x80 + $t - 25 );
- } elseif( $t ) {
+ } elseif ( $t ) {
$out .= "\xe1\x86" . chr( 0xa7 + $t );
}
continue;
@@ -557,6 +559,7 @@ class UtfNormal {
}
$out .= $c;
}
+
return $out;
}
@@ -573,23 +576,23 @@ class UtfNormal {
$out = '';
$combiners = array();
$lastClass = -1;
- for( $i = 0; $i < $len; $i++ ) {
+ for ( $i = 0; $i < $len; $i++ ) {
$c = $string[$i];
$n = ord( $c );
- if( $n >= 0x80 ) {
- if( $n >= 0xf0 ) {
+ if ( $n >= 0x80 ) {
+ if ( $n >= 0xf0 ) {
$c = substr( $string, $i, 4 );
$i += 3;
- } elseif( $n >= 0xe0 ) {
+ } elseif ( $n >= 0xe0 ) {
$c = substr( $string, $i, 3 );
$i += 2;
- } elseif( $n >= 0xc0 ) {
+ } elseif ( $n >= 0xc0 ) {
$c = substr( $string, $i, 2 );
$i++;
}
- if( isset( self::$utfCombiningClass[$c] ) ) {
+ if ( isset( self::$utfCombiningClass[$c] ) ) {
$lastClass = self::$utfCombiningClass[$c];
- if( isset( $combiners[$lastClass] ) ) {
+ if ( isset( $combiners[$lastClass] ) ) {
$combiners[$lastClass] .= $c;
} else {
$combiners[$lastClass] = $c;
@@ -597,7 +600,7 @@ class UtfNormal {
continue;
}
}
- if( $lastClass ) {
+ if ( $lastClass ) {
ksort( $combiners );
$out .= implode( '', $combiners );
$combiners = array();
@@ -605,10 +608,11 @@ class UtfNormal {
$out .= $c;
$lastClass = 0;
}
- if( $lastClass ) {
+ if ( $lastClass ) {
ksort( $combiners );
$out .= implode( '', $combiners );
}
+
return $out;
}
@@ -616,8 +620,10 @@ class UtfNormal {
* Produces canonically composed sequences, i.e. normal form C or KC.
*
* @private
- * @param string $string a valid UTF-8 string in sorted normal form D or KD. Input is not validated.
- * @return string a UTF-8 string with canonical precomposed characters used where possible
+ * @param string $string a valid UTF-8 string in sorted normal form D or KD.
+ * Input is not validated.
+ * @return string a UTF-8 string with canonical precomposed characters used
+ * where possible.
*/
static function fastCompose( $string ) {
UtfNormal::loadData();
@@ -627,12 +633,12 @@ class UtfNormal {
$lastHangul = 0;
$startChar = '';
$combining = '';
- $x1 = ord(substr(UTF8_HANGUL_VBASE, 0, 1));
- $x2 = ord(substr(UTF8_HANGUL_TEND, 0, 1));
- for( $i = 0; $i < $len; $i++ ) {
+ $x1 = ord( substr( UTF8_HANGUL_VBASE, 0, 1 ) );
+ $x2 = ord( substr( UTF8_HANGUL_TEND, 0, 1 ) );
+ for ( $i = 0; $i < $len; $i++ ) {
$c = $string[$i];
$n = ord( $c );
- if( $n < 0x80 ) {
+ if ( $n < 0x80 ) {
# No combining characters here...
$out .= $startChar;
$out .= $combining;
@@ -640,25 +646,26 @@ class UtfNormal {
$combining = '';
$lastClass = 0;
continue;
- } elseif( $n >= 0xf0 ) {
+ } elseif ( $n >= 0xf0 ) {
$c = substr( $string, $i, 4 );
$i += 3;
- } elseif( $n >= 0xe0 ) {
+ } elseif ( $n >= 0xe0 ) {
$c = substr( $string, $i, 3 );
$i += 2;
- } elseif( $n >= 0xc0 ) {
+ } elseif ( $n >= 0xc0 ) {
$c = substr( $string, $i, 2 );
$i++;
}
$pair = $startChar . $c;
- if( $n > 0x80 ) {
- if( isset( self::$utfCombiningClass[$c] ) ) {
+ if ( $n > 0x80 ) {
+ if ( isset( self::$utfCombiningClass[$c] ) ) {
# A combining char; see what we can do with it
$class = self::$utfCombiningClass[$c];
- if( !empty( $startChar ) &&
+ if ( !empty( $startChar ) &&
$lastClass < $class &&
$class > 0 &&
- isset( self::$utfCanonicalComp[$pair] ) ) {
+ isset( self::$utfCanonicalComp[$pair] )
+ ) {
$startChar = self::$utfCanonicalComp[$pair];
$class = 0;
} else {
@@ -670,56 +677,58 @@ class UtfNormal {
}
}
# New start char
- if( $lastClass == 0 ) {
- if( isset( self::$utfCanonicalComp[$pair] ) ) {
+ if ( $lastClass == 0 ) {
+ if ( isset( self::$utfCanonicalComp[$pair] ) ) {
$startChar = self::$utfCanonicalComp[$pair];
$lastHangul = 0;
continue;
}
- if( $n >= $x1 && $n <= $x2 ) {
+ if ( $n >= $x1 && $n <= $x2 ) {
# WARNING: Hangul code is painfully slow.
# I apologize for this ugly, ugly code; however
# performance is even more teh suck if we call
# out to nice clean functions. Lookup tables are
# marginally faster, but require a lot of space.
#
- if( $c >= UTF8_HANGUL_VBASE &&
+ if ( $c >= UTF8_HANGUL_VBASE &&
$c <= UTF8_HANGUL_VEND &&
$startChar >= UTF8_HANGUL_LBASE &&
- $startChar <= UTF8_HANGUL_LEND ) {
+ $startChar <= UTF8_HANGUL_LEND
+ ) {
#
#$lIndex = utf8ToCodepoint( $startChar ) - UNICODE_HANGUL_LBASE;
#$vIndex = utf8ToCodepoint( $c ) - UNICODE_HANGUL_VBASE;
$lIndex = ord( $startChar[2] ) - 0x80;
- $vIndex = ord( $c[2] ) - 0xa1;
+ $vIndex = ord( $c[2] ) - 0xa1;
$hangulPoint = UNICODE_HANGUL_FIRST +
UNICODE_HANGUL_TCOUNT *
- (UNICODE_HANGUL_VCOUNT * $lIndex + $vIndex);
+ ( UNICODE_HANGUL_VCOUNT * $lIndex + $vIndex );
# Hardcode the limited-range UTF-8 conversion:
$startChar = chr( $hangulPoint >> 12 & 0x0f | 0xe0 ) .
- chr( $hangulPoint >> 6 & 0x3f | 0x80 ) .
- chr( $hangulPoint & 0x3f | 0x80 );
+ chr( $hangulPoint >> 6 & 0x3f | 0x80 ) .
+ chr( $hangulPoint & 0x3f | 0x80 );
$lastHangul = 0;
continue;
- } elseif( $c >= UTF8_HANGUL_TBASE &&
- $c <= UTF8_HANGUL_TEND &&
- $startChar >= UTF8_HANGUL_FIRST &&
- $startChar <= UTF8_HANGUL_LAST &&
- !$lastHangul ) {
+ } elseif ( $c >= UTF8_HANGUL_TBASE &&
+ $c <= UTF8_HANGUL_TEND &&
+ $startChar >= UTF8_HANGUL_FIRST &&
+ $startChar <= UTF8_HANGUL_LAST &&
+ !$lastHangul
+ ) {
# $tIndex = utf8ToCodepoint( $c ) - UNICODE_HANGUL_TBASE;
$tIndex = ord( $c[2] ) - 0xa7;
- if( $tIndex < 0 ) $tIndex = ord( $c[2] ) - 0x80 + (0x11c0 - 0x11a7);
+ if ( $tIndex < 0 ) $tIndex = ord( $c[2] ) - 0x80 + ( 0x11c0 - 0x11a7 );
# Increment the code point by $tIndex, without
# the function overhead of decoding and recoding UTF-8
#
$tail = ord( $startChar[2] ) + $tIndex;
- if( $tail > 0xbf ) {
+ if ( $tail > 0xbf ) {
$tail -= 0x40;
$mid = ord( $startChar[1] ) + 1;
- if( $mid > 0xbf ) {
+ if ( $mid > 0xbf ) {
$startChar[0] = chr( ord( $startChar[0] ) + 1 );
$mid -= 0x40;
}
@@ -741,6 +750,7 @@ class UtfNormal {
$lastHangul = 0;
}
$out .= $startChar . $combining;
+
return $out;
}
@@ -753,11 +763,13 @@ class UtfNormal {
static function placebo( $string ) {
$len = strlen( $string );
$out = '';
- for( $i = 0; $i < $len; $i++ ) {
+ for ( $i = 0; $i < $len; $i++ ) {
$out .= $string[$i];
}
+
return $out;
}
+
/**
* Function to replace some characters that we don't want
* but most of the native normalize functions keep.
@@ -772,6 +784,7 @@ class UtfNormal {
$string );
$string = str_replace( UTF8_FFFE, UTF8_REPLACEMENT, $string );
$string = str_replace( UTF8_FFFF, UTF8_REPLACEMENT, $string );
+
return $string;
}
}
generated by cgit v1.2.3-54-g00ecf (git 2.45.2) at 2024-11-18 19:54:49 +0000