<?php
/*
* Created on Oct 19, 2006
*
* API for MediaWiki 1.8+
*
* Copyright (C) 2006 Yuri Astrakhan <Firstname><Lastname>@gmail.com
*
* 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
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License along
* with this program; if not, write to the Free Software Foundation, Inc.,
* 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
* http://www.gnu.org/copyleft/gpl.html
*/
if ( !defined( 'MEDIAWIKI' ) ) {
// Eclipse helper - will be ignored in production
require_once ( 'ApiQueryBase.php' );
}
/**
* A query action to enumerate the recent changes that were done to the wiki.
* Various filters are supported.
*
* @ingroup API
*/
class ApiQueryRecentChanges extends ApiQueryBase {
public function __construct( $query, $moduleName ) {
parent :: __construct( $query, $moduleName, 'rc' );
}
private $fld_comment = false, $fld_parsedcomment = false, $fld_user = false, $fld_flags = false,
$fld_timestamp = false, $fld_title = false, $fld_ids = false,
$fld_sizes = false;
/**
* Get an array mapping token names to their handler functions.
* The prototype for a token function is func($pageid, $title, $rc)
* it should return a token or false (permission denied)
* @return array(tokenname => function)
*/
protected function getTokenFunctions() {
// Don't call the hooks twice
if ( isset( $this->tokenFunctions ) )
return $this->tokenFunctions;
// If we're in JSON callback mode, no tokens can be obtained
if ( !is_null( $this->getMain()->getRequest()->getVal( 'callback' ) ) )
return array();
$this->tokenFunctions = array(
'patrol' => array( 'ApiQueryRecentChanges', 'getPatrolToken' )
);
wfRunHooks( 'APIQueryRecentChangesTokens', array( &$this->tokenFunctions ) );
return $this->tokenFunctions;
}
public static function getPatrolToken( $pageid, $title, $rc )
{
global $wgUser;
if ( !$wgUser->useRCPatrol() && ( !$wgUser->useNPPatrol() ||
$rc->getAttribute( 'rc_type' ) != RC_NEW ) )
return false;
// The patrol token is always the same, let's exploit that
static $cachedPatrolToken = null;
if ( !is_null( $cachedPatrolToken ) )
return $cachedPatrolToken;
$cachedPatrolToken = $wgUser->editToken();
return $cachedPatrolToken;
}
/**
* Sets internal state to include the desired properties in the output.
* @param $prop associative array of properties, only keys are used here
*/
public function initProperties( $prop ) {
$this->fld_comment = isset ( $prop['comment'] );
$this->fld_parsedcomment = isset ( $prop['parsedcomment'] );
$this->fld_user = isset ( $prop['user'] );
$this->fld_flags = isset ( $prop['flags'] );
$this->fld_timestamp = isset ( $prop['timestamp'] );
$this->fld_title = isset ( $prop['title'] );
$this->fld_ids = isset ( $prop['ids'] );
$this->fld_sizes = isset ( $prop['sizes'] );
$this->fld_redirect = isset( $prop['redirect'] );
$this->fld_patrolled = isset( $prop['patrolled'] );
$this->fld_loginfo = isset( $prop['loginfo'] );
$this->fld_tags = isset( $prop['tags'] );
}
/**
* Generates and outputs the result of this query based upon the provided parameters.
*/
public function execute() {
/* Get the parameters of the request. */
$params = $this->extractRequestParams();
/* Build our basic query. Namely, something along the lines of:
* SELECT * FROM recentchanges WHERE rc_timestamp > $start
* AND rc_timestamp < $end AND rc_namespace = $namespace
* AND rc_deleted = '0'
*/
$db = $this->getDB();
$this->addTables( 'recentchanges' );
$index = array( 'recentchanges' => 'rc_timestamp' ); // May change
$this->addWhereRange( 'rc_timestamp', $params['dir'], $params['start'], $params['end'] );
$this->addWhereFld( 'rc_namespace', $params['namespace'] );
$this->addWhereFld( 'rc_deleted', 0 );
if ( !is_null( $params['type'] ) )
$this->addWhereFld( 'rc_type', $this->parseRCType( $params['type'] ) );
if ( !is_null( $params['show'] ) ) {
$show = array_flip( $params['show'] );
/* Check for conflicting parameters. */
if ( ( isset ( $show['minor'] ) && isset ( $show['!minor'] ) )
|| ( isset ( $show['bot'] ) && isset ( $show['!bot'] ) )
|| ( isset ( $show['anon'] ) && isset ( $show['!anon'] ) )
|| ( isset ( $show['redirect'] ) && isset ( $show['!redirect'] ) )
|| ( isset ( $show['patrolled'] ) && isset ( $show['!patrolled'] ) ) ) {
$this->dieUsageMsg( array( 'show' ) );
}
// Check permissions
global $wgUser;
if ( ( isset( $show['patrolled'] ) || isset( $show['!patrolled'] ) ) && !$wgUser->useRCPatrol() && !$wgUser->useNPPatrol() )
$this->dieUsage( "You need the patrol right to request the patrolled flag", 'permissiondenied' );
/* Add additional conditions to query depending upon parameters. */
$this->addWhereIf( 'rc_minor = 0', isset ( $show['!minor'] ) );
$this->addWhereIf( 'rc_minor != 0', isset ( $show['minor'] ) );
$this->addWhereIf( 'rc_bot = 0', isset ( $show['!bot'] ) );
$this->addWhereIf( 'rc_bot != 0', isset ( $show['bot'] ) );
$this->addWhereIf( 'rc_user = 0', isset ( $show['anon'] ) );
$this->addWhereIf( 'rc_user != 0', isset ( $show['!anon'] ) );
$this->addWhereIf( 'rc_patrolled = 0', isset( $show['!patrolled'] ) );
$this->addWhereIf( 'rc_patrolled != 0', isset( $show['patrolled'] ) );
$this->addWhereIf( 'page_is_redirect = 1', isset ( $show['redirect'] ) );
// Don't throw log entries out the window here
$this->addWhereIf( 'page_is_redirect = 0 OR page_is_redirect IS NULL', isset ( $show['!redirect'] ) );
}
if ( !is_null( $params['user'] ) && !is_null( $param['excludeuser'] ) )
$this->dieUsage( 'user and excludeuser cannot be used together', 'user-excludeuser' );
if ( !is_null( $params['user'] ) )
{
$this->addWhereFld( 'rc_user_text', $params['user'] );
$index['recentchanges'] = 'rc_user_text';
}
if ( !is_null( $params['excludeuser'] ) )
// We don't use the rc_user_text index here because
// * it would require us to sort by rc_user_text before rc_timestamp
// * the != condition doesn't throw out too many rows anyway
$this->addWhere( 'rc_user_text != ' . $this->getDB()->addQuotes( $params['excludeuser'] ) );
/* Add the fields we're concerned with to our query. */
$this->addFields( array (
'rc_timestamp',
'rc_namespace',
'rc_title',
'rc_cur_id',
'rc_type',
'rc_moved_to_ns',
'rc_moved_to_title'
) );
/* Determine what properties we need to display. */
if ( !is_null( $params['prop'] ) ) {
$prop = array_flip( $params['prop'] );
/* Set up internal members based upon params. */
$this->initProperties( $prop );
global $wgUser;
if ( $this->fld_patrolled && !$wgUser->useRCPatrol() && !$wgUser->useNPPatrol() )
$this->dieUsage( "You need the patrol right to request the patrolled flag", 'permissiondenied' );
/* Add fields to our query if they are specified as a needed parameter. */
$this->addFieldsIf( 'rc_id', $this->fld_ids );
$this->addFieldsIf( 'rc_this_oldid', $this->fld_ids );
$this->addFieldsIf( 'rc_last_oldid', $this->fld_ids );
$this->addFieldsIf( 'rc_comment', $this->fld_comment || $this->fld_parsedcomment );
$this->addFieldsIf( 'rc_user', $this->fld_user );
$this->addFieldsIf( 'rc_user_text', $this->fld_user );
$this->addFieldsIf( 'rc_minor', $this->fld_flags );
$this->addFieldsIf( 'rc_bot', $this->fld_flags );
$this->addFieldsIf( 'rc_new', $this->fld_flags );
$this->addFieldsIf( 'rc_old_len', $this->fld_sizes );
$this->addFieldsIf( 'rc_new_len', $this->fld_sizes );
$this->addFieldsIf( 'rc_patrolled', $this->fld_patrolled );
$this->addFieldsIf( 'rc_logid', $this->fld_loginfo );
$this->addFieldsIf( 'rc_log_type', $this->fld_loginfo );
$this->addFieldsIf( 'rc_log_action', $this->fld_loginfo );
$this->addFieldsIf( 'rc_params', $this->fld_loginfo );
if ( $this->fld_redirect || isset( $show['redirect'] ) || isset( $show['!redirect'] ) )
{
$this->addTables( 'page' );
$this->addJoinConds( array( 'page' => array( 'LEFT JOIN', array( 'rc_namespace=page_namespace', 'rc_title=page_title' ) ) ) );
$this->addFields( 'page_is_redirect' );
}
}
if ( $this->fld_tags ) {
$this->addTables( 'tag_summary' );
$this->addJoinConds( array( 'tag_summary' => array( 'LEFT JOIN', array( 'rc_id=ts_rc_id' ) ) ) );
$this->addFields( 'ts_tags' );
}
if ( !is_null( $params['tag'] ) ) {
$this->addTables( 'change_tag' );
$this->addJoinConds( array( 'change_tag' => array( 'INNER JOIN', array( 'rc_id=ct_rc_id' ) ) ) );
$this->addWhereFld( 'ct_tag' , $params['tag'] );
global $wgOldChangeTagsIndex;
$index['change_tag'] = $wgOldChangeTagsIndex ? 'ct_tag' : 'change_tag_tag_id';
}
$this->token = $params['token'];
$this->addOption( 'LIMIT', $params['limit'] + 1 );
$this->addOption( 'USE INDEX', $index );
$count = 0;
/* Perform the actual query. */
$db = $this->getDB();
$res = $this->select( __METHOD__ );
/* Iterate through the rows, adding data extracted from them to our query result. */
while ( $row = $db->fetchObject( $res ) ) {
if ( ++ $count > $params['limit'] ) {
// We've reached the one extra which shows that there are additional pages to be had. Stop here...
$this->setContinueEnumParameter( 'start', wfTimestamp( TS_ISO_8601, $row->rc_timestamp ) );
break;
}
/* Extract the data from a single row. */
$vals = $this->extractRowInfo( $row );
/* Add that row's data to our final output. */
if ( !$vals )
continue;
$fit = $this->getResult()->addValue( array( 'query', $this->getModuleName() ), null, $vals );
if ( !$fit )
{
$this->setContinueEnumParameter( 'start', wfTimestamp( TS_ISO_8601, $row->rc_timestamp ) );
break;
}
}
$db->freeResult( $res );
/* Format the result */
$this->getResult()->setIndexedTagName_internal( array( 'query', $this->getModuleName() ), 'rc' );
}
/**
* Extracts from a single sql row the data needed to describe one recent change.
*
* @param $row The row from which to extract the data.
* @return An array mapping strings (descriptors) to their respective string values.
* @access public
*/
public function extractRowInfo( $row ) {
/* If page was moved somewhere, get the title of the move target. */
$movedToTitle = false;
if ( isset( $row->rc_moved_to_title ) && $row->rc_moved_to_title !== '' )
$movedToTitle = Title :: makeTitle( $row->rc_moved_to_ns, $row->rc_moved_to_title );
/* Determine the title of the page that has been changed. */
$title = Title :: makeTitle( $row->rc_namespace, $row->rc_title );
/* Our output data. */
$vals = array ();
$type = intval ( $row->rc_type );
/* Determine what kind of change this was. */
switch ( $type ) {
case RC_EDIT:
$vals['type'] = 'edit';
break;
case RC_NEW:
$vals['type'] = 'new';
break;
case RC_MOVE:
$vals['type'] = 'move';
break;
case RC_LOG:
$vals['type'] = 'log';
break;
case RC_MOVE_OVER_REDIRECT:
$vals['type'] = 'move over redirect';
break;
default:
$vals['type'] = $type;
}
/* Create a new entry in the result for the title. */
if ( $this->fld_title ) {
ApiQueryBase :: addTitleInfo( $vals, $title );
if ( $movedToTitle )
ApiQueryBase :: addTitleInfo( $vals, $movedToTitle, "new_" );
}
/* Add ids, such as rcid, pageid, revid, and oldid to the change's info. */
if ( $this->fld_ids ) {
$vals['rcid'] = intval( $row->rc_id );
$vals['pageid'] = intval( $row->rc_cur_id );
$vals['revid'] = intval( $row->rc_this_oldid );
$vals['old_revid'] = intval( $row->rc_last_oldid );
}
/* Add user data and 'anon' flag, if use is anonymous. */
if ( $this->fld_user ) {
$vals['user'] = $row->rc_user_text;
if ( !$row->rc_user )
$vals['anon'] = '';
}
/* Add flags, such as new, minor, bot. */
if ( $this->fld_flags ) {
if ( $row->rc_bot )
$vals['bot'] = '';
if ( $row->rc_new )
$vals['new'] = '';
if ( $row->rc_minor )
$vals['minor'] = '';
}
/* Add sizes of each revision. (Only available on 1.10+) */
if ( $this->fld_sizes ) {
$vals['oldlen'] = intval( $row->rc_old_len );
$vals['newlen'] = intval( $row->rc_new_len );
}
/* Add the timestamp. */
if ( $this->fld_timestamp )
$vals['timestamp'] = wfTimestamp( TS_ISO_8601, $row->rc_timestamp );
/* Add edit summary / log summary. */
if ( $this->fld_comment && isset( $row->rc_comment ) )
$vals['comment'] = $row->rc_comment;
if ( $this->fld_parsedcomment && isset( $row->rc_comment ) ) {
global $wgUser;
$vals['parsedcomment'] = $wgUser->getSkin()->formatComment( $row->rc_comment, $title );
}
if ( $this->fld_redirect )
if ( $row->page_is_redirect )
$vals['redirect'] = '';
/* Add the patrolled flag */
if ( $this->fld_patrolled && $row->rc_patrolled == 1 )
$vals['patrolled'] = '';
if ( $this->fld_loginfo && $row->rc_type == RC_LOG ) {
$vals['logid'] = intval( $row->rc_logid );
$vals['logtype'] = $row->rc_log_type;
$vals['logaction'] = $row->rc_log_action;
ApiQueryLogEvents::addLogParams( $this->getResult(),
$vals, $row->rc_params,
$row->rc_log_type, $row->rc_timestamp );
}
if ( $this->fld_tags ) {
if ( $row->ts_tags ) {
$tags = explode( ',', $row->ts_tags );
$this->getResult()->setIndexedTagName( $tags, 'tag' );
$vals['tags'] = $tags;
} else {
$vals['tags'] = array();
}
}
if ( !is_null( $this->token ) )
{
$tokenFunctions = $this->getTokenFunctions();
foreach ( $this->token as $t )
{
$val = call_user_func( $tokenFunctions[$t], $row->rc_cur_id,
$title, RecentChange::newFromRow( $row ) );
if ( $val === false )
$this->setWarning( "Action '$t' is not allowed for the current user" );
else
$vals[$t . 'token'] = $val;
}
}
return $vals;
}
private function parseRCType( $type )
{
if ( is_array( $type ) )
{
$retval = array();
foreach ( $type as $t )
$retval[] = $this->parseRCType( $t );
return $retval;
}
switch( $type )
{
case 'edit': return RC_EDIT;
case 'new': return RC_NEW;
case 'log': return RC_LOG;
}
}
public function getCacheMode( $params ) {
if ( isset( $params['show'] ) ) {
foreach ( $params['show'] as $show ) {
if ( $show === 'patrolled' || $show === '!patrolled' ) {
return 'private';
}
}
}
if ( isset( $params['token'] ) ) {
return 'private';
}
if ( !is_null( $params['prop'] ) && in_array( 'parsedcomment', $params['prop'] ) ) {
// formatComment() calls wfMsg() among other things
return 'anon-public-user-private';
}
return 'public';
}
public function getAllowedParams() {
return array (
'start' => array (
ApiBase :: PARAM_TYPE => 'timestamp'
),
'end' => array (
ApiBase :: PARAM_TYPE => 'timestamp'
),
'dir' => array (
ApiBase :: PARAM_DFLT => 'older',
ApiBase :: PARAM_TYPE => array (
'newer',
'older'
)
),
'namespace' => array (
ApiBase :: PARAM_ISMULTI => true,
ApiBase :: PARAM_TYPE => 'namespace'
),
'user' => array(
ApiBase :: PARAM_TYPE => 'user'
),
'excludeuser' => array(
ApiBase :: PARAM_TYPE => 'user'
),
'tag' => null,
'prop' => array (
ApiBase :: PARAM_ISMULTI => true,
ApiBase :: PARAM_DFLT => 'title|timestamp|ids',
ApiBase :: PARAM_TYPE => array (
'user',
'comment',
'parsedcomment',
'flags',
'timestamp',
'title',
'ids',
'sizes',
'redirect',
'patrolled',
'loginfo',
'tags'
)
),
'token' => array(
ApiBase :: PARAM_TYPE => array_keys( $this->getTokenFunctions() ),
ApiBase :: PARAM_ISMULTI => true
),
'show' => array (
ApiBase :: PARAM_ISMULTI => true,
ApiBase :: PARAM_TYPE => array (
'minor',
'!minor',
'bot',
'!bot',
'anon',
'!anon',
'redirect',
'!redirect',
'patrolled',
'!patrolled'
)
),
'limit' => array (
ApiBase :: PARAM_DFLT => 10,
ApiBase :: PARAM_TYPE => 'limit',
ApiBase :: PARAM_MIN => 1,
ApiBase :: PARAM_MAX => ApiBase :: LIMIT_BIG1,
ApiBase :: PARAM_MAX2 => ApiBase :: LIMIT_BIG2
),
'type' => array (
ApiBase :: PARAM_ISMULTI => true,
ApiBase :: PARAM_TYPE => array (
'edit',
'new',
'log'
)
)
);
}
public function getParamDescription() {
return array (
'start' => 'The timestamp to start enumerating from.',
'end' => 'The timestamp to end enumerating.',
'dir' => 'In which direction to enumerate.',
'namespace' => 'Filter log entries to only this namespace(s)',
'user' => 'Only list changes by this user',
'excludeuser' => 'Don\'t list changes by this user',
'prop' => 'Include additional pieces of information',
'token' => 'Which tokens to obtain for each change',
'show' => array (
'Show only items that meet this criteria.',
'For example, to see only minor edits done by logged-in users, set show=minor|!anon'
),
'type' => 'Which types of changes to show.',
'limit' => 'How many total changes to return.',
'tag' => 'Only list changes tagged with this tag.',
);
}
public function getDescription() {
return 'Enumerate recent changes';
}
public function getPossibleErrors() {
return array_merge( parent::getPossibleErrors(), array(
array( 'show' ),
array( 'code' => 'permissiondenied', 'info' => 'You need the patrol right to request the patrolled flag' ),
array( 'code' => 'user-excludeuser', 'info' => 'user and excludeuser cannot be used together' ),
) );
}
protected function getExamples() {
return array (
'api.php?action=query&list=recentchanges'
);
}
public function getVersion() {
return __CLASS__ . ': $Id: ApiQueryRecentChanges.php 69932 2010-07-26 08:03:21Z tstarling $';
}
}
/**
* @defgroup Language Language
*
* @file
* @ingroup Language
*/
if( !defined( 'MEDIAWIKI' ) ) {
echo "This file is part of MediaWiki, it is not a valid entry point.\n";
exit( 1 );
}
# Read language names
global $wgLanguageNames;
require_once( dirname(__FILE__) . '/Names.php' ) ;
global $wgInputEncoding, $wgOutputEncoding;
/**
* These are always UTF-8, they exist only for backwards compatibility
*/
$wgInputEncoding = "UTF-8";
$wgOutputEncoding = "UTF-8";
if( function_exists( 'mb_strtoupper' ) ) {
mb_internal_encoding('UTF-8');
}
/**
* a fake language converter
*
* @ingroup Language
*/
class FakeConverter {
var $mLang;
function FakeConverter( $langobj ) { $this->mLang = $langobj; }
function autoConvertToAllVariants( $text ) { return $text; }
function convert( $t ) { return $t; }
function convertTitle( $t ) { return $t->getPrefixedText(); }
function getVariants() { return array( $this->mLang->getCode() ); }
function getPreferredVariant() { return $this->mLang->getCode(); }
function getConvRuleTitle() { return false; }
function findVariantLink(&$l, &$n, $ignoreOtherCond = false) {}
function getExtraHashOptions() {return '';}
function getParsedTitle() {return '';}
function markNoConversion($text, $noParse=false) {return $text;}
function convertCategoryKey( $key ) {return $key; }
function convertLinkToAllVariants($text){ return array( $this->mLang->getCode() => $text); }
function armourMath($text){ return $text; }
}
/**
* Internationalisation code
* @ingroup Language
*/
class Language {
var $mConverter, $mVariants, $mCode, $mLoaded = false;
var $mMagicExtensions = array(), $mMagicHookDone = false;
var $mNamespaceIds, $namespaceNames, $namespaceAliases;
var $dateFormatStrings = array();
var $mExtendedSpecialPageAliases;
/**
* ReplacementArray object caches
*/
var $transformData = array();
static public $dataCache;
static public $mLangObjCache = array();
static public $mWeekdayMsgs = array(
'sunday', 'monday', 'tuesday', 'wednesday', 'thursday',
'friday', 'saturday'
);
static public $mWeekdayAbbrevMsgs = array(
'sun', 'mon', 'tue', 'wed', 'thu', 'fri', 'sat'
);
static public $mMonthMsgs = array(
'january', 'february', 'march', 'april', 'may_long', 'june',
'july', 'august', 'september', 'october', 'november',
'december'
);
static public $mMonthGenMsgs = array(
'january-gen', 'february-gen', 'march-gen', 'april-gen', 'may-gen', 'june-gen',
'july-gen', 'august-gen', 'september-gen', 'october-gen', 'november-gen',
'december-gen'
);
static public $mMonthAbbrevMsgs = array(
'jan', 'feb', 'mar', 'apr', 'may', 'jun', 'jul', 'aug',
'sep', 'oct', 'nov', 'dec'
);
static public $mIranianCalendarMonthMsgs = array(
'iranian-calendar-m1', 'iranian-calendar-m2', 'iranian-calendar-m3',
'iranian-calendar-m4', 'iranian-calendar-m5', 'iranian-calendar-m6',
'iranian-calendar-m7', 'iranian-calendar-m8', 'iranian-calendar-m9',
'iranian-calendar-m10', 'iranian-calendar-m11', 'iranian-calendar-m12'
);
static public $mHebrewCalendarMonthMsgs = array(
'hebrew-calendar-m1', 'hebrew-calendar-m2', 'hebrew-calendar-m3',
'hebrew-calendar-m4', 'hebrew-calendar-m5', 'hebrew-calendar-m6',
'hebrew-calendar-m7', 'hebrew-calendar-m8', 'hebrew-calendar-m9',
'hebrew-calendar-m10', 'hebrew-calendar-m11', 'hebrew-calendar-m12',
'hebrew-calendar-m6a', 'hebrew-calendar-m6b'
);
static public $mHebrewCalendarMonthGenMsgs = array(
'hebrew-calendar-m1-gen', 'hebrew-calendar-m2-gen', 'hebrew-calendar-m3-gen',
'hebrew-calendar-m4-gen', 'hebrew-calendar-m5-gen', 'hebrew-calendar-m6-gen',
'hebrew-calendar-m7-gen', 'hebrew-calendar-m8-gen', 'hebrew-calendar-m9-gen',
'hebrew-calendar-m10-gen', 'hebrew-calendar-m11-gen', 'hebrew-calendar-m12-gen',
'hebrew-calendar-m6a-gen', 'hebrew-calendar-m6b-gen'
);
static public $mHijriCalendarMonthMsgs = array(
'hijri-calendar-m1', 'hijri-calendar-m2', 'hijri-calendar-m3',
'hijri-calendar-m4', 'hijri-calendar-m5', 'hijri-calendar-m6',
'hijri-calendar-m7', 'hijri-calendar-m8', 'hijri-calendar-m9',
'hijri-calendar-m10', 'hijri-calendar-m11', 'hijri-calendar-m12'
);
/**
* Get a cached language object for a given language code
*/
static function factory( $code ) {
if ( !isset( self::$mLangObjCache[$code] ) ) {
if( count( self::$mLangObjCache ) > 10 ) {
// Don't keep a billion objects around, that's stupid.
self::$mLangObjCache = array();
}
self::$mLangObjCache[$code] = self::newFromCode( $code );
}
return self::$mLangObjCache[$code];
}
/**
* Create a language object for a given language code
*/
protected static function newFromCode( $code ) {
global $IP;
static $recursionLevel = 0;
// Protect against path traversal below
if ( !Language::isValidCode( $code )
|| strcspn( $code, "/\\\000" ) !== strlen( $code ) )
{
throw new MWException( "Invalid language code \"$code\"" );
}
if ( $code == 'en' ) {
$class = 'Language';
} else {
$class = 'Language' . str_replace( '-', '_', ucfirst( $code ) );
// Preload base classes to work around APC/PHP5 bug
if ( file_exists( "$IP/languages/classes/$class.deps.php" ) ) {
include_once("$IP/languages/classes/$class.deps.php");
}
if ( file_exists( "$IP/languages/classes/$class.php" ) ) {
include_once("$IP/languages/classes/$class.php");
}
}
if ( $recursionLevel > 5 ) {
throw new MWException( "Language fallback loop detected when creating class $class\n" );
}
if( ! class_exists( $class ) ) {
$fallback = Language::getFallbackFor( $code );
++$recursionLevel;
$lang = Language::newFromCode( $fallback );
--$recursionLevel;
$lang->setCode( $code );
} else {
$lang = new $class;
}
return $lang;
}
/**
* Returns true if a language code string is of a valid form, whether or
* not it exists.
*/
public static function isValidCode( $code ) {
return strcspn( $code, "/\\\000" ) === strlen( $code );
}
/**
* Get the LocalisationCache instance
*/
public static function getLocalisationCache() {
if ( is_null( self::$dataCache ) ) {
global $wgLocalisationCacheConf;
$class = $wgLocalisationCacheConf['class'];
self::$dataCache = new $class( $wgLocalisationCacheConf );
}
return self::$dataCache;
}
function __construct() {
$this->mConverter = new FakeConverter($this);
// Set the code to the name of the descendant
if ( get_class( $this ) == 'Language' ) {
$this->mCode = 'en';
} else {
$this->mCode = str_replace( '_', '-', strtolower( substr( get_class( $this ), 8 ) ) );
}
self::getLocalisationCache();
}
/**
* Reduce memory usage
*/
function __destruct() {
foreach ( $this as $name => $value ) {
unset( $this->$name );
}
}
/**
* Hook which will be called if this is the content language.
* Descendants can use this to register hook functions or modify globals
*/
function initContLang() {}
/**
* @deprecated Use User::getDefaultOptions()
* @return array
*/
function getDefaultUserOptions() {
wfDeprecated( __METHOD__ );
return User::getDefaultOptions();
}
function getFallbackLanguageCode() {
if ( $this->mCode === 'en' ) {
return false;
} else {
return self::$dataCache->getItem( $this->mCode, 'fallback' );
}
}
/**
* Exports $wgBookstoreListEn
* @return array
*/
function getBookstoreList() {
return self::$dataCache->getItem( $this->mCode, 'bookstoreList' );
}
/**
* @return array
*/
function getNamespaces() {
if ( is_null( $this->namespaceNames ) ) {
global $wgExtraNamespaces, $wgMetaNamespace, $wgMetaNamespaceTalk;
$this->namespaceNames = self::$dataCache->getItem( $this->mCode, 'namespaceNames' );
if ( $wgExtraNamespaces ) {
$this->namespaceNames = $wgExtraNamespaces + $this->namespaceNames;
}
$this->namespaceNames[NS_PROJECT] = $wgMetaNamespace;
if ( $wgMetaNamespaceTalk ) {
$this->namespaceNames[NS_PROJECT_TALK] = $wgMetaNamespaceTalk;
} else {
$talk = $this->namespaceNames[NS_PROJECT_TALK];
$this->namespaceNames[NS_PROJECT_TALK] =
$this->fixVariableInNamespace( $talk );
}
# The above mixing may leave namespaces out of canonical order.
# Re-order by namespace ID number...
ksort( $this->namespaceNames );
}
return $this->namespaceNames;
}
/**
* A convenience function that returns the same thing as
* getNamespaces() except with the array values changed to ' '
* where it found '_', useful for producing output to be displayed
* e.g. in <select> forms.
*
* @return array
*/
function getFormattedNamespaces() {
$ns = $this->getNamespaces();
foreach($ns as $k => $v) {
$ns[$k] = strtr($v, '_', ' ');
}
return $ns;
}
/**
* Get a namespace value by key
* <code>
* $mw_ns = $wgContLang->getNsText( NS_MEDIAWIKI );
* echo $mw_ns; // prints 'MediaWiki'
* </code>
*
* @param $index Int: the array key of the namespace to return
* @return mixed, string if the namespace value exists, otherwise false
*/
function getNsText( $index ) {
$ns = $this->getNamespaces();
return isset( $ns[$index] ) ? $ns[$index] : false;
}
/**
* A convenience function that returns the same thing as
* getNsText() except with '_' changed to ' ', useful for
* producing output.
*
* @return array
*/
function getFormattedNsText( $index ) {
$ns = $this->getNsText( $index );
return strtr($ns, '_', ' ');
}
/**
* Get a namespace key by value, case insensitive.
* Only matches namespace names for the current language, not the
* canonical ones defined in Namespace.php.
*
* @param $text String
* @return mixed An integer if $text is a valid value otherwise false
*/
function getLocalNsIndex( $text ) {
$lctext = $this->lc($text);
$ids = $this->getNamespaceIds();
return isset( $ids[$lctext] ) ? $ids[$lctext] : false;
}
function getNamespaceAliases() {
if ( is_null( $this->namespaceAliases ) ) {
$aliases = self::$dataCache->getItem( $this->mCode, 'namespaceAliases' );
if ( !$aliases ) {
$aliases = array();
} else {
foreach ( $aliases as $name => $index ) {
if ( $index === NS_PROJECT_TALK ) {
unset( $aliases[$name] );
$name = $this->fixVariableInNamespace( $name );
$aliases[$name] = $index;
}
}
}
$this->namespaceAliases = $aliases;
}
return $this->namespaceAliases;
}
function getNamespaceIds() {
if ( is_null( $this->mNamespaceIds ) ) {
global $wgNamespaceAliases;
# Put namespace names and aliases into a hashtable.
# If this is too slow, then we should arrange it so that it is done
# before caching. The catch is that at pre-cache time, the above
# class-specific fixup hasn't been done.
$this->mNamespaceIds = array();
foreach ( $this->getNamespaces() as $index => $name ) {
$this->mNamespaceIds[$this->lc($name)] = $index;
}
foreach ( $this->getNamespaceAliases() as $name => $index ) {
$this->mNamespaceIds[$this->lc($name)] = $index;
}
if ( $wgNamespaceAliases ) {
foreach ( $wgNamespaceAliases as $name => $index ) {
$this->mNamespaceIds[$this->lc($name)] = $index;
}
}
}
return $this->mNamespaceIds;
}
/**
* Get a namespace key by value, case insensitive. Canonical namespace
* names override custom ones defined for the current language.
*
* @param $text String
* @return mixed An integer if $text is a valid value otherwise false
*/
function getNsIndex( $text ) {
$lctext = $this->lc($text);
if ( ( $ns = MWNamespace::getCanonicalIndex( $lctext ) ) !== null ) {
return $ns;
}
$ids = $this->getNamespaceIds();
return isset( $ids[$lctext] ) ? $ids[$lctext] : false;
}
/**
* short names for language variants used for language conversion links.
*
* @param $code String
* @return string
*/
function getVariantname( $code ) {
return $this->getMessageFromDB( "variantname-$code" );
}
function specialPage( $name ) {
$aliases = $this->getSpecialPageAliases();
if ( isset( $aliases[$name][0] ) ) {
$name = $aliases[$name][0];
}
return $this->getNsText( NS_SPECIAL ) . ':' . $name;
}
function getQuickbarSettings() {
return array(
$this->getMessage( 'qbsettings-none' ),
$this->getMessage( 'qbsettings-fixedleft' ),
$this->getMessage( 'qbsettings-fixedright' ),
$this->getMessage( 'qbsettings-floatingleft' ),
$this->getMessage( 'qbsettings-floatingright' )
);
}
function getMathNames() {
return self::$dataCache->getItem( $this->mCode, 'mathNames' );
}
function getDatePreferences() {
return self::$dataCache->getItem( $this->mCode, 'datePreferences' );
}
function getDateFormats() {
return self::$dataCache->getItem( $this->mCode, 'dateFormats' );
}
function getDefaultDateFormat() {
$df = self::$dataCache->getItem( $this->mCode, 'defaultDateFormat' );
if ( $df === 'dmy or mdy' ) {
global $wgAmericanDates;
return $wgAmericanDates ? 'mdy' : 'dmy';
} else {
return $df;
}
}
function getDatePreferenceMigrationMap() {
return self::$dataCache->getItem( $this->mCode, 'datePreferenceMigrationMap' );
}
function getImageFile( $image ) {
return self::$dataCache->getSubitem( $this->mCode, 'imageFiles', $image );
}
function getDefaultUserOptionOverrides() {
return self::$dataCache->getItem( $this->mCode, 'defaultUserOptionOverrides' );
}
function getExtraUserToggles() {
return self::$dataCache->getItem( $this->mCode, 'extraUserToggles' );
}
function getUserToggle( $tog ) {
return $this->getMessageFromDB( "tog-$tog" );
}
/**
* Get language names, indexed by code.
* If $customisedOnly is true, only returns codes with a messages file
*/
public static function getLanguageNames( $customisedOnly = false ) {
global $wgLanguageNames, $wgExtraLanguageNames;
$allNames = $wgExtraLanguageNames + $wgLanguageNames;
if ( !$customisedOnly ) {
return $allNames;
}
global $IP;
$names = array();
$dir = opendir( "$IP/languages/messages" );
while( false !== ( $file = readdir( $dir ) ) ) {
$code = self::getCodeFromFileName( $file, 'Messages' );
if ( $code && isset( $allNames[$code] ) ) {
$names[$code] = $allNames[$code];
}
}
closedir( $dir );
return $names;
}
/**
* Get a message from the MediaWiki namespace.
*
* @param $msg String: message name
* @return string
*/
function getMessageFromDB( $msg ) {
return wfMsgExt( $msg, array( 'parsemag', 'language' => $this ) );
}
function getLanguageName( $code ) {
$names = self::getLanguageNames();
if ( !array_key_exists( $code, $names ) ) {
return '';
}
return $names[$code];
}
function getMonthName( $key ) {
return $this->getMessageFromDB( self::$mMonthMsgs[$key-1] );
}
function getMonthNameGen( $key ) {
return $this->getMessageFromDB( self::$mMonthGenMsgs[$key-1] );
}
function getMonthAbbreviation( $key ) {
return $this->getMessageFromDB( self::$mMonthAbbrevMsgs[$key-1] );
}
function getWeekdayName( $key ) {
return $this->getMessageFromDB( self::$mWeekdayMsgs[$key-1] );
}
function getWeekdayAbbreviation( $key ) {
return $this->getMessageFromDB( self::$mWeekdayAbbrevMsgs[$key-1] );
}
function getIranianCalendarMonthName( $key ) {
return $this->getMessageFromDB( self::$mIranianCalendarMonthMsgs[$key-1] );
}
function getHebrewCalendarMonthName( $key ) {
return $this->getMessageFromDB( self::$mHebrewCalendarMonthMsgs[$key-1] );
}
function getHebrewCalendarMonthNameGen( $key ) {
return $this->getMessageFromDB( self::$mHebrewCalendarMonthGenMsgs[$key-1] );
}
function getHijriCalendarMonthName( $key ) {
return $this->getMessageFromDB( self::$mHijriCalendarMonthMsgs[$key-1] );
}
/**
* Used by date() and time() to adjust the time output.
*
* @param $ts Int the time in date('YmdHis') format
* @param $tz Mixed: adjust the time by this amount (default false, mean we
* get user timecorrection setting)
* @return int
*/
function userAdjust( $ts, $tz = false ) {
global $wgUser, $wgLocalTZoffset;
if ( $tz === false ) {
$tz = $wgUser->getOption( 'timecorrection' );
}
$data = explode( '|', $tz, 3 );
if ( $data[0] == 'ZoneInfo' ) {
if ( function_exists( 'timezone_open' ) && @timezone_open( $data[2] ) !== false ) {
$date = date_create( $ts, timezone_open( 'UTC' ) );
date_timezone_set( $date, timezone_open( $data[2] ) );
$date = date_format( $date, 'YmdHis' );
return $date;
}
# Unrecognized timezone, default to 'Offset' with the stored offset.
$data[0] = 'Offset';
}
$minDiff = 0;
if ( $data[0] == 'System' || $tz == '' ) {
# Global offset in minutes.
if( isset($wgLocalTZoffset) ) $minDiff = $wgLocalTZoffset;
} else if ( $data[0] == 'Offset' ) {
$minDiff = intval( $data[1] );
} else {
$data = explode( ':', $tz );
if( count( $data ) == 2 ) {
$data[0] = intval( $data[0] );
$data[1] = intval( $data[1] );
$minDiff = abs( $data[0] ) * 60 + $data[1];
if ( $data[0] < 0 ) $minDiff = -$minDiff;
} else {
$minDiff = intval( $data[0] ) * 60;
}
}
# No difference ? Return time unchanged
if ( 0 == $minDiff ) return $ts;
wfSuppressWarnings(); // E_STRICT system time bitching
# Generate an adjusted date; take advantage of the fact that mktime
# will normalize out-of-range values so we don't have to split $minDiff
# into hours and minutes.
$t = mktime( (
(int)substr( $ts, 8, 2) ), # Hours
(int)substr( $ts, 10, 2 ) + $minDiff, # Minutes
(int)substr( $ts, 12, 2 ), # Seconds
(int)substr( $ts, 4, 2 ), # Month
(int)substr( $ts, 6, 2 ), # Day
(int)substr( $ts, 0, 4 ) ); #Year
$date = date( 'YmdHis', $t );
wfRestoreWarnings();
return $date;
}
/**
* This is a workalike of PHP's date() function, but with better
* internationalisation, a reduced set of format characters, and a better
* escaping format.
*
* Supported format characters are dDjlNwzWFmMntLoYyaAgGhHiscrU. See the
* PHP manual for definitions. "o" format character is supported since
* PHP 5.1.0, previous versions return literal o.
* There are a number of extensions, which start with "x":
*
* xn Do not translate digits of the next numeric format character
* xN Toggle raw digit (xn) flag, stays set until explicitly unset
* xr Use roman numerals for the next numeric format character
* xh Use hebrew numerals for the next numeric format character
* xx Literal x
* xg Genitive month name
*
* xij j (day number) in Iranian calendar
* xiF F (month name) in Iranian calendar
* xin n (month number) in Iranian calendar
* xiY Y (full year) in Iranian calendar
*
* xjj j (day number) in Hebrew calendar
* xjF F (month name) in Hebrew calendar
* xjt t (days in month) in Hebrew calendar
* xjx xg (genitive month name) in Hebrew calendar
* xjn n (month number) in Hebrew calendar
* xjY Y (full year) in Hebrew calendar
*
* xmj j (day number) in Hijri calendar
* xmF F (month name) in Hijri calendar
* xmn n (month number) in Hijri calendar
* xmY Y (full year) in Hijri calendar
*
* xkY Y (full year) in Thai solar calendar. Months and days are
* identical to the Gregorian calendar
* xoY Y (full year) in Minguo calendar or Juche year.
* Months and days are identical to the
* Gregorian calendar
* xtY Y (full year) in Japanese nengo. Months and days are
* identical to the Gregorian calendar
*
* Characters enclosed in double quotes will be considered literal (with
* the quotes themselves removed). Unmatched quotes will be considered
* literal quotes. Example:
*
* "The month is" F => The month is January
* i's" => 20'11"
*
* Backslash escaping is also supported.
*
* Input timestamp is assumed to be pre-normalized to the desired local
* time zone, if any.
*
* @param $format String
* @param $ts String: 14-character timestamp
* YYYYMMDDHHMMSS
* 01234567890123
* @todo emulation of "o" format character for PHP pre 5.1.0
* @todo handling of "o" format character for Iranian, Hebrew, Hijri & Thai?
*/
function sprintfDate( $format, $ts ) {
$s = '';
$raw = false;
$roman = false;
$hebrewNum = false;
$unix = false;
$rawToggle = false;
$iranian = false;
$hebrew = false;
$hijri = false;
$thai = false;
$minguo = false;
$tenno = false;
for ( $p = 0; $p < strlen( $format ); $p++ ) {
$num = false;
$code = $format[$p];
if ( $code == 'x' && $p < strlen( $format ) - 1 ) {
$code .= $format[++$p];
}
if ( ( $code === 'xi' || $code == 'xj' || $code == 'xk' || $code == 'xm' || $code == 'xo' || $code == 'xt' ) && $p < strlen( $format ) - 1 ) {
$code .= $format[++$p];
}
switch ( $code ) {
case 'xx':
$s .= 'x';
break;
case 'xn':
$raw = true;
break;
case 'xN':
$rawToggle = !$rawToggle;
break;
case 'xr':
$roman = true;
break;
case 'xh':
$hebrewNum = true;
break;
case 'xg':
$s .= $this->getMonthNameGen( substr( $ts, 4, 2 ) );
break;
case 'xjx':
if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts );
$s .= $this->getHebrewCalendarMonthNameGen( $hebrew[1] );
break;
case 'd':
$num = substr( $ts, 6, 2 );
break;
case 'D':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$s .= $this->getWeekdayAbbreviation( gmdate( 'w', $unix ) + 1 );
break;
case 'j':
$num = intval( substr( $ts, 6, 2 ) );
break;
case 'xij':
if ( !$iranian ) $iranian = self::tsToIranian( $ts );
$num = $iranian[2];
break;
case 'xmj':
if ( !$hijri ) $hijri = self::tsToHijri( $ts );
$num = $hijri[2];
break;
case 'xjj':
if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts );
$num = $hebrew[2];
break;
case 'l':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$s .= $this->getWeekdayName( gmdate( 'w', $unix ) + 1 );
break;
case 'N':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$w = gmdate( 'w', $unix );
$num = $w ? $w : 7;
break;
case 'w':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$num = gmdate( 'w', $unix );
break;
case 'z':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$num = gmdate( 'z', $unix );
break;
case 'W':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$num = gmdate( 'W', $unix );
break;
case 'F':
$s .= $this->getMonthName( substr( $ts, 4, 2 ) );
break;
case 'xiF':
if ( !$iranian ) $iranian = self::tsToIranian( $ts );
$s .= $this->getIranianCalendarMonthName( $iranian[1] );
break;
case 'xmF':
if ( !$hijri ) $hijri = self::tsToHijri( $ts );
$s .= $this->getHijriCalendarMonthName( $hijri[1] );
break;
case 'xjF':
if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts );
$s .= $this->getHebrewCalendarMonthName( $hebrew[1] );
break;
case 'm':
$num = substr( $ts, 4, 2 );
break;
case 'M':
$s .= $this->getMonthAbbreviation( substr( $ts, 4, 2 ) );
break;
case 'n':
$num = intval( substr( $ts, 4, 2 ) );
break;
case 'xin':
if ( !$iranian ) $iranian = self::tsToIranian( $ts );
$num = $iranian[1];
break;
case 'xmn':
if ( !$hijri ) $hijri = self::tsToHijri ( $ts );
$num = $hijri[1];
break;
case 'xjn':
if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts );
$num = $hebrew[1];
break;
case 't':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$num = gmdate( 't', $unix );
break;
case 'xjt':
if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts );
$num = $hebrew[3];
break;
case 'L':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$num = gmdate( 'L', $unix );
break;
# 'o' is supported since PHP 5.1.0
# return literal if not supported
# TODO: emulation for pre 5.1.0 versions
case 'o':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
if ( version_compare(PHP_VERSION, '5.1.0') === 1 )
$num = date( 'o', $unix );
else
$s .= 'o';
break;
case 'Y':
$num = substr( $ts, 0, 4 );
break;
case 'xiY':
if ( !$iranian ) $iranian = self::tsToIranian( $ts );
$num = $iranian[0];
break;
case 'xmY':
if ( !$hijri ) $hijri = self::tsToHijri( $ts );
$num = $hijri[0];
break;
case 'xjY':
if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts );
$num = $hebrew[0];
break;
case 'xkY':
if ( !$thai ) $thai = self::tsToYear( $ts, 'thai' );
$num = $thai[0];
break;
case 'xoY':
if ( !$minguo ) $minguo = self::tsToYear( $ts, 'minguo' );
$num = $minguo[0];
break;
case 'xtY':
if ( !$tenno ) $tenno = self::tsToYear( $ts, 'tenno' );
$num = $tenno[0];
break;
case 'y':
$num = substr( $ts, 2, 2 );
break;
case 'a':
$s .= intval( substr( $ts, 8, 2 ) ) < 12 ? 'am' : 'pm';
break;
case 'A':
$s .= intval( substr( $ts, 8, 2 ) ) < 12 ? 'AM' : 'PM';
break;
case 'g':
$h = substr( $ts, 8, 2 );
$num = $h % 12 ? $h % 12 : 12;
break;
case 'G':
$num = intval( substr( $ts, 8, 2 ) );
break;
case 'h':
$h = substr( $ts, 8, 2 );
$num = sprintf( '%02d', $h % 12 ? $h % 12 : 12 );
break;
case 'H':
$num = substr( $ts, 8, 2 );
break;
case 'i':
$num = substr( $ts, 10, 2 );
break;
case 's':
$num = substr( $ts, 12, 2 );
break;
case 'c':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$s .= gmdate( 'c', $unix );
break;
case 'r':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$s .= gmdate( 'r', $unix );
break;
case 'U':
if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts );
$num = $unix;
break;
case '\\':
# Backslash escaping
if ( $p < strlen( $format ) - 1 ) {
$s .= $format[++$p];
} else {
$s .= '\\';
}
break;
case '"':
# Quoted literal
if ( $p < strlen( $format ) - 1 ) {
$endQuote = strpos( $format, '"', $p + 1 );
if ( $endQuote === false ) {
# No terminating quote, assume literal "
$s .= '"';
} else {
$s .= substr( $format, $p + 1, $endQuote - $p - 1 );
$p = $endQuote;
}
} else {
# Quote at end of string, assume literal "
$s .= '"';
}
break;
default:
$s .= $format[$p];
}
if ( $num !== false ) {
if ( $rawToggle || $raw ) {
$s .= $num;
$raw = false;
} elseif ( $roman ) {
$s .= self::romanNumeral( $num );
$roman = false;
} elseif( $hebrewNum ) {
$s .= self::hebrewNumeral( $num );
$hebrewNum = false;
} else {
$s .= $this->formatNum( $num, true );
}
$num = false;
}
}
return $s;
}
private static $GREG_DAYS = array( 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31 );
private static $IRANIAN_DAYS = array( 31, 31, 31, 31, 31, 31, 30, 30, 30, 30, 30, 29 );
/**
* Algorithm by Roozbeh Pournader and Mohammad Toossi to convert
* Gregorian dates to Iranian dates. Originally written in C, it
* is released under the terms of GNU Lesser General Public
* License. Conversion to PHP was performed by Niklas Laxström.
*
* Link: http://www.farsiweb.info/jalali/jalali.c
*/
private static function tsToIranian( $ts ) {
$gy = substr( $ts, 0, 4 ) -1600;
$gm = substr( $ts, 4, 2 ) -1;
$gd = substr( $ts, 6, 2 ) -1;
# Days passed from the beginning (including leap years)
$gDayNo = 365*$gy
+ floor(($gy+3) / 4)
- floor(($gy+99) / 100)
+ floor(($gy+399) / 400);
// Add days of the past months of this year
for( $i = 0; $i < $gm; $i++ ) {
$gDayNo += self::$GREG_DAYS[$i];
}
// Leap years
if ( $gm > 1 && (($gy%4===0 && $gy%100!==0 || ($gy%400==0)))) {
$gDayNo++;
}
// Days passed in current month
$gDayNo += $gd;
$jDayNo = $gDayNo - 79;
$jNp = floor($jDayNo / 12053);
$jDayNo %= 12053;
$jy = 979 + 33*$jNp + 4*floor($jDayNo/1461);
$jDayNo %= 1461;
if ( $jDayNo >= 366 ) {
$jy += floor(($jDayNo-1)/365);
$jDayNo = floor(($jDayNo-1)%365);
}
for ( $i = 0; $i < 11 && $jDayNo >= self::$IRANIAN_DAYS[$i]; $i++ ) {
$jDayNo -= self::$IRANIAN_DAYS[$i];
}
$jm= $i+1;
$jd= $jDayNo+1;
return array($jy, $jm, $jd);
}
/**
* Converting Gregorian dates to Hijri dates.
*
* Based on a PHP-Nuke block by Sharjeel which is released under GNU/GPL license
*
* @link http://phpnuke.org/modules.php?name=News&file=article&sid=8234&mode=thread&order=0&thold=0
*/
private static function tsToHijri ( $ts ) {
$year = substr( $ts, 0, 4 );
$month = substr( $ts, 4, 2 );
$day = substr( $ts, 6, 2 );
$zyr = $year;
$zd=$day;
$zm=$month;
$zy=$zyr;
if (($zy>1582)||(($zy==1582)&&($zm>10))||(($zy==1582)&&($zm==10)&&($zd>14)))
{
$zjd=(int)((1461*($zy + 4800 + (int)( ($zm-14) /12) ))/4) + (int)((367*($zm-2-12*((int)(($zm-14)/12))))/12)-(int)((3*(int)(( ($zy+4900+(int)(($zm-14)/12))/100)))/4)+$zd-32075;
}
else
{
$zjd = 367*$zy-(int)((7*($zy+5001+(int)(($zm-9)/7)))/4)+(int)((275*$zm)/9)+$zd+1729777;
}
$zl=$zjd-1948440+10632;
$zn=(int)(($zl-1)/10631);
$zl=$zl-10631*$zn+354;
$zj=((int)((10985-$zl)/5316))*((int)((50*$zl)/17719))+((int)($zl/5670))*((int)((43*$zl)/15238));
$zl=$zl-((int)((30-$zj)/15))*((int)((17719*$zj)/50))-((int)($zj/16))*((int)((15238*$zj)/43))+29;
$zm=(int)((24*$zl)/709);
$zd=$zl-(int)((709*$zm)/24);
$zy=30*$zn+$zj-30;
return array ($zy, $zm, $zd);
}
/**
* Converting Gregorian dates to Hebrew dates.
*
* Based on a JavaScript code by Abu Mami and Yisrael Hersch
* (abu-mami@kaluach.net, http://www.kaluach.net), who permitted
* to translate the relevant functions into PHP and release them under
* GNU GPL.
*
* The months are counted from Tishrei = 1. In a leap year, Adar I is 13
* and Adar II is 14. In a non-leap year, Adar is 6.
*/
private static function tsToHebrew( $ts ) {
# Parse date
$year = substr( $ts, 0, 4 );
$month = substr( $ts, 4, 2 );
$day = substr( $ts, 6, 2 );
# Calculate Hebrew year
$hebrewYear = $year + 3760;
# Month number when September = 1, August = 12
$month += 4;
if( $month > 12 ) {
# Next year
$month -= 12;
$year++;
$hebrewYear++;
}
# Calculate day of year from 1 September
$dayOfYear = $day;
for( $i = 1; $i < $month; $i++ ) {
if( $i == 6 ) {
# February
$dayOfYear += 28;
# Check if the year is leap
if( $year % 400 == 0 || ( $year % 4 == 0 && $year % 100 > 0 ) ) {
$dayOfYear++;
}
} elseif( $i == 8 || $i == 10 || $i == 1 || $i == 3 ) {
$dayOfYear += 30;
} else {
$dayOfYear += 31;
}
}
# Calculate the start of the Hebrew year
$start = self::hebrewYearStart( $hebrewYear );
# Calculate next year's start
if( $dayOfYear <= $start ) {
# Day is before the start of the year - it is the previous year
# Next year's start
$nextStart = $start;
# Previous year
$year--;
$hebrewYear--;
# Add days since previous year's 1 September
$dayOfYear += 365;
if( ( $year % 400 == 0 ) || ( $year % 100 != 0 && $year % 4 == 0 ) ) {
# Leap year
$dayOfYear++;
}
# Start of the new (previous) year
$start = self::hebrewYearStart( $hebrewYear );
} else {
# Next year's start
$nextStart = self::hebrewYearStart( $hebrewYear + 1 );
}
# Calculate Hebrew day of year
$hebrewDayOfYear = $dayOfYear - $start;
# Difference between year's days
$diff = $nextStart - $start;
# Add 12 (or 13 for leap years) days to ignore the difference between
# Hebrew and Gregorian year (353 at least vs. 365/6) - now the
# difference is only about the year type
if( ( $year % 400 == 0 ) || ( $year % 100 != 0 && $year % 4 == 0 ) ) {
$diff += 13;
} else {
$diff += 12;
}
# Check the year pattern, and is leap year
# 0 means an incomplete year, 1 means a regular year, 2 means a complete year
# This is mod 30, to work on both leap years (which add 30 days of Adar I)
# and non-leap years
$yearPattern = $diff % 30;
# Check if leap year
$isLeap = $diff >= 30;
# Calculate day in the month from number of day in the Hebrew year
# Don't check Adar - if the day is not in Adar, we will stop before;
# if it is in Adar, we will use it to check if it is Adar I or Adar II
$hebrewDay = $hebrewDayOfYear;
$hebrewMonth = 1;
$days = 0;
while( $hebrewMonth <= 12 ) {
# Calculate days in this month
if( $isLeap && $hebrewMonth == 6 ) {
# Adar in a leap year
if( $isLeap ) {
# Leap year - has Adar I, with 30 days, and Adar II, with 29 days
$days = 30;
if( $hebrewDay <= $days ) {
# Day in Adar I
$hebrewMonth = 13;
} else {
# Subtract the days of Adar I
$hebrewDay -= $days;
# Try Adar II
$days = 29;
if( $hebrewDay <= $days ) {
# Day in Adar II
$hebrewMonth = 14;
}
}
}
} elseif( $hebrewMonth == 2 && $yearPattern == 2 ) {
# Cheshvan in a complete year (otherwise as the rule below)
$days = 30;
} elseif( $hebrewMonth == 3 && $yearPattern == 0 ) {
# Kislev in an incomplete year (otherwise as the rule below)
$days = 29;
} else {
# Odd months have 30 days, even have 29
$days = 30 - ( $hebrewMonth - 1 ) % 2;
}
if( $hebrewDay <= $days ) {
# In the current month
break;
} else {
# Subtract the days of the current month
$hebrewDay -= $days;
# Try in the next month
$hebrewMonth++;
}
}
return array( $hebrewYear, $hebrewMonth, $hebrewDay, $days );
}
/**
* This calculates the Hebrew year start, as days since 1 September.
* Based on Carl Friedrich Gauss algorithm for finding Easter date.
* Used for Hebrew date.
*/
private static function hebrewYearStart( $year ) {
$a = intval( ( 12 * ( $year - 1 ) + 17 ) % 19 );
$b = intval( ( $year - 1 ) % 4 );
$m = 32.044093161144 + 1.5542417966212 * $a + $b / 4.0 - 0.0031777940220923 * ( $year - 1 );
if( $m < 0 ) {
$m--;
}
$Mar = intval( $m );
if( $m < 0 ) {
$m++;
}
$m -= $Mar;
$c = intval( ( $Mar + 3 * ( $year - 1 ) + 5 * $b + 5 ) % 7);
if( $c == 0 && $a > 11 && $m >= 0.89772376543210 ) {
$Mar++;
} else if( $c == 1 && $a > 6 && $m >= 0.63287037037037 ) {
$Mar += 2;
} else if( $c == 2 || $c == 4 || $c == 6 ) {
$Mar++;
}
$Mar += intval( ( $year - 3761 ) / 100 ) - intval( ( $year - 3761 ) / 400 ) - 24;
return $Mar;
}
/**
* Algorithm to convert Gregorian dates to Thai solar dates,
* Minguo dates or Minguo dates.
*
* Link: http://en.wikipedia.org/wiki/Thai_solar_calendar
* http://en.wikipedia.org/wiki/Minguo_calendar
* http://en.wikipedia.org/wiki/Japanese_era_name
*
* @param $ts String: 14-character timestamp, calender name
* @return array converted year, month, day
*/
private static function tsToYear( $ts, $cName ) {
$gy = substr( $ts, 0, 4 );
$gm = substr( $ts, 4, 2 );
$gd = substr( $ts, 6, 2 );
if (!strcmp($cName,'thai')) {
# Thai solar dates
# Add 543 years to the Gregorian calendar
# Months and days are identical
$gy_offset = $gy + 543;
} else if ((!strcmp($cName,'minguo')) || !strcmp($cName,'juche')) {
# Minguo dates
# Deduct 1911 years from the Gregorian calendar
# Months and days are identical
$gy_offset = $gy - 1911;
} else if (!strcmp($cName,'tenno')) {
# Nengō dates up to Meiji period
# Deduct years from the Gregorian calendar
# depending on the nengo periods
# Months and days are identical
if (($gy < 1912) || (($gy == 1912) && ($gm < 7)) || (($gy == 1912) && ($gm == 7) && ($gd < 31))) {
# Meiji period
$gy_gannen = $gy - 1868 + 1;
$gy_offset = $gy_gannen;
if ($gy_gannen == 1)
$gy_offset = '元';
$gy_offset = '明治'.$gy_offset;
} else if ((($gy == 1912) && ($gm == 7) && ($gd == 31)) || (($gy == 1912) && ($gm >= 8)) || (($gy > 1912) && ($gy < 1926)) || (($gy == 1926) && ($gm < 12)) || (($gy == 1926) && ($gm == 12) && ($gd < 26))) {
# Taishō period
$gy_gannen = $gy - 1912 + 1;
$gy_offset = $gy_gannen;
if ($gy_gannen == 1)
$gy_offset = '元';
$gy_offset = '大正'.$gy_offset;
} else if ((($gy == 1926) && ($gm == 12) && ($gd >= 26)) || (($gy > 1926) && ($gy < 1989)) || (($gy == 1989) && ($gm == 1) && ($gd < 8))) {
# Shōwa period
$gy_gannen = $gy - 1926 + 1;
$gy_offset = $gy_gannen;
if ($gy_gannen == 1)
$gy_offset = '元';
$gy_offset = '昭和'.$gy_offset;
} else {
# Heisei period
$gy_gannen = $gy - 1989 + 1;
$gy_offset = $gy_gannen;
if ($gy_gannen == 1)
$gy_offset = '元';
$gy_offset = '平成'.$gy_offset;
}
} else {
$gy_offset = $gy;
}
return array( $gy_offset, $gm, $gd );
}
/**
* Roman number formatting up to 3000
*/
static function romanNumeral( $num ) {
static $table = array(
array( '', 'I', 'II', 'III', 'IV', 'V', 'VI', 'VII', 'VIII', 'IX', 'X' ),
array( '', 'X', 'XX', 'XXX', 'XL', 'L', 'LX', 'LXX', 'LXXX', 'XC', 'C' ),
array( '', 'C', 'CC', 'CCC', 'CD', 'D', 'DC', 'DCC', 'DCCC', 'CM', 'M' ),
array( '', 'M', 'MM', 'MMM' )
);
$num = intval( $num );
if ( $num > 3000 || $num <= 0 ) {
return $num;
}
$s = '';
for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
if ( $num >= $pow10 ) {
$s .= $table[$i][floor($num / $pow10)];
}
$num = $num % $pow10;
}
return $s;
}
/**
* Hebrew Gematria number formatting up to 9999
*/
static function hebrewNumeral( $num ) {
static $table = array(
array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' ),
array( '', 'י', 'כ', 'ל', 'מ', 'נ', 'ס', 'ע', 'פ', 'צ', 'ק' ),
array( '', 'ק', 'ר', 'ש', 'ת', 'תק', 'תר', 'תש', 'תת', 'תתק', 'תתר' ),
array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' )
);
$num = intval( $num );
if ( $num > 9999 || $num <= 0 ) {
return $num;
}
$s = '';
for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
if ( $num >= $pow10 ) {
if ( $num == 15 || $num == 16 ) {
$s .= $table[0][9] . $table[0][$num - 9];
$num = 0;
} else {
$s .= $table[$i][intval( ( $num / $pow10 ) )];
if( $pow10 == 1000 ) {
$s .= "'";
}
}
}
$num = $num % $pow10;
}
if( strlen( $s ) == 2 ) {
$str = $s . "'";
} else {
$str = substr( $s, 0, strlen( $s ) - 2 ) . '"';
$str .= substr( $s, strlen( $s ) - 2, 2 );
}
$start = substr( $str, 0, strlen( $str ) - 2 );
$end = substr( $str, strlen( $str ) - 2 );
switch( $end ) {
case 'כ':
$str = $start . 'ך';
break;
case 'מ':
$str = $start . 'ם';
break;
case 'נ':
$str = $start . 'ן';
break;
case 'פ':
$str = $start . 'ף';
break;
case 'צ':
$str = $start . 'ץ';
break;
}
return $str;
}
/**
* This is meant to be used by time(), date(), and timeanddate() to get
* the date preference they're supposed to use, it should be used in
* all children.
*
*<code>
* function timeanddate([...], $format = true) {
* $datePreference = $this->dateFormat($format);
* [...]
* }
*</code>
*
* @param $usePrefs Mixed: if true, the user's preference is used
* if false, the site/language default is used
* if int/string, assumed to be a format.
* @return string
*/
function dateFormat( $usePrefs = true ) {
global $wgUser;
if( is_bool( $usePrefs ) ) {
if( $usePrefs ) {
$datePreference = $wgUser->getDatePreference();
} else {
$options = User::getDefaultOptions();
$datePreference = (string)$options['date'];
}
} else {
$datePreference = (string)$usePrefs;
}
// return int
if( $datePreference == '' ) {
return 'default';
}
return $datePreference;
}
/**
* Get a format string for a given type and preference
* @param $type May be date, time or both
* @param $pref The format name as it appears in Messages*.php
*/
function getDateFormatString( $type, $pref ) {
if ( !isset( $this->dateFormatStrings[$type][$pref] ) ) {
if ( $pref == 'default' ) {
$pref = $this->getDefaultDateFormat();
$df = self::$dataCache->getSubitem( $this->mCode, 'dateFormats', "$pref $type" );
} else {
$df = self::$dataCache->getSubitem( $this->mCode, 'dateFormats', "$pref $type" );
if ( is_null( $df ) ) {
$pref = $this->getDefaultDateFormat();
$df = self::$dataCache->getSubitem( $this->mCode, 'dateFormats', "$pref $type" );
}
}
$this->dateFormatStrings[$type][$pref] = $df;
}
return $this->dateFormatStrings[$type][$pref];
}
/**
* @param $ts Mixed: the time format which needs to be turned into a
* date('YmdHis') format with wfTimestamp(TS_MW,$ts)
* @param $adj Bool: whether to adjust the time output according to the
* user configured offset ($timecorrection)
* @param $format Mixed: true to use user's date format preference
* @param $timecorrection String: the time offset as returned by
* validateTimeZone() in Special:Preferences
* @return string
*/
function date( $ts, $adj = false, $format = true, $timecorrection = false ) {
if ( $adj ) {
$ts = $this->userAdjust( $ts, $timecorrection );
}
$df = $this->getDateFormatString( 'date', $this->dateFormat( $format ) );
return $this->sprintfDate( $df, $ts );
}
/**
* @param $ts Mixed: the time format which needs to be turned into a
* date('YmdHis') format with wfTimestamp(TS_MW,$ts)
* @param $adj Bool: whether to adjust the time output according to the
* user configured offset ($timecorrection)
* @param $format Mixed: true to use user's date format preference
* @param $timecorrection String: the time offset as returned by
* validateTimeZone() in Special:Preferences
* @return string
*/
function time( $ts, $adj = false, $format = true, $timecorrection = false ) {
if ( $adj ) {
$ts = $this->userAdjust( $ts, $timecorrection );
}
$df = $this->getDateFormatString( 'time', $this->dateFormat( $format ) );
return $this->sprintfDate( $df, $ts );
}
/**
* @param $ts Mixed: the time format which needs to be turned into a
* date('YmdHis') format with wfTimestamp(TS_MW,$ts)
* @param $adj Bool: whether to adjust the time output according to the
* user configured offset ($timecorrection)
* @param $format Mixed: what format to return, if it's false output the
* default one (default true)
* @param $timecorrection String: the time offset as returned by
* validateTimeZone() in Special:Preferences
* @return string
*/
function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false) {
$ts = wfTimestamp( TS_MW, $ts );
if ( $adj ) {
$ts = $this->userAdjust( $ts, $timecorrection );
}
$df = $this->getDateFormatString( 'both', $this->dateFormat( $format ) );
return $this->sprintfDate( $df, $ts );
}
function getMessage( $key ) {
return self::$dataCache->getSubitem( $this->mCode, 'messages', $key );
}
function getAllMessages() {
return self::$dataCache->getItem( $this->mCode, 'messages' );
}
function iconv( $in, $out, $string ) {
# This is a wrapper for iconv in all languages except esperanto,
# which does some nasty x-conversions beforehand
# Even with //IGNORE iconv can whine about illegal characters in
# *input* string. We just ignore those too.
# REF: http://bugs.php.net/bug.php?id=37166
# REF: https://bugzilla.wikimedia.org/show_bug.cgi?id=16885
wfSuppressWarnings();
$text = iconv( $in, $out . '//IGNORE', $string );
wfRestoreWarnings();
return $text;
}
// callback functions for uc(), lc(), ucwords(), ucwordbreaks()
function ucwordbreaksCallbackAscii($matches){
return $this->ucfirst($matches[1]);
}
function ucwordbreaksCallbackMB($matches){
return mb_strtoupper($matches[0]);
}
function ucCallback($matches){
list( $wikiUpperChars ) = self::getCaseMaps();
return strtr( $matches[1], $wikiUpperChars );
}
function lcCallback($matches){
list( , $wikiLowerChars ) = self::getCaseMaps();
return strtr( $matches[1], $wikiLowerChars );
}
function ucwordsCallbackMB($matches){
return mb_strtoupper($matches[0]);
}
function ucwordsCallbackWiki($matches){
list( $wikiUpperChars ) = self::getCaseMaps();
return strtr( $matches[0], $wikiUpperChars );
}
function ucfirst( $str ) {
$o = ord( $str );
if ( $o < 96 ) {
return $str;
} elseif ( $o < 128 ) {
return ucfirst($str);
} else {
// fall back to more complex logic in case of multibyte strings
return self::uc($str,true);
}
}
function uc( $str, $first = false ) {
if ( function_exists( 'mb_strtoupper' ) ) {
if ( $first ) {
if ( self::isMultibyte( $str ) ) {
return mb_strtoupper( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
} else {
return ucfirst( $str );
}
} else {
return self::isMultibyte( $str ) ? mb_strtoupper( $str ) : strtoupper( $str );
}
} else {
if ( self::isMultibyte( $str ) ) {
list( $wikiUpperChars ) = $this->getCaseMaps();
$x = $first ? '^' : '';
return preg_replace_callback(
"/$x([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
array($this,"ucCallback"),
$str
);
} else {
return $first ? ucfirst( $str ) : strtoupper( $str );
}
}
}
function lcfirst( $str ) {
$o = ord( $str );
if ( !$o ) {
return strval( $str );
} elseif ( $o >= 128 ) {
return self::lc( $str, true );
} elseif ( $o > 96 ) {
return $str;
} else {
$str[0] = strtolower( $str[0] );
return $str;
}
}
function lc( $str, $first = false ) {
if ( function_exists( 'mb_strtolower' ) )
if ( $first )
if ( self::isMultibyte( $str ) )
return mb_strtolower( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
else
return strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 );
else
return self::isMultibyte( $str ) ? mb_strtolower( $str ) : strtolower( $str );
else
if ( self::isMultibyte( $str ) ) {
list( , $wikiLowerChars ) = self::getCaseMaps();
$x = $first ? '^' : '';
return preg_replace_callback(
"/$x([A-Z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
array($this,"lcCallback"),
$str
);
} else
return $first ? strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 ) : strtolower( $str );
}
function isMultibyte( $str ) {
return (bool)preg_match( '/[\x80-\xff]/', $str );
}
function ucwords($str) {
if ( self::isMultibyte( $str ) ) {
$str = self::lc($str);
// regexp to find first letter in each word (i.e. after each space)
$replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)| ([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
// function to use to capitalize a single char
if ( function_exists( 'mb_strtoupper' ) )
return preg_replace_callback(
$replaceRegexp,
array($this,"ucwordsCallbackMB"),
$str
);
else
return preg_replace_callback(
$replaceRegexp,
array($this,"ucwordsCallbackWiki"),
$str
);
}
else
return ucwords( strtolower( $str ) );
}
# capitalize words at word breaks
function ucwordbreaks($str){
if (self::isMultibyte( $str ) ) {
$str = self::lc($str);
// since \b doesn't work for UTF-8, we explicitely define word break chars
$breaks= "[ \-\(\)\}\{\.,\?!]";
// find first letter after word break
$replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)|$breaks([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
if ( function_exists( 'mb_strtoupper' ) )
return preg_replace_callback(
$replaceRegexp,
array($this,"ucwordbreaksCallbackMB"),
$str
);
else
return preg_replace_callback(
$replaceRegexp,
array($this,"ucwordsCallbackWiki"),
$str
);
}
else
return preg_replace_callback(
'/\b([\w\x80-\xff]+)\b/',
array($this,"ucwordbreaksCallbackAscii"),
$str );
}
/**
* Return a case-folded representation of $s
*
* This is a representation such that caseFold($s1)==caseFold($s2) if $s1
* and $s2 are the same except for the case of their characters. It is not
* necessary for the value returned to make sense when displayed.
*
* Do *not* perform any other normalisation in this function. If a caller
* uses this function when it should be using a more general normalisation
* function, then fix the caller.
*/
function caseFold( $s ) {
return $this->uc( $s );
}
function checkTitleEncoding( $s ) {
if( is_array( $s ) ) {
wfDebugDieBacktrace( 'Given array to checkTitleEncoding.' );
}
# Check for non-UTF-8 URLs
$ishigh = preg_match( '/[\x80-\xff]/', $s);
if(!$ishigh) return $s;
$isutf8 = preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
'[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s );
if( $isutf8 ) return $s;
return $this->iconv( $this->fallback8bitEncoding(), "utf-8", $s );
}
function fallback8bitEncoding() {
return self::$dataCache->getItem( $this->mCode, 'fallback8bitEncoding' );
}
/**
* Most writing systems use whitespace to break up words.
* Some languages such as Chinese don't conventionally do this,
* which requires special handling when breaking up words for
* searching etc.
*/
function hasWordBreaks() {
return true;
}
/**
* Some languages such as Chinese require word segmentation,
* Specify such segmentation when overridden in derived class.
*
* @param $string String
* @return String
*/
function wordSegmentation( $string ) {
return $string;
}
/**
* Some languages have special punctuation need to be normalized.
* Make such changes here.
*
* @param $string String
* @return String
*/
function normalizeForSearch( $string ) {
return $string;
}
/**
* convert double-width roman characters to single-width.
* range: ff00-ff5f ~= 0020-007f
*/
protected static function convertDoubleWidth( $string ) {
$string = preg_replace( '/\xef\xbc([\x80-\xbf])/e', 'chr((ord("$1") & 0x3f) + 0x20)', $string );
$string = preg_replace( '/\xef\xbd([\x80-\x99])/e', 'chr((ord("$1") & 0x3f) + 0x60)', $string );
return $string;
}
protected static function insertSpace( $string, $pattern ) {
$string = preg_replace( $pattern, " $1 ", $string );
$string = preg_replace( '/ +/', ' ', $string );
return $string;
}
function convertForSearchResult( $termsArray ) {
# some languages, e.g. Chinese, need to do a conversion
# in order for search results to be displayed correctly
return $termsArray;
}
/**
* Get the first character of a string.
*
* @param $s string
* @return string
*/
function firstChar( $s ) {
$matches = array();
preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
'[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/', $s, $matches);
if ( isset( $matches[1] ) ) {
if ( strlen( $matches[1] ) != 3 ) {
return $matches[1];
}
// Break down Hangul syllables to grab the first jamo
$code = utf8ToCodepoint( $matches[1] );
if ( $code < 0xac00 || 0xd7a4 <= $code) {
return $matches[1];
} elseif ( $code < 0xb098 ) {
return "\xe3\x84\xb1";
} elseif ( $code < 0xb2e4 ) {
return "\xe3\x84\xb4";
} elseif ( $code < 0xb77c ) {
return "\xe3\x84\xb7";
} elseif ( $code < 0xb9c8 ) {
return "\xe3\x84\xb9";
} elseif ( $code < 0xbc14 ) {
return "\xe3\x85\x81";
} elseif ( $code < 0xc0ac ) {
return "\xe3\x85\x82";
} elseif ( $code < 0xc544 ) {
return "\xe3\x85\x85";
} elseif ( $code < 0xc790 ) {
return "\xe3\x85\x87";
} elseif ( $code < 0xcc28 ) {
return "\xe3\x85\x88";
} elseif ( $code < 0xce74 ) {
return "\xe3\x85\x8a";
} elseif ( $code < 0xd0c0 ) {
return "\xe3\x85\x8b";
} elseif ( $code < 0xd30c ) {
return "\xe3\x85\x8c";
} elseif ( $code < 0xd558 ) {
return "\xe3\x85\x8d";
} else {
return "\xe3\x85\x8e";
}
} else {
return "";
}
}
function initEncoding() {
# Some languages may have an alternate char encoding option
# (Esperanto X-coding, Japanese furigana conversion, etc)
# If this language is used as the primary content language,
# an override to the defaults can be set here on startup.
}
function recodeForEdit( $s ) {
# For some languages we'll want to explicitly specify
# which characters make it into the edit box raw
# or are converted in some way or another.
# Note that if wgOutputEncoding is different from
# wgInputEncoding, this text will be further converted
# to wgOutputEncoding.
global $wgEditEncoding;
if( $wgEditEncoding == '' or
$wgEditEncoding == 'UTF-8' ) {
return $s;
} else {
return $this->iconv( 'UTF-8', $wgEditEncoding, $s );
}
}
function recodeInput( $s ) {
# Take the previous into account.
global $wgEditEncoding;
if($wgEditEncoding != "") {
$enc = $wgEditEncoding;
} else {
$enc = 'UTF-8';
}
if( $enc == 'UTF-8' ) {
return $s;
} else {
return $this->iconv( $enc, 'UTF-8', $s );
}
}
/**
* Convert a UTF-8 string to normal form C. In Malayalam and Arabic, this
* also cleans up certain backwards-compatible sequences, converting them
* to the modern Unicode equivalent.
*
* This is language-specific for performance reasons only.
*/
function normalize( $s ) {
return UtfNormal::cleanUp( $s );
}
/**
* Transform a string using serialized data stored in the given file (which
* must be in the serialized subdirectory of $IP). The file contains pairs
* mapping source characters to destination characters.
*
* The data is cached in process memory. This will go faster if you have the
* FastStringSearch extension.
*/
function transformUsingPairFile( $file, $string ) {
if ( !isset( $this->transformData[$file] ) ) {
$data = wfGetPrecompiledData( $file );
if ( $data === false ) {
throw new MWException( __METHOD__.": The transformation file $file is missing" );
}
$this->transformData[$file] = new ReplacementArray( $data );
}
return $this->transformData[$file]->replace( $string );
}
/**
* For right-to-left language support
*
* @return bool
*/
function isRTL() {
return self::$dataCache->getItem( $this->mCode, 'rtl' );
}
/**
* Return the correct HTML 'dir' attribute value for this language.
* @return String
*/
function getDir() {
return $this->isRTL() ? 'rtl' : 'ltr';
}
/**
* Return 'left' or 'right' as appropriate alignment for line-start
* for this language's text direction.
*
* Should be equivalent to CSS3 'start' text-align value....
*
* @return String
*/
function alignStart() {
return $this->isRTL() ? 'right' : 'left';
}
/**
* Return 'right' or 'left' as appropriate alignment for line-end
* for this language's text direction.
*
* Should be equivalent to CSS3 'end' text-align value....
*
* @return String
*/
function alignEnd() {
return $this->isRTL() ? 'left' : 'right';
}
/**
* A hidden direction mark (LRM or RLM), depending on the language direction
*
* @return string
*/
function getDirMark() {
return $this->isRTL() ? "\xE2\x80\x8F" : "\xE2\x80\x8E";
}
function capitalizeAllNouns() {
return self::$dataCache->getItem( $this->mCode, 'capitalizeAllNouns' );
}
/**
* An arrow, depending on the language direction
*
* @return string
*/
function getArrow() {
return $this->isRTL() ? '←' : '→';
}
/**
* To allow "foo[[bar]]" to extend the link over the whole word "foobar"
*
* @return bool
*/
function linkPrefixExtension() {
return self::$dataCache->getItem( $this->mCode, 'linkPrefixExtension' );
}
function getMagicWords() {
return self::$dataCache->getItem( $this->mCode, 'magicWords' );
}
# Fill a MagicWord object with data from here
function getMagic( $mw ) {
if ( !$this->mMagicHookDone ) {
$this->mMagicHookDone = true;
wfProfileIn( 'LanguageGetMagic' );
wfRunHooks( 'LanguageGetMagic', array( &$this->mMagicExtensions, $this->getCode() ) );
wfProfileOut( 'LanguageGetMagic' );
}
if ( isset( $this->mMagicExtensions[$mw->mId] ) ) {
$rawEntry = $this->mMagicExtensions[$mw->mId];
} else {
$magicWords = $this->getMagicWords();
if ( isset( $magicWords[$mw->mId] ) ) {
$rawEntry = $magicWords[$mw->mId];
} else {
$rawEntry = false;
}
}
if( !is_array( $rawEntry ) ) {
error_log( "\"$rawEntry\" is not a valid magic thingie for \"$mw->mId\"" );
} else {
$mw->mCaseSensitive = $rawEntry[0];
$mw->mSynonyms = array_slice( $rawEntry, 1 );
}
}
/**
* Add magic words to the extension array
*/
function addMagicWordsByLang( $newWords ) {
$code = $this->getCode();
$fallbackChain = array();
while ( $code && !in_array( $code, $fallbackChain ) ) {
$fallbackChain[] = $code;
$code = self::getFallbackFor( $code );
}
if ( !in_array( 'en', $fallbackChain ) ) {
$fallbackChain[] = 'en';
}
$fallbackChain = array_reverse( $fallbackChain );
foreach ( $fallbackChain as $code ) {
if ( isset( $newWords[$code] ) ) {
$this->mMagicExtensions = $newWords[$code] + $this->mMagicExtensions;
}
}
}
/**
* Get special page names, as an associative array
* case folded alias => real name
*/
function getSpecialPageAliases() {
// Cache aliases because it may be slow to load them
if ( is_null( $this->mExtendedSpecialPageAliases ) ) {
// Initialise array
$this->mExtendedSpecialPageAliases =
self::$dataCache->getItem( $this->mCode, 'specialPageAliases' );
wfRunHooks( 'LanguageGetSpecialPageAliases',
array( &$this->mExtendedSpecialPageAliases, $this->getCode() ) );
}
return $this->mExtendedSpecialPageAliases;
}
/**
* Italic is unsuitable for some languages
*
* @param $text String: the text to be emphasized.
* @return string
*/
function emphasize( $text ) {
return "<em>$text</em>";
}
/**
* Normally we output all numbers in plain en_US style, that is
* 293,291.235 for twohundredninetythreethousand-twohundredninetyone
* point twohundredthirtyfive. However this is not sutable for all
* languages, some such as Pakaran want ੨੯੩,੨੯੫.੨੩੫ and others such as
* Icelandic just want to use commas instead of dots, and dots instead
* of commas like "293.291,235".
*
* An example of this function being called:
* <code>
* wfMsg( 'message', $wgLang->formatNum( $num ) )
* </code>
*
* See LanguageGu.php for the Gujarati implementation and
* $separatorTransformTable on MessageIs.php for
* the , => . and . => , implementation.
*
* @todo check if it's viable to use localeconv() for the decimal
* separator thing.
* @param $number Mixed: the string to be formatted, should be an integer
* or a floating point number.
* @param $nocommafy Bool: set to true for special numbers like dates
* @return string
*/
function formatNum( $number, $nocommafy = false ) {
global $wgTranslateNumerals;
if (!$nocommafy) {
$number = $this->commafy($number);
$s = $this->separatorTransformTable();
if ($s) { $number = strtr($number, $s); }
}
if ($wgTranslateNumerals) {
$s = $this->digitTransformTable();
if ($s) { $number = strtr($number, $s); }
}
return $number;
}
function parseFormattedNumber( $number ) {
$s = $this->digitTransformTable();
if ($s) { $number = strtr($number, array_flip($s)); }
$s = $this->separatorTransformTable();
if ($s) { $number = strtr($number, array_flip($s)); }
$number = strtr( $number, array (',' => '') );
return $number;
}
/**
* Adds commas to a given number
*
* @param $_ mixed
* @return string
*/
function commafy($_) {
return strrev((string)preg_replace('/(\d{3})(?=\d)(?!\d*\.)/','$1,',strrev($_)));
}
function digitTransformTable() {
return self::$dataCache->getItem( $this->mCode, 'digitTransformTable' );
}
function separatorTransformTable() {
return self::$dataCache->getItem( $this->mCode, 'separatorTransformTable' );
}
/**
* Take a list of strings and build a locale-friendly comma-separated
* list, using the local comma-separator message.
* The last two strings are chained with an "and".
*
* @param $l Array
* @return string
*/
function listToText( $l ) {
$s = '';
$m = count( $l ) - 1;
if( $m == 1 ) {
return $l[0] . $this->getMessageFromDB( 'and' ) . $this->getMessageFromDB( 'word-separator' ) . $l[1];
}
else {
for ( $i = $m; $i >= 0; $i-- ) {
if ( $i == $m ) {
$s = $l[$i];
} else if( $i == $m - 1 ) {
$s = $l[$i] . $this->getMessageFromDB( 'and' ) . $this->getMessageFromDB( 'word-separator' ) . $s;
} else {
$s = $l[$i] . $this->getMessageFromDB( 'comma-separator' ) . $s;
}
}
return $s;
}
}
/**
* Take a list of strings and build a locale-friendly comma-separated
* list, using the local comma-separator message.
* @param $list array of strings to put in a comma list
* @return string
*/
function commaList( $list ) {
return implode(
$list,
wfMsgExt( 'comma-separator', array( 'parsemag', 'escapenoentities', 'language' => $this ) ) );
}
/**
* Take a list of strings and build a locale-friendly semicolon-separated
* list, using the local semicolon-separator message.
* @param $list array of strings to put in a semicolon list
* @return string
*/
function semicolonList( $list ) {
return implode(
$list,
wfMsgExt( 'semicolon-separator', array( 'parsemag', 'escapenoentities', 'language' => $this ) ) );
}
/**
* Same as commaList, but separate it with the pipe instead.
* @param $list array of strings to put in a pipe list
* @return string
*/
function pipeList( $list ) {
return implode(
$list,
wfMsgExt( 'pipe-separator', array( 'escapenoentities', 'language' => $this ) ) );
}
/**
* Truncate a string to a specified length in bytes, appending an optional
* string (e.g. for ellipses)
*
* The database offers limited byte lengths for some columns in the database;
* multi-byte character sets mean we need to ensure that only whole characters
* are included, otherwise broken characters can be passed to the user
*
* If $length is negative, the string will be truncated from the beginning
*
* @param $string String to truncate
* @param $length Int: maximum length (excluding ellipses)
* @param $ellipsis String to append to the truncated text
* @return string
*/
function truncate( $string, $length, $ellipsis = '...' ) {
# Use the localized ellipsis character
if( $ellipsis == '...' ) {
$ellipsis = wfMsgExt( 'ellipsis', array( 'escapenoentities', 'language' => $this ) );
}
if( $length == 0 ) {
return $ellipsis;
}
if ( strlen( $string ) <= abs( $length ) ) {
return $string;
}
$stringOriginal = $string;
if( $length > 0 ) {
$string = substr( $string, 0, $length );
$char = ord( $string[strlen( $string ) - 1] );
$m = array();
if ($char >= 0xc0) {
# We got the first byte only of a multibyte char; remove it.
$string = substr( $string, 0, -1 );
} elseif( $char >= 0x80 &&
preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
'[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) ) {
# We chopped in the middle of a character; remove it
$string = $m[1];
}
$string = $string . $ellipsis;
} else {
$string = substr( $string, $length );
$char = ord( $string[0] );
if( $char >= 0x80 && $char < 0xc0 ) {
# We chopped in the middle of a character; remove the whole thing
$string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
}
$string = $ellipsis . $string;
}
# Do not truncate if the ellipsis actually make the string longer. Bug 22181
if ( strlen( $string ) < strlen( $stringOriginal ) ) {
return $string;
} else {
return $stringOriginal;
}
}
/**
* Grammatical transformations, needed for inflected languages
* Invoked by putting {{grammar:case|word}} in a message
*
* @param $word string
* @param $case string
* @return string
*/
function convertGrammar( $word, $case ) {
global $wgGrammarForms;
if ( isset($wgGrammarForms[$this->getCode()][$case][$word]) ) {
return $wgGrammarForms[$this->getCode()][$case][$word];
}
return $word;
}
/**
* Provides an alternative text depending on specified gender.
* Usage {{gender:username|masculine|feminine|neutral}}.
* username is optional, in which case the gender of current user is used,
* but only in (some) interface messages; otherwise default gender is used.
* If second or third parameter are not specified, masculine is used.
* These details may be overriden per language.
*/
function gender( $gender, $forms ) {
if ( !count($forms) ) { return ''; }
$forms = $this->preConvertPlural( $forms, 2 );
if ( $gender === 'male' ) return $forms[0];
if ( $gender === 'female' ) return $forms[1];
return isset($forms[2]) ? $forms[2] : $forms[0];
}
/**
* Plural form transformations, needed for some languages.
* For example, there are 3 form of plural in Russian and Polish,
* depending on "count mod 10". See [[w:Plural]]
* For English it is pretty simple.
*
* Invoked by putting {{plural:count|wordform1|wordform2}}
* or {{plural:count|wordform1|wordform2|wordform3}}
*
* Example: {{plural:{{NUMBEROFARTICLES}}|article|articles}}
*
* @param $count Integer: non-localized number
* @param $forms Array: different plural forms
* @return string Correct form of plural for $count in this language
*/
function convertPlural( $count, $forms ) {
if ( !count($forms) ) { return ''; }
$forms = $this->preConvertPlural( $forms, 2 );
return ( $count == 1 ) ? $forms[0] : $forms[1];
}
/**
* Checks that convertPlural was given an array and pads it to requested
* amound of forms by copying the last one.
*
* @param $count Integer: How many forms should there be at least
* @param $forms Array of forms given to convertPlural
* @return array Padded array of forms or an exception if not an array
*/
protected function preConvertPlural( /* Array */ $forms, $count ) {
while ( count($forms) < $count ) {
$forms[] = $forms[count($forms)-1];
}
return $forms;
}
/**
* For translaing of expiry times
* @param $str String: the validated block time in English
* @return Somehow translated block time
* @see LanguageFi.php for example implementation
*/
function translateBlockExpiry( $str ) {
$scBlockExpiryOptions = $this->getMessageFromDB( 'ipboptions' );
if ( $scBlockExpiryOptions == '-') {
return $str;
}
foreach (explode(',', $scBlockExpiryOptions) as $option) {
if ( strpos($option, ":") === false )
continue;
list($show, $value) = explode(":", $option);
if ( strcmp ( $str, $value) == 0 ) {
return htmlspecialchars( trim( $show ) );
}
}
return $str;
}
/**
* languages like Chinese need to be segmented in order for the diff
* to be of any use
*
* @param $text String
* @return String
*/
function segmentForDiff( $text ) {
return $text;
}
/**
* and unsegment to show the result
*
* @param $text String
* @return String
*/
function unsegmentForDiff( $text ) {
return $text;
}
# convert text to all supported variants
function autoConvertToAllVariants($text) {
return $this->mConverter->autoConvertToAllVariants($text);
}
# convert text to different variants of a language.
function convert( $text ) {
return $this->mConverter->convert( $text );
}
# Convert a Title object to a string in the preferred variant
function convertTitle( $title ) {
return $this->mConverter->convertTitle( $title );
}
# Check if this is a language with variants
function hasVariants(){
return sizeof($this->getVariants())>1;
}
# Put custom tags (e.g. -{ }-) around math to prevent conversion
function armourMath($text){
return $this->mConverter->armourMath($text);
}
/**
* Perform output conversion on a string, and encode for safe HTML output.
* @param $text String text to be converted
* @param $isTitle Bool whether this conversion is for the article title
* @return string
* @todo this should get integrated somewhere sane
*/
function convertHtml( $text, $isTitle = false ) {
return htmlspecialchars( $this->convert( $text, $isTitle ) );
}
function convertCategoryKey( $key ) {
return $this->mConverter->convertCategoryKey( $key );
}
/**
* get the list of variants supported by this langauge
* see sample implementation in LanguageZh.php
*
* @return array an array of language codes
*/
function getVariants() {
return $this->mConverter->getVariants();
}
function getPreferredVariant( $fromUser = true, $fromHeader = false ) {
return $this->mConverter->getPreferredVariant( $fromUser, $fromHeader );
}
/**
* if a language supports multiple variants, it is
* possible that non-existing link in one variant
* actually exists in another variant. this function
* tries to find it. See e.g. LanguageZh.php
*
* @param $link String: the name of the link
* @param $nt Mixed: the title object of the link
* @param boolean $ignoreOtherCond: to disable other conditions when
* we need to transclude a template or update a category's link
* @return null the input parameters may be modified upon return
*/
function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
$this->mConverter->findVariantLink( $link, $nt, $ignoreOtherCond );
}
/**
* If a language supports multiple variants, converts text
* into an array of all possible variants of the text:
* 'variant' => text in that variant
*/
function convertLinkToAllVariants($text){
return $this->mConverter->convertLinkToAllVariants($text);
}
/**
* returns language specific options used by User::getPageRenderHash()
* for example, the preferred language variant
*
* @return string
*/
function getExtraHashOptions() {
return $this->mConverter->getExtraHashOptions();
}
/**
* for languages that support multiple variants, the title of an
* article may be displayed differently in different variants. this
* function returns the apporiate title defined in the body of the article.
*
* @return string
*/
function getParsedTitle() {
return $this->mConverter->getParsedTitle();
}
/**
* Enclose a string with the "no conversion" tag. This is used by
* various functions in the Parser
*
* @param $text String: text to be tagged for no conversion
* @param $noParse
* @return string the tagged text
*/
function markNoConversion( $text, $noParse=false ) {
return $this->mConverter->markNoConversion( $text, $noParse );
}
/**
* A regular expression to match legal word-trailing characters
* which should be merged onto a link of the form [[foo]]bar.
*
* @return string
*/
function linkTrail() {
return self::$dataCache->getItem( $this->mCode, 'linkTrail' );
}
function getLangObj() {
return $this;
}
/**
* Get the RFC 3066 code for this language object
*/
function getCode() {
return $this->mCode;
}
function setCode( $code ) {
$this->mCode = $code;
}
/**
* Get the name of a file for a certain language code
* @param $prefix string Prepend this to the filename
* @param $code string Language code
* @param $suffix string Append this to the filename
* @return string $prefix . $mangledCode . $suffix
*/
static function getFileName( $prefix = 'Language', $code, $suffix = '.php' ) {
// Protect against path traversal
if ( !Language::isValidCode( $code )
|| strcspn( $code, "/\\\000" ) !== strlen( $code ) )
{
throw new MWException( "Invalid language code \"$code\"" );
}
return $prefix . str_replace( '-', '_', ucfirst( $code ) ) . $suffix;
}
/**
* Get the language code from a file name. Inverse of getFileName()
* @param $filename string $prefix . $languageCode . $suffix
* @param $prefix string Prefix before the language code
* @param $suffix string Suffix after the language code
* @return Language code, or false if $prefix or $suffix isn't found
*/
static function getCodeFromFileName( $filename, $prefix = 'Language', $suffix = '.php' ) {
$m = null;
preg_match( '/' . preg_quote( $prefix, '/' ) . '([A-Z][a-z_]+)' .
preg_quote( $suffix, '/' ) . '/', $filename, $m );
if ( !count( $m ) ) {
return false;
}
return str_replace( '_', '-', strtolower( $m[1] ) );
}
static function getMessagesFileName( $code ) {
global $IP;
return self::getFileName( "$IP/languages/messages/Messages", $code, '.php' );
}
static function getClassFileName( $code ) {
global $IP;
return self::getFileName( "$IP/languages/classes/Language", $code, '.php' );
}
/**
* Get the fallback for a given language
*/
static function getFallbackFor( $code ) {
if ( $code === 'en' ) {
// Shortcut
return false;
} else {
return self::getLocalisationCache()->getItem( $code, 'fallback' );
}
}
/**
* Get all messages for a given language
* WARNING: this may take a long time
*/
static function getMessagesFor( $code ) {
return self::getLocalisationCache()->getItem( $code, 'messages' );
}
/**
* Get a message for a given language
*/
static function getMessageFor( $key, $code ) {
return self::getLocalisationCache()->getSubitem( $code, 'messages', $key );
}
function fixVariableInNamespace( $talk ) {
if ( strpos( $talk, '$1' ) === false ) return $talk;
global $wgMetaNamespace;
$talk = str_replace( '$1', $wgMetaNamespace, $talk );
# Allow grammar transformations
# Allowing full message-style parsing would make simple requests
# such as action=raw much more expensive than they need to be.
# This will hopefully cover most cases.
$talk = preg_replace_callback( '/{{grammar:(.*?)\|(.*?)}}/i',
array( &$this, 'replaceGrammarInNamespace' ), $talk );
return str_replace( ' ', '_', $talk );
}
function replaceGrammarInNamespace( $m ) {
return $this->convertGrammar( trim( $m[2] ), trim( $m[1] ) );
}
static function getCaseMaps() {
static $wikiUpperChars, $wikiLowerChars;
if ( isset( $wikiUpperChars ) ) {
return array( $wikiUpperChars, $wikiLowerChars );
}
wfProfileIn( __METHOD__ );
$arr = wfGetPrecompiledData( 'Utf8Case.ser' );
if ( $arr === false ) {
throw new MWException(
"Utf8Case.ser is missing, please run \"make\" in the serialized directory\n" );
}
extract( $arr );
wfProfileOut( __METHOD__ );
return array( $wikiUpperChars, $wikiLowerChars );
}
function formatTimePeriod( $seconds ) {
if ( $seconds < 10 ) {
return $this->formatNum( sprintf( "%.1f", $seconds ) ) . wfMsg( 'seconds-abbrev' );
} elseif ( $seconds < 60 ) {
return $this->formatNum( round( $seconds ) ) . wfMsg( 'seconds-abbrev' );
} elseif ( $seconds < 3600 ) {
return $this->formatNum( floor( $seconds / 60 ) ) . wfMsg( 'minutes-abbrev' ) .
$this->formatNum( round( fmod( $seconds, 60 ) ) ) . wfMsg( 'seconds-abbrev' );
} else {
$hours = floor( $seconds / 3600 );
$minutes = floor( ( $seconds - $hours * 3600 ) / 60 );
$secondsPart = round( $seconds - $hours * 3600 - $minutes * 60 );
return $this->formatNum( $hours ) . wfMsg( 'hours-abbrev' ) .
$this->formatNum( $minutes ) . wfMsg( 'minutes-abbrev' ) .
$this->formatNum( $secondsPart ) . wfMsg( 'seconds-abbrev' );
}
}
function formatBitrate( $bps ) {
$units = array( 'bps', 'kbps', 'Mbps', 'Gbps' );
if ( $bps <= 0 ) {
return $this->formatNum( $bps ) . $units[0];
}
$unitIndex = floor( log10( $bps ) / 3 );
$mantissa = $bps / pow( 1000, $unitIndex );
if ( $mantissa < 10 ) {
$mantissa = round( $mantissa, 1 );
} else {
$mantissa = round( $mantissa );
}
return $this->formatNum( $mantissa ) . $units[$unitIndex];
}
/**
* Format a size in bytes for output, using an appropriate
* unit (B, KB, MB or GB) according to the magnitude in question
*
* @param $size Size to format
* @return string Plain text (not HTML)
*/
function formatSize( $size ) {
// For small sizes no decimal places necessary
$round = 0;
if( $size > 1024 ) {
$size = $size / 1024;
if( $size > 1024 ) {
$size = $size / 1024;
// For MB and bigger two decimal places are smarter
$round = 2;
if( $size > 1024 ) {
$size = $size / 1024;
$msg = 'size-gigabytes';
} else {
$msg = 'size-megabytes';
}
} else {
$msg = 'size-kilobytes';
}
} else {
$msg = 'size-bytes';
}
$size = round( $size, $round );
$text = $this->getMessageFromDB( $msg );
return str_replace( '$1', $this->formatNum( $size ), $text );
}
/**
* Get the conversion rule title, if any.
*/
function getConvRuleTitle() {
return $this->mConverter->getConvRuleTitle();
}
}
else ()
<?php
/**
* Undelete a page by fetching it from the archive table
*
* @file
* @ingroup Maintenance
*/
require_once( dirname(__FILE__) . '/Maintenance.php' );
class Undelete extends Maintenance {
public function __construct() {
parent::__construct();
$this->mDescription = "Undelete a page";
$this->addOption( 'u', 'The user to perform the undeletion', false, true );
$this->addOption( 'r', 'The reason to undelete', false, true );
$this->addArg( 'pagename', 'Page to undelete' );
}
public function execute() {
global $wgUser;
$user = $this->getOption( 'u', 'Command line script' );
$reason = $this->getOption( 'r', '' );
$pageName = $this->getArg();
$title = Title::newFromText( $pageName );
if ( !$title ) {
$this->error( "Invalid title", true );
}
$wgUser = User::newFromName( $user );
$archive = new PageArchive( $title );
$this->output( "Undeleting " . $title->getPrefixedDBkey() . '...' );
$archive->undelete( array(), $reason );
$this->output( "done\n" );
}
}
$maintClass = "Undelete";
require_once( DO_MAINTENANCE );
?>
<?php
<?php
/**
* CheckUser extension - grants users with the appropriate permission the
* ability to check user's IP addresses and other information.
*
* @file
* @ingroup Extensions
* @version 2.3
* @author Tim Starling
* @author Aaron Schulz
* @license http://www.gnu.org/copyleft/gpl.html GNU General Public License 2.0 or later
* @link http://www.mediawiki.org/wiki/Extension:CheckUser Documentation
*/
# Not a valid entry point, skip unless MEDIAWIKI is defined
if ( !defined( 'MEDIAWIKI' ) ) {
echo "CheckUser extension";
exit( 1 );
}
# Internationalisation file
$dir = dirname( __FILE__ ) . '/';
$wgExtensionMessagesFiles['CheckUser'] = $dir . 'CheckUser.i18n.php';
$wgExtensionAliasesFiles['CheckUser'] = $dir . 'CheckUser.alias.php';
// Extension credits that will show up on Special:Version
$wgExtensionCredits['specialpage'][] = array(
'path' => __FILE__,
'author' => array( 'Tim Starling', 'Aaron Schulz' ),
'name' => 'CheckUser',
'version' => '2.3',
'url' => 'http://www.mediawiki.org/wiki/Extension:CheckUser',
'descriptionmsg' => 'checkuser-desc',
);
// New user rights
// 'checkuser' right is required to query IPs/users through Special:CheckUser
// 'checkuser-log' is required to view the private log of checkuser checks
$wgAvailableRights[] = 'checkuser';
$wgAvailableRights[] = 'checkuser-log';
$wgGroupPermissions['checkuser']['checkuser'] = true;
$wgGroupPermissions['checkuser']['checkuser-log'] = true;
// Legacy variable, no longer used. Used to point to a file in the server where
// CheckUser would log all queries done through Special:CheckUser.
// If this file exists, the installer will try to import data from this file to
// the 'cu_log' table in the database.
$wgCheckUserLog = '/home/wikipedia/logs/checkuser.log';
# How long to keep CU data?
$wgCUDMaxAge = 3 * 30 * 24 * 3600; // 3 months
# Mass block limits
$wgCheckUserMaxBlocks = 200;
// Set this to true if you want to force checkusers into giving a reason for
// each check they do through Special:CheckUser.
$wgCheckUserForceSummary = false;
// Increment this number every time you touch the .js file.
$wgCheckUserStyleVersion = 5;
# Recent changes data hook
$wgHooks['RecentChange_save'][] = 'efUpdateCheckUserData';
$wgHooks['EmailUser'][] = 'efUpdateCUEmailData';
$wgHooks['User::mailPasswordInternal'][] = 'efUpdateCUPasswordResetData';
$wgHooks['AuthPluginAutoCreate'][] = 'efUpdateAutoCreateData';
$wgHooks['ParserTestTables'][] = 'efCheckUserParserTestTables';
$wgHooks['LoadExtensionSchemaUpdates'][] = 'efCheckUserSchemaUpdates';
$wgHooks['ContributionsToolLinks'][] = 'efLoadCheckUserLink';
$wgResourceModules['ext.checkUser'] = array(
'scripts' => 'checkuser.js',
'dependencies' => array( 'mediawiki.legacy.block' ), // IP stuff
'localBasePath' => dirname( __FILE__ ),
'remoteExtPath' => 'CheckUser',
);
// TODO: move hooks to CheckUser.hooks.php
/**
* Hook function for RecentChange_save
* Saves user data into the cu_changes table
*/
function efUpdateCheckUserData( $rc ) {
// Extract params
extract( $rc->mAttribs );
// Get IP
$ip = wfGetIP();
// Get XFF header
$xff = wfGetForwardedFor();
list( $xff_ip, $trusted ) = efGetClientIPfromXFF( $xff );
// Our squid XFFs can flood this up sometimes
$isSquidOnly = efXFFChainIsSquid( $xff );
// Get agent
$agent = wfGetAgent();
// Store the log action text for log events
// $rc_comment should just be the log_comment
// BC: check if log_type and log_action exists
// If not, then $rc_comment is the actiontext and comment
if ( isset( $rc_log_type ) && $rc_type == RC_LOG ) {
$target = Title::makeTitle( $rc_namespace, $rc_title );
$actionText = LogPage::actionText( $rc_log_type, $rc_log_action, $target,
null, LogPage::extractParams( $rc_params )
);
} else {
$actionText = '';
}
$dbw = wfGetDB( DB_MASTER );
$cuc_id = $dbw->nextSequenceValue( 'cu_changes_cu_id_seq' );
$rcRow = array(
'cuc_id' => $cuc_id,
'cuc_namespace' => $rc_namespace,
'cuc_title' => $rc_title,
'cuc_minor' => $rc_minor,
'cuc_user' => $rc_user,
'cuc_user_text' => $rc_user_text,
'cuc_actiontext' => $actionText,
'cuc_comment' => $rc_comment,
'cuc_this_oldid' => $rc_this_oldid,
'cuc_last_oldid' => $rc_last_oldid,
'cuc_type' => $rc_type,
'cuc_timestamp' => $rc_timestamp,
'cuc_ip' => IP::sanitizeIP( $ip ),
'cuc_ip_hex' => $ip ? IP::toHex( $ip ) : null,
'cuc_xff' => !$isSquidOnly ? $xff : '',
'cuc_xff_hex' => ( $xff_ip && !$isSquidOnly ) ? IP::toHex( $xff_ip ) : null,
'cuc_agent' => $agent
);
# On PG, MW unsets cur_id due to schema incompatibilites. So it may not be set!
if ( isset( $rc_cur_id ) ) {
$rcRow['cuc_page_id'] = $rc_cur_id;
}
$dbw->insert( 'cu_changes', $rcRow, __METHOD__ );
# Every 100th edit, prune the checkuser changes table.
if ( 0 == mt_rand( 0, 99 ) ) {
# Periodically flush old entries from the recentchanges table.
global $wgCUDMaxAge;
$cutoff = $dbw->timestamp( time() - $wgCUDMaxAge );
$recentchanges = $dbw->tableName( 'cu_changes' );
$sql = "DELETE FROM $recentchanges WHERE cuc_timestamp < '{$cutoff}'";
$dbw->query( $sql, __METHOD__ );
}
return true;
}
/**
* Hook function to store password reset
* Saves user data into the cu_changes table
*/
function efUpdateCUPasswordResetData( $user, $ip, $account ) {
// Get XFF header
$xff = wfGetForwardedFor();
list( $xff_ip, $trusted ) = efGetClientIPfromXFF( $xff );
// Our squid XFFs can flood this up sometimes
$isSquidOnly = efXFFChainIsSquid( $xff );
// Get agent
$agent = wfGetAgent();
$dbw = wfGetDB( DB_MASTER );
$cuc_id = $dbw->nextSequenceValue( 'cu_changes_cu_id_seq' );
$rcRow = array(
'cuc_id' => $cuc_id,
'cuc_namespace' => NS_USER,
'cuc_title' => '',
'cuc_minor' => 0,
'cuc_user' => $user->getId(),
'cuc_user_text' => $user->getName(),
'cuc_actiontext' => wfMsgForContent( 'checkuser-reset-action', $account->getName() ),
'cuc_comment' => '',
'cuc_this_oldid' => 0,
'cuc_last_oldid' => 0,
'cuc_type' => RC_LOG,
'cuc_timestamp' => $dbw->timestamp( wfTimestampNow() ),
'cuc_ip' => IP::sanitizeIP( $ip ),
'cuc_ip_hex' => $ip ? IP::toHex( $ip ) : null,
'cuc_xff' => !$isSquidOnly ? $xff : '',
'cuc_xff_hex' => ( $xff_ip && !$isSquidOnly ) ? IP::toHex( $xff_ip ) : null,
'cuc_agent' => $agent
);
$dbw->insert( 'cu_changes', $rcRow, __METHOD__ );
return true;
}
/**
* Hook function to store email data
* Saves user data into the cu_changes table
*/
function efUpdateCUEmailData( $to, $from, $subject, $text ) {
global $wgSecretKey;
if ( !$wgSecretKey || $from->name == $to->name ) {
return true;
}
$userFrom = User::newFromName( $from->name );
$userTo = User::newFromName( $to->name );
$hash = md5( $userTo->getEmail() . $userTo->getId() . $wgSecretKey );
// Get IP
$ip = wfGetIP();
// Get XFF header
$xff = wfGetForwardedFor();
list( $xff_ip, $trusted ) = efGetClientIPfromXFF( $xff );
// Our squid XFFs can flood this up sometimes
$isSquidOnly = efXFFChainIsSquid( $xff );
// Get agent
$agent = wfGetAgent();
$dbw = wfGetDB( DB_MASTER );
$cuc_id = $dbw->nextSequenceValue( 'cu_changes_cu_id_seq' );
$rcRow = array(
'cuc_id' => $cuc_id,
'cuc_namespace' => NS_USER,
'cuc_title' => '',
'cuc_minor' => 0,
'cuc_user' => $userFrom->getId(),
'cuc_user_text' => $userFrom->getName(),
'cuc_actiontext' => wfMsgForContent( 'checkuser-email-action', $hash ),
'cuc_comment' => '',
'cuc_this_oldid' => 0,
'cuc_last_oldid' => 0,
'cuc_type' => RC_LOG,
'cuc_timestamp' => $dbw->timestamp( wfTimestampNow() ),
'cuc_ip' => IP::sanitizeIP( $ip ),
'cuc_ip_hex' => $ip ? IP::toHex( $ip ) : null,
'cuc_xff' => !$isSquidOnly ? $xff : '',
'cuc_xff_hex' => ( $xff_ip && !$isSquidOnly ) ? IP::toHex( $xff_ip ) : null,
'cuc_agent' => $agent
);
$dbw->insert( 'cu_changes', $rcRow, __METHOD__ );
return true;
}
/**
* Hook function to store autocreation data from the auth plugin
* Saves user data into the cu_changes table
*/
function efUpdateAutoCreateData( $user ) {
// Get IP
$ip = wfGetIP();
// Get XFF header
$xff = wfGetForwardedFor();
list( $xff_ip, $trusted ) = efGetClientIPfromXFF( $xff );
// Our squid XFFs can flood this up sometimes
$isSquidOnly = efXFFChainIsSquid( $xff );
// Get agent
$agent = wfGetAgent();
$dbw = wfGetDB( DB_MASTER );
$cuc_id = $dbw->nextSequenceValue( 'cu_changes_cu_id_seq' );
$rcRow = array(
'cuc_id' => $cuc_id,
'cuc_page_id' => 0,
'cuc_namespace' => NS_USER,
'cuc_title' => '',
'cuc_minor' => 0,
'cuc_user' => $user->getId(),
'cuc_user_text' => $user->getName(),
'cuc_actiontext' => wfMsgForContent( 'checkuser-autocreate-action' ),
'cuc_comment' => '',
'cuc_this_oldid' => 0,
'cuc_last_oldid' => 0,
'cuc_type' => RC_LOG,
'cuc_timestamp' => $dbw->timestamp( wfTimestampNow() ),
'cuc_ip' => IP::sanitizeIP( $ip ),
'cuc_ip_hex' => $ip ? IP::toHex( $ip ) : null,
'cuc_xff' => !$isSquidOnly ? $xff : '',
'cuc_xff_hex' => ( $xff_ip && !$isSquidOnly ) ? IP::toHex( $xff_ip ) : null,
'cuc_agent' => $agent
);
$dbw->insert( 'cu_changes', $rcRow, __METHOD__ );
return true;
}
/**
* Locates the client IP within a given XFF string
* @param string $xff
* @param string $address, the ip that sent this header (optional)
* @return array( string, bool )
*/
function efGetClientIPfromXFF( $xff, $address = null ) {
if ( !$xff ) {
return array( null, false );
}
// Avoid annoyingly long xff hacks
$xff = trim( substr( $xff, 0, 255 ) );
$client = null;
$trusted = true;
// Check each IP, assuming they are separated by commas
$ips = explode( ',', $xff );
foreach ( $ips as $ip ) {
$ip = trim( $ip );
// If it is a valid IP, not a hash or such
if ( IP::isIPAddress( $ip ) ) {
# The first IP should be the client.
# Start only from the first public IP.
if ( is_null( $client ) ) {
if ( IP::isPublic( $ip ) ) {
$client = $ip;
}
# Check that all servers are trusted
} elseif ( !wfIsTrustedProxy( $ip ) ) {
$trusted = false;
break;
}
}
}
// We still have to test if the IP that sent
// this header is trusted to confirm results
if ( $client != $address && ( !$address || !wfIsTrustedProxy( $address ) ) ) {
$trusted = false;
}
return array( $client, $trusted );
}
function efXFFChainIsSquid( $xff ) {
global $wgSquidServers, $wgSquidServersNoPurge;
if ( !$xff ) {
false;
}
// Avoid annoyingly long xff hacks
$xff = trim( substr( $xff, 0, 255 ) );
$squidOnly = true;
// Check each IP, assuming they are separated by commas
$ips = explode( ',', $xff );
foreach ( $ips as $n => $ip ) {
$ip = trim( $ip );
// If it is a valid IP, not a hash or such
if ( IP::isIPAddress( $ip ) ) {
if ( $n == 0 ) {
// The first IP should be the client...
} elseif ( !in_array( $ip, $wgSquidServers ) && !in_array( $ip, $wgSquidServersNoPurge ) ) {
$squidOnly = false;
break;
}
}
}
return $squidOnly;
}
function efCheckUserSchemaUpdates( $updater = null ) {
$base = dirname( __FILE__ );
if ( $updater === null ) {
global $wgDBtype, $wgExtNewIndexes;
efCheckUserCreateTables();
if ( $wgDBtype == 'mysql' ) {
$wgExtNewIndexes[] = array(
'cu_changes', 'cuc_ip_hex_time',
"$base/archives/patch-cu_changes_indexes.sql"
);
$wgExtNewIndexes[] = array(
'cu_changes', 'cuc_user_ip_time',
"$base/archives/patch-cu_changes_indexes2.sql"
);
}
} else {
$updater->addExtensionUpdate( array( 'efCheckUserCreateTables' ) );
if ( $updater->getDB()->getType() == 'mysql' ) {
$updater->addExtensionUpdate( array( 'addIndex', 'cu_changes',
'cuc_ip_hex_time', "$base/archives/patch-cu_changes_indexes.sql", true ) );
$updater->addExtensionUpdate( array( 'addIndex', 'cu_changes',
'cuc_user_ip_time', "$base/archives/patch-cu_changes_indexes2.sql", true ) );
}
}
return true;
}
function efCheckUserCreateTables( $updater = null ) {
if ( $updater === null ) {
$db = wfGetDB( DB_MASTER );
} else {
$db = $updater->getDB();
}
$base = dirname( __FILE__ );
if ( $db->tableExists( 'cu_changes' ) ) {
wfOut( "...cu_changes table already exists.\n" );
} else {
require_once "$base/install.inc";
create_cu_changes( $db );
}
if ( $db->tableExists( 'cu_log' ) ) {
wfOut( "...cu_log table already exists.\n" );
} else {
require_once "$base/install.inc";
create_cu_log( $db );
}
}
/**
* Tell the parser test engine to create a stub cu_changes table,
* or temporary pages won't save correctly during the test run.
*/
function efCheckUserParserTestTables( &$tables ) {
$tables[] = 'cu_changes';
return true;
}
// Set up the new special page
$wgSpecialPages['CheckUser'] = 'CheckUser';
$wgSpecialPageGroups['CheckUser'] = 'users';
$wgAutoloadClasses['CheckUser'] = dirname( __FILE__ ) . '/CheckUser_body.php';
/**
* Add a link to Special:CheckUser on Special:Contributions/<username> for
* privileged users.
* @param $id Integer: user ID
* @param $nt Title: user page title
* @param $links Array: tool links
* @return true
*/
function efLoadCheckUserLink( $id, $nt, &$links ) {
global $wgUser;
if ( $wgUser->isAllowed( 'checkuser' ) ) {
$links[] = $wgUser->getSkin()->makeKnownLinkObj(
SpecialPage::getTitleFor( 'CheckUser' ),
wfMsgHtml( 'checkuser-contribs' ),
'user=' . urlencode( $nt->getText() )
);
}
return true;
}
?>