From d9022f63880ce039446fba8364f68e656b7bf4cb Mon Sep 17 00:00:00 2001 From: Pierre Schmitz Date: Thu, 3 May 2012 13:01:35 +0200 Subject: Update to MediaWiki 1.19.0 --- maintenance/language/generateCollationData.php | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) (limited to 'maintenance/language/generateCollationData.php') diff --git a/maintenance/language/generateCollationData.php b/maintenance/language/generateCollationData.php index 2c3ffedc..e5ce5c87 100644 --- a/maintenance/language/generateCollationData.php +++ b/maintenance/language/generateCollationData.php @@ -1,4 +1,8 @@ addOption( 'data-dir', 'A directory on the local filesystem ' . - 'containing allkeys.txt and ucd.all.grouped.xml from unicode.org', + 'containing allkeys.txt and ucd.all.grouped.xml from unicode.org', false, true ); $this->addOption( 'debug-output', 'Filename for sending debug output to', false, true ); @@ -72,7 +76,7 @@ class GenerateCollationData extends Maintenance { // but do not skip a normal space (U+0020) since // people like to use that as a fake no header symbol. $category = substr( $data['gc'], 0, 1 ); - if ( strpos( 'LNPS', $category ) === false + if ( strpos( 'LNPS', $category ) === false && $data['cp'] !== '0020' ) { return; } @@ -84,7 +88,7 @@ class GenerateCollationData extends Maintenance { return; } - // Skip the composed Hangul syllables, we will use the bare Jamo + // Skip the composed Hangul syllables, we will use the bare Jamo // as first letters if ( $data['block'] == 'Hangul Syllables' ) { return; @@ -163,7 +167,7 @@ class GenerateCollationData extends Maintenance { } $this->weights[$cp] = $primary; if ( $tertiary === '.0008' - || $tertiary === '.000E' ) + || $tertiary === '.000E' ) { $goodTertiaryChars[$cp] = true; } @@ -192,8 +196,8 @@ class GenerateCollationData extends Maintenance { } // If one character has a given primary weight sequence, and a second - // character has a longer primary weight sequence with an initial - // portion equal to the first character, then remove the second + // character has a longer primary weight sequence with an initial + // portion equal to the first character, then remove the second // character. This avoids having characters like U+A732 (double A) // polluting the basic latin sort area. @@ -306,11 +310,12 @@ class UcdXmlReader { while ( $this->xml->name !== 'ucd' && $this->xml->read() ); $this->xml->read(); return $this->xml; - } + } /** - * Read the attributes of the current element node and return them + * Read the attributes of the current element node and return them * as an array + * @return array */ protected function readAttributes() { $attrs = array(); -- cgit v1.2.2