summaryrefslogtreecommitdiff
path: root/languages/classes/LanguageJa.php
diff options
context:
space:
mode:
authorPierre Schmitz <pierre@archlinux.de>2010-07-28 11:52:48 +0200
committerPierre Schmitz <pierre@archlinux.de>2010-07-28 11:52:48 +0200
commit222b01f5169f1c7e69762e0e8904c24f78f71882 (patch)
tree8e932e12546bb991357ec48eb1638d1770be7a35 /languages/classes/LanguageJa.php
parent00ab76a6b686e98a914afc1975812d2b1aaa7016 (diff)
update to MediaWiki 1.16.0
Diffstat (limited to 'languages/classes/LanguageJa.php')
-rw-r--r--languages/classes/LanguageJa.php27
1 files changed, 13 insertions, 14 deletions
diff --git a/languages/classes/LanguageJa.php b/languages/classes/LanguageJa.php
index 72c06e19..4a24260b 100644
--- a/languages/classes/LanguageJa.php
+++ b/languages/classes/LanguageJa.php
@@ -6,15 +6,11 @@
* @ingroup Language
*/
class LanguageJa extends Language {
- function stripForSearch( $string ) {
- # MySQL fulltext index doesn't grok utf-8, so we
- # need to fold cases and convert to hex
- $s = $string;
+ function wordSegmentation( $string ) {
+ // Strip known punctuation ?
+ // $s = preg_replace( '/\xe3\x80[\x80-\xbf]/', '', $s ); # U3000-303f
- # Strip known punctuation ?
- #$s = preg_replace( '/\xe3\x80[\x80-\xbf]/', '', $s ); # U3000-303f
-
- # Space strings of like hiragana/katakana/kanji
+ // Space strings of like hiragana/katakana/kanji
$hiragana = '(?:\xe3(?:\x81[\x80-\xbf]|\x82[\x80-\x9f]))'; # U3040-309f
$katakana = '(?:\xe3(?:\x82[\xa0-\xbf]|\x83[\x80-\xbf]))'; # U30a0-30ff
$kanji = '(?:\xe3[\x88-\xbf][\x80-\xbf]'
@@ -22,14 +18,17 @@ class LanguageJa extends Language {
. '|\xe9[\x80-\xa5][\x80-\xbf]'
. '|\xe9\xa6[\x80-\x99])';
# U3200-9999 = \xe3\x88\x80-\xe9\xa6\x99
- $s = preg_replace( "/({$hiragana}+|{$katakana}+|{$kanji}+)/", ' $1 ', $s );
-
- # Double-width roman characters: ff00-ff5f ~= 0020-007f
- $s = preg_replace( '/\xef\xbc([\x80-\xbf])/e', 'chr((ord("$1") & 0x3f) + 0x20)', $s );
- $s = preg_replace( '/\xef\xbd([\x80-\x99])/e', 'chr((ord("$1") & 0x3f) + 0x60)', $s );
+ $reg = "/({$hiragana}+|{$katakana}+|{$kanji}+)/";
+ $s = self::insertSpace( $string, $reg );
+ return $s;
+ }
+ function normalizeForSearch( $string ) {
+ // Double-width roman characters
+ $s = self::convertDoubleWidth( $string );
+
# Do general case folding and UTF-8 armoring
- return parent::stripForSearch( $s );
+ return parent::normalizeForSearch( $s );
}
# Italic is not appropriate for Japanese script