LanguageConverter.php

Go to the documentation of this file.

<?php
use MediaWiki\MediaWikiServices;
 
use MediaWiki\Logger\LoggerFactory;
 
class LanguageConverter {
    static public $languagesWithVariants = [
        'gan',
        'iu',
        'kk',
        'ku',
        'shi',
        'sr',
        'tg',
        'uz',
        'zh',
    ];
 
    public $mMainLanguageCode;
    public $mVariants;
    public $mVariantFallbacks;
    public $mVariantNames;
    public $mTablesLoaded = false;
    public $mTables;
    // 'bidirectional' 'unidirectional' 'disable' for each variant
    public $mManualLevel;
 
    public $mCacheKey;
 
    public $mLangObj;
    public $mFlags;
    public $mDescCodeSep = ':', $mDescVarSep = ';';
    public $mUcfirst = false;
    public $mConvRuleTitle = false;
    public $mURLVariant;
    public $mUserVariant;
    public $mHeaderVariant;
    public $mMaxDepth = 10;
    public $mVarSeparatorPattern;
 
    const CACHE_VERSION_KEY = 'VERSION 7';
 
    public function __construct( $langobj, $maincode, $variants = [],
                                $variantfallbacks = [], $flags = [],
                                $manualLevel = [] ) {
        global $wgDisabledVariants;
        $this->mLangObj = $langobj;
        $this->mMainLanguageCode = $maincode;
        $this->mVariants = array_diff( $variants, $wgDisabledVariants );
        $this->mVariantFallbacks = $variantfallbacks;
        $this->mVariantNames = Language::fetchLanguageNames();
        $this->mCacheKey = wfMemcKey( 'conversiontables', $maincode );
        $defaultflags = [
            // 'S' show converted text
            // '+' add rules for alltext
            // 'E' the gave flags is error
            // these flags above are reserved for program
            'A' => 'A',   // add rule for convert code (all text convert)
            'T' => 'T',   // title convert
            'R' => 'R',   // raw content
            'D' => 'D',   // convert description (subclass implement)
            '-' => '-',   // remove convert (not implement)
            'H' => 'H',   // add rule for convert code (but no display in placed code)
            'N' => 'N'    // current variant name
        ];
        $this->mFlags = array_merge( $defaultflags, $flags );
        foreach ( $this->mVariants as $v ) {
            if ( array_key_exists( $v, $manualLevel ) ) {
                $this->mManualLevel[$v] = $manualLevel[$v];
            } else {
                $this->mManualLevel[$v] = 'bidirectional';
            }
            $this->mFlags[$v] = $v;
        }
    }
 
    public function getVariants() {
        return $this->mVariants;
    }
 
    public function getVariantFallbacks( $variant ) {
        if ( isset( $this->mVariantFallbacks[$variant] ) ) {
            return $this->mVariantFallbacks[$variant];
        }
        return $this->mMainLanguageCode;
    }
 
    public function getConvRuleTitle() {
        return $this->mConvRuleTitle;
    }
 
    public function getPreferredVariant() {
        global $wgDefaultLanguageVariant, $wgUser;
 
        $req = $this->getURLVariant();
 
        if ( $wgUser->isSafeToLoad() && $wgUser->isLoggedIn() && !$req ) {
            $req = $this->getUserVariant();
        } elseif ( !$req ) {
            $req = $this->getHeaderVariant();
        }
 
        if ( $wgDefaultLanguageVariant && !$req ) {
            $req = $this->validateVariant( $wgDefaultLanguageVariant );
        }
 
        // This function, unlike the other get*Variant functions, is
        // not memoized (i.e. there return value is not cached) since
        // new information might appear during processing after this
        // is first called.
        if ( $this->validateVariant( $req ) ) {
            return $req;
        }
        return $this->mMainLanguageCode;
    }
 
    public function getDefaultVariant() {
        global $wgDefaultLanguageVariant;
 
        $req = $this->getURLVariant();
 
        if ( !$req ) {
            $req = $this->getHeaderVariant();
        }
 
        if ( $wgDefaultLanguageVariant && !$req ) {
            $req = $this->validateVariant( $wgDefaultLanguageVariant );
        }
 
        if ( $req ) {
            return $req;
        }
        return $this->mMainLanguageCode;
    }
 
    public function validateVariant( $variant = null ) {
        if ( $variant !== null && in_array( $variant, $this->mVariants ) ) {
            return $variant;
        }
        return null;
    }
 
    public function getURLVariant() {
        global $wgRequest;
 
        if ( $this->mURLVariant ) {
            return $this->mURLVariant;
        }
 
        // see if the preference is set in the request
        $ret = $wgRequest->getText( 'variant' );
 
        if ( !$ret ) {
            $ret = $wgRequest->getVal( 'uselang' );
        }
 
        $this->mURLVariant = $this->validateVariant( $ret );
        return $this->mURLVariant;
    }
 
    protected function getUserVariant() {
        global $wgUser, $wgContLang;
 
        // memoizing this function wreaks havoc on parserTest.php
        /*
        if ( $this->mUserVariant ) {
            return $this->mUserVariant;
        }
        */
 
        // Get language variant preference from logged in users
        // Don't call this on stub objects because that causes infinite
        // recursion during initialisation
        if ( !$wgUser->isSafeToLoad() ) {
            return false;
        }
        if ( $wgUser->isLoggedIn() ) {
            if ( $this->mMainLanguageCode == $wgContLang->getCode() ) {
                $ret = $wgUser->getOption( 'variant' );
            } else {
                $ret = $wgUser->getOption( 'variant-' . $this->mMainLanguageCode );
            }
        } else {
            // figure out user lang without constructing wgLang to avoid
            // infinite recursion
            $ret = $wgUser->getOption( 'language' );
        }
 
        $this->mUserVariant = $this->validateVariant( $ret );
        return $this->mUserVariant;
    }
 
    protected function getHeaderVariant() {
        global $wgRequest;
 
        if ( $this->mHeaderVariant ) {
            return $this->mHeaderVariant;
        }
 
        // see if some supported language variant is set in the
        // HTTP header.
        $languages = array_keys( $wgRequest->getAcceptLang() );
        if ( empty( $languages ) ) {
            return null;
        }
 
        $fallbackLanguages = [];
        foreach ( $languages as $language ) {
            $this->mHeaderVariant = $this->validateVariant( $language );
            if ( $this->mHeaderVariant ) {
                break;
            }
 
            // To see if there are fallbacks of current language.
            // We record these fallback variants, and process
            // them later.
            $fallbacks = $this->getVariantFallbacks( $language );
            if ( is_string( $fallbacks ) && $fallbacks !== $this->mMainLanguageCode ) {
                $fallbackLanguages[] = $fallbacks;
            } elseif ( is_array( $fallbacks ) ) {
                $fallbackLanguages =
                    array_merge( $fallbackLanguages, $fallbacks );
            }
        }
 
        if ( !$this->mHeaderVariant ) {
            // process fallback languages now
            $fallback_languages = array_unique( $fallbackLanguages );
            foreach ( $fallback_languages as $language ) {
                $this->mHeaderVariant = $this->validateVariant( $language );
                if ( $this->mHeaderVariant ) {
                    break;
                }
            }
        }
 
        return $this->mHeaderVariant;
    }
 
    public function autoConvert( $text, $toVariant = false ) {
 
        $this->loadTables();
 
        if ( !$toVariant ) {
            $toVariant = $this->getPreferredVariant();
            if ( !$toVariant ) {
                return $text;
            }
        }
 
        if ( $this->guessVariant( $text, $toVariant ) ) {
            return $text;
        }
        /* we convert everything except:
           1. HTML markups (anything between < and >)
           2. HTML entities
           3. placeholders created by the parser
        */
        $marker = '|' . Parser::MARKER_PREFIX . '[^\x7f]++\x7f';
 
        // this one is needed when the text is inside an HTML markup
        $htmlfix = '|<[^>\004]++(?=\004$)|^[^<>]*+>';
 
        // Optimize for the common case where these tags have
        // few or no children. Thus try and possesively get as much as
        // possible, and only engage in backtracking when we hit a '<'.
 
        // disable convert to variants between <code> tags
        $codefix = '<code>[^<]*+(?:(?:(?!<\/code>).)[^<]*+)*+<\/code>|';
        // disable conversion of <script> tags
        $scriptfix = '<script[^>]*+>[^<]*+(?:(?:(?!<\/script>).)[^<]*+)*+<\/script>|';
        // disable conversion of <pre> tags
        $prefix = '<pre[^>]*+>[^<]*+(?:(?:(?!<\/pre>).)[^<]*+)*+<\/pre>|';
        // The "|.*+)" at the end, is in case we missed some part of html syntax,
        // we will fail securely (hopefully) by matching the rest of the string.
        $htmlFullTag = '<(?:[^>=]*+(?>[^>=]*+=\s*+(?:"[^"]*"|\'[^\']*\'|[^\'">\s]*+))*+[^>=]*+>|.*+)|';
 
        $reg = '/' . $codefix . $scriptfix . $prefix . $htmlFullTag .
            '&[a-zA-Z#][a-z0-9]++;' . $marker . $htmlfix . '|\004$/s';
        $startPos = 0;
        $sourceBlob = '';
        $literalBlob = '';
 
        // Guard against delimiter nulls in the input
        $text = str_replace( "\000", '', $text );
        $text = str_replace( "\004", '', $text );
 
        $markupMatches = null;
        $elementMatches = null;
 
        // We add a marker (\004) at the end of text, to ensure we always match the
        // entire text (Otherwise, pcre.backtrack_limit might cause silent failure)
        while ( $startPos < strlen( $text ) ) {
            if ( preg_match( $reg, $text . "\004", $markupMatches, PREG_OFFSET_CAPTURE, $startPos ) ) {
                $elementPos = $markupMatches[0][1];
                $element = $markupMatches[0][0];
                if ( $element === "\004" ) {
                    // We hit the end.
                    $elementPos = strlen( $text );
                    $element = '';
                } elseif ( substr( $element, -1 ) === "\004" ) {
                    // This can sometimes happen if we have
                    // unclosed html tags (For example
                    // when converting a title attribute
                    // during a recursive call that contains
                    // a &lt; e.g. <div title="&lt;">.
                    $element = substr( $element, 0, -1 );
                }
            } else {
                // If we hit here, then Language Converter could be tricked
                // into doing an XSS, so we refuse to translate.
                // If non-crazy input manages to reach this code path,
                // we should consider it a bug.
                $log = LoggerFactory::getInstance( 'languageconverter' );
                $log->error( "Hit pcre.backtrack_limit in " . __METHOD__
                    . ". Disabling language conversion for this page.",
                    [
                        "method" => __METHOD__,
                        "variant" => $toVariant,
                        "startOfText" => substr( $text, 0, 500 )
                    ]
                );
                return $text;
            }
            // Queue the part before the markup for translation in a batch
            $sourceBlob .= substr( $text, $startPos, $elementPos - $startPos ) . "\000";
 
            // Advance to the next position
            $startPos = $elementPos + strlen( $element );
 
            // Translate any alt or title attributes inside the matched element
            if ( $element !== ''
                && preg_match( '/^(<[^>\s]*+)\s([^>]*+)(.*+)$/', $element, $elementMatches )
            ) {
                // FIXME, this decodes entities, so if you have something
                // like <div title="foo&lt;bar"> the bar won't get
                // translated since after entity decoding it looks like
                // unclosed html and we call this method recursively
                // on attributes.
                $attrs = Sanitizer::decodeTagAttributes( $elementMatches[2] );
                // Ensure self-closing tags stay self-closing.
                $close = substr( $elementMatches[2], -1 ) === '/' ? ' /' : '';
                $changed = false;
                foreach ( [ 'title', 'alt' ] as $attrName ) {
                    if ( !isset( $attrs[$attrName] ) ) {
                        continue;
                    }
                    $attr = $attrs[$attrName];
                    // Don't convert URLs
                    if ( !strpos( $attr, '://' ) ) {
                        $attr = $this->recursiveConvertTopLevel( $attr, $toVariant );
                    }
 
                    // Remove HTML tags to avoid disrupting the layout
                    $attr = preg_replace( '/<[^>]+>/', '', $attr );
                    if ( $attr !== $attrs[$attrName] ) {
                        $attrs[$attrName] = $attr;
                        $changed = true;
                    }
                }
                if ( $changed ) {
                    $element = $elementMatches[1] . Html::expandAttributes( $attrs ) .
                        $close . $elementMatches[3];
                }
            }
            $literalBlob .= $element . "\000";
        }
 
        // Do the main translation batch
        $translatedBlob = $this->translate( $sourceBlob, $toVariant );
 
        // Put the output back together
        $translatedIter = StringUtils::explode( "\000", $translatedBlob );
        $literalIter = StringUtils::explode( "\000", $literalBlob );
        $output = '';
        while ( $translatedIter->valid() && $literalIter->valid() ) {
            $output .= $translatedIter->current();
            $output .= $literalIter->current();
            $translatedIter->next();
            $literalIter->next();
        }
 
        return $output;
    }
 
    public function translate( $text, $variant ) {
        // If $text is empty or only includes spaces, do nothing
        // Otherwise translate it
        if ( trim( $text ) ) {
            $this->loadTables();
            $text = $this->mTables[$variant]->replace( $text );
        }
        return $text;
    }
 
    public function autoConvertToAllVariants( $text ) {
        $this->loadTables();
 
        $ret = [];
        foreach ( $this->mVariants as $variant ) {
            $ret[$variant] = $this->translate( $text, $variant );
        }
 
        return $ret;
    }
 
    protected function applyManualConv( $convRule ) {
        // Use syntax -{T|zh-cn:TitleCN; zh-tw:TitleTw}- to custom
        // title conversion.
        // Bug 24072: $mConvRuleTitle was overwritten by other manual
        // rule(s) not for title, this breaks the title conversion.
        $newConvRuleTitle = $convRule->getTitle();
        if ( $newConvRuleTitle ) {
            // So I add an empty check for getTitle()
            $this->mConvRuleTitle = $newConvRuleTitle;
        }
 
        // merge/remove manual conversion rules to/from global table
        $convTable = $convRule->getConvTable();
        $action = $convRule->getRulesAction();
        foreach ( $convTable as $variant => $pair ) {
            if ( !$this->validateVariant( $variant ) ) {
                continue;
            }
 
            if ( $action == 'add' ) {
                // More efficient than array_merge(), about 2.5 times.
                foreach ( $pair as $from => $to ) {
                    $this->mTables[$variant]->setPair( $from, $to );
                }
            } elseif ( $action == 'remove' ) {
                $this->mTables[$variant]->removeArray( $pair );
            }
        }
    }
 
    public function convertTitle( $title ) {
        $variant = $this->getPreferredVariant();
        $index = $title->getNamespace();
        if ( $index !== NS_MAIN ) {
            $text = $this->convertNamespace( $index, $variant ) . ':';
        } else {
            $text = '';
        }
        $text .= $this->translate( $title->getText(), $variant );
        return $text;
    }
 
    public function convertNamespace( $index, $variant = null ) {
        if ( $index === NS_MAIN ) {
            return '';
        }
 
        if ( $variant === null ) {
            $variant = $this->getPreferredVariant();
        }
 
        $cache = MediaWikiServices::getInstance()->getLocalServerObjectCache();
        $key = $cache->makeKey( 'languageconverter', 'namespace-text', $index, $variant );
        $nsVariantText = $cache->get( $key );
        if ( $nsVariantText !== false ) {
            return $nsVariantText;
        }
 
        // First check if a message gives a converted name in the target variant.
        $nsConvMsg = wfMessage( 'conversion-ns' . $index )->inLanguage( $variant );
        if ( $nsConvMsg->exists() ) {
            $nsVariantText = $nsConvMsg->plain();
        }
 
        // Then check if a message gives a converted name in content language
        // which needs extra translation to the target variant.
        if ( $nsVariantText === false ) {
            $nsConvMsg = wfMessage( 'conversion-ns' . $index )->inContentLanguage();
            if ( $nsConvMsg->exists() ) {
                $nsVariantText = $this->translate( $nsConvMsg->plain(), $variant );
            }
        }
 
        if ( $nsVariantText === false ) {
            // No message exists, retrieve it from the target variant's namespace names.
            $langObj = $this->mLangObj->factory( $variant );
            $nsVariantText = $langObj->getFormattedNsText( $index );
        }
 
        $cache->set( $key, $nsVariantText, 60 );
 
        return $nsVariantText;
    }
 
    public function convert( $text ) {
        $variant = $this->getPreferredVariant();
        return $this->convertTo( $text, $variant );
    }
 
    public function convertTo( $text, $variant ) {
        global $wgDisableLangConversion;
        if ( $wgDisableLangConversion ) {
            return $text;
        }
        // Reset converter state for a new converter run.
        $this->mConvRuleTitle = false;
        return $this->recursiveConvertTopLevel( $text, $variant );
    }
 
    protected function recursiveConvertTopLevel( $text, $variant, $depth = 0 ) {
        $startPos = 0;
        $out = '';
        $length = strlen( $text );
        $shouldConvert = !$this->guessVariant( $text, $variant );
        $continue = 1;
 
        $noScript = '<script.*?>.*?<\/script>(*SKIP)(*FAIL)';
        $noStyle = '<style.*?>.*?<\/style>(*SKIP)(*FAIL)';
        // @codingStandardsIgnoreStart Generic.Files.LineLength.TooLong
        $noHtml = '<(?:[^>=]*+(?>[^>=]*+=\s*+(?:"[^"]*"|\'[^\']*\'|[^\'">\s]*+))*+[^>=]*+>|.*+)(*SKIP)(*FAIL)';
        // @codingStandardsIgnoreEnd
        while ( $startPos < $length && $continue ) {
            $continue = preg_match(
                // Only match -{ outside of html.
                "/$noScript|$noStyle|$noHtml|-\{/",
                $text,
                $m,
                PREG_OFFSET_CAPTURE,
                $startPos
            );
 
            if ( !$continue ) {
                // No more markup, append final segment
                $fragment = substr( $text, $startPos );
                $out .= $shouldConvert ? $this->autoConvert( $fragment, $variant ) : $fragment;
                return $out;
            }
 
            // Offset of the match of the regex pattern.
            $pos = $m[0][1];
 
            // Append initial segment
            $fragment = substr( $text, $startPos, $pos - $startPos );
            $out .= $shouldConvert ? $this->autoConvert( $fragment, $variant ) : $fragment;
            // -{ marker found, not in attribute
            // Advance position up to -{ marker.
            $startPos = $pos;
            // Do recursive conversion
            // Note: This passes $startPos by reference, and advances it.
            $out .= $this->recursiveConvertRule( $text, $variant, $startPos, $depth + 1 );
        }
        return $out;
    }
 
    protected function recursiveConvertRule( $text, $variant, &$startPos, $depth = 0 ) {
        // Quick sanity check (no function calls)
        if ( $text[$startPos] !== '-' || $text[$startPos + 1] !== '{' ) {
            throw new MWException( __METHOD__ . ': invalid input string' );
        }
 
        $startPos += 2;
        $inner = '';
        $warningDone = false;
        $length = strlen( $text );
 
        while ( $startPos < $length ) {
            $m = false;
            preg_match( '/-\{|\}-/', $text, $m, PREG_OFFSET_CAPTURE, $startPos );
            if ( !$m ) {
                // Unclosed rule
                break;
            }
 
            $token = $m[0][0];
            $pos = $m[0][1];
 
            // Markup found
            // Append initial segment
            $inner .= substr( $text, $startPos, $pos - $startPos );
 
            // Advance position
            $startPos = $pos;
 
            switch ( $token ) {
                case '-{':
                    // Check max depth
                    if ( $depth >= $this->mMaxDepth ) {
                        $inner .= '-{';
                        if ( !$warningDone ) {
                            $inner .= '<span class="error">' .
                                wfMessage( 'language-converter-depth-warning' )
                                    ->numParams( $this->mMaxDepth )->inContentLanguage()->text() .
                                '</span>';
                            $warningDone = true;
                        }
                        $startPos += 2;
                        continue;
                    }
                    // Recursively parse another rule
                    $inner .= $this->recursiveConvertRule( $text, $variant, $startPos, $depth + 1 );
                    break;
                case '}-':
                    // Apply the rule
                    $startPos += 2;
                    $rule = new ConverterRule( $inner, $this );
                    $rule->parse( $variant );
                    $this->applyManualConv( $rule );
                    return $rule->getDisplay();
                default:
                    throw new MWException( __METHOD__ . ': invalid regex match' );
            }
        }
 
        // Unclosed rule
        if ( $startPos < $length ) {
            $inner .= substr( $text, $startPos );
        }
        $startPos = $length;
        return '-{' . $this->autoConvert( $inner, $variant );
    }
 
    public function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
        # If the article has already existed, there is no need to
        # check it again, otherwise it may cause a fault.
        if ( is_object( $nt ) && $nt->exists() ) {
            return;
        }
 
        global $wgDisableLangConversion, $wgDisableTitleConversion, $wgRequest;
        $isredir = $wgRequest->getText( 'redirect', 'yes' );
        $action = $wgRequest->getText( 'action' );
        if ( $action == 'edit' && $wgRequest->getBool( 'redlink' ) ) {
            $action = 'view';
        }
        $linkconvert = $wgRequest->getText( 'linkconvert', 'yes' );
        $disableLinkConversion = $wgDisableLangConversion
            || $wgDisableTitleConversion;
        $linkBatch = new LinkBatch();
 
        $ns = NS_MAIN;
 
        if ( $disableLinkConversion ||
            ( !$ignoreOtherCond &&
                ( $isredir == 'no'
                    || $action == 'edit'
                    || $action == 'submit'
                    || $linkconvert == 'no' ) ) ) {
            return;
        }
 
        if ( is_object( $nt ) ) {
            $ns = $nt->getNamespace();
        }
 
        $variants = $this->autoConvertToAllVariants( $link );
        if ( !$variants ) { // give up
            return;
        }
 
        $titles = [];
 
        foreach ( $variants as $v ) {
            if ( $v != $link ) {
                $varnt = Title::newFromText( $v, $ns );
                if ( !is_null( $varnt ) ) {
                    $linkBatch->addObj( $varnt );
                    $titles[] = $varnt;
                }
            }
        }
 
        // fetch all variants in single query
        $linkBatch->execute();
 
        foreach ( $titles as $varnt ) {
            if ( $varnt->getArticleID() > 0 ) {
                $nt = $varnt;
                $link = $varnt->getText();
                break;
            }
        }
    }
 
    public function getExtraHashOptions() {
        $variant = $this->getPreferredVariant();
 
        return '!' . $variant;
    }
 
    public function guessVariant( $text, $variant ) {
        return false;
    }
 
    function loadDefaultTables() {
        $name = get_class( $this );
 
        throw new MWException( "Must implement loadDefaultTables() method in class $name" );
    }
 
    function loadTables( $fromCache = true ) {
        global $wgLanguageConverterCacheType;
 
        if ( $this->mTablesLoaded ) {
            return;
        }
 
        $this->mTablesLoaded = true;
        $this->mTables = false;
        $cache = ObjectCache::getInstance( $wgLanguageConverterCacheType );
        if ( $fromCache ) {
            wfProfileIn( __METHOD__ . '-cache' );
            $this->mTables = $cache->get( $this->mCacheKey );
            wfProfileOut( __METHOD__ . '-cache' );
        }
        if ( !$this->mTables || !array_key_exists( self::CACHE_VERSION_KEY, $this->mTables ) ) {
            wfProfileIn( __METHOD__ . '-recache' );
            // not in cache, or we need a fresh reload.
            // We will first load the default tables
            // then update them using things in MediaWiki:Conversiontable/*
            $this->loadDefaultTables();
            foreach ( $this->mVariants as $var ) {
                $cached = $this->parseCachedTable( $var );
                $this->mTables[$var]->mergeArray( $cached );
            }
 
            $this->postLoadTables();
            $this->mTables[self::CACHE_VERSION_KEY] = true;
 
            $cache->set( $this->mCacheKey, $this->mTables, 43200 );
            wfProfileOut( __METHOD__ . '-recache' );
        }
    }
 
    function postLoadTables() {
    }
 
    function reloadTables() {
        if ( $this->mTables ) {
            unset( $this->mTables );
        }
 
        $this->mTablesLoaded = false;
        $this->loadTables( false );
    }
 
    function parseCachedTable( $code, $subpage = '', $recursive = true ) {
        static $parsed = [];
 
        $key = 'Conversiontable/' . $code;
        if ( $subpage ) {
            $key .= '/' . $subpage;
        }
        if ( array_key_exists( $key, $parsed ) ) {
            return [];
        }
 
        $parsed[$key] = true;
 
        if ( $subpage === '' ) {
            $txt = MessageCache::singleton()->getMsgFromNamespace( $key, $code );
        } else {
            $txt = false;
            $title = Title::makeTitleSafe( NS_MEDIAWIKI, $key );
            if ( $title && $title->exists() ) {
                $revision = Revision::newFromTitle( $title );
                if ( $revision ) {
                    if ( $revision->getContentModel() == CONTENT_MODEL_WIKITEXT ) {
                        $txt = $revision->getContent( Revision::RAW )->getNativeData();
                    }
 
                    // @todo in the future, use a specialized content model, perhaps based on json!
                }
            }
        }
 
        # Nothing to parse if there's no text
        if ( $txt === false || $txt === null || $txt === '' ) {
            return [];
        }
 
        // get all subpage links of the form
        // [[MediaWiki:Conversiontable/zh-xx/...|...]]
        $linkhead = $this->mLangObj->getNsText( NS_MEDIAWIKI ) .
            ':Conversiontable';
        $subs = StringUtils::explode( '[[', $txt );
        $sublinks = [];
        foreach ( $subs as $sub ) {
            $link = explode( ']]', $sub, 2 );
            if ( count( $link ) != 2 ) {
                continue;
            }
            $b = explode( '|', $link[0], 2 );
            $b = explode( '/', trim( $b[0] ), 3 );
            if ( count( $b ) == 3 ) {
                $sublink = $b[2];
            } else {
                $sublink = '';
            }
 
            if ( $b[0] == $linkhead && $b[1] == $code ) {
                $sublinks[] = $sublink;
            }
        }
 
        // parse the mappings in this page
        $blocks = StringUtils::explode( '-{', $txt );
        $ret = [];
        $first = true;
        foreach ( $blocks as $block ) {
            if ( $first ) {
                // Skip the part before the first -{
                $first = false;
                continue;
            }
            $mappings = explode( '}-', $block, 2 )[0];
            $stripped = str_replace( [ "'", '"', '*', '#' ], '', $mappings );
            $table = StringUtils::explode( ';', $stripped );
            foreach ( $table as $t ) {
                $m = explode( '=>', $t, 3 );
                if ( count( $m ) != 2 ) {
                    continue;
                }
                // trim any trailling comments starting with '//'
                $tt = explode( '//', $m[1], 2 );
                $ret[trim( $m[0] )] = trim( $tt[0] );
            }
        }
 
        // recursively parse the subpages
        if ( $recursive ) {
            foreach ( $sublinks as $link ) {
                $s = $this->parseCachedTable( $code, $link, $recursive );
                $ret = $s + $ret;
            }
        }
 
        if ( $this->mUcfirst ) {
            foreach ( $ret as $k => $v ) {
                $ret[$this->mLangObj->ucfirst( $k )] = $this->mLangObj->ucfirst( $v );
            }
        }
        return $ret;
    }
 
    public function markNoConversion( $text, $noParse = false ) {
        # don't mark if already marked
        if ( strpos( $text, '-{' ) || strpos( $text, '}-' ) ) {
            return $text;
        }
 
        $ret = "-{R|$text}-";
        return $ret;
    }
 
    function convertCategoryKey( $key ) {
        return $key;
    }
 
    public function updateConversionTable( Title $titleobj ) {
        if ( $titleobj->getNamespace() == NS_MEDIAWIKI ) {
            $title = $titleobj->getDBkey();
            $t = explode( '/', $title, 3 );
            $c = count( $t );
            if ( $c > 1 && $t[0] == 'Conversiontable' ) {
                if ( $this->validateVariant( $t[1] ) ) {
                    $this->reloadTables();
                }
            }
        }
    }
 
    function getVarSeparatorPattern() {
        if ( is_null( $this->mVarSeparatorPattern ) ) {
            // varsep_pattern for preg_split:
            // text should be splited by ";" only if a valid variant
            // name exist after the markup, for example:
            //  -{zh-hans:<span style="font-size:120%;">xxx</span>;zh-hant:\
            //  <span style="font-size:120%;">yyy</span>;}-
            // we should split it as:
            //  [
            //    [0] => 'zh-hans:<span style="font-size:120%;">xxx</span>'
            //    [1] => 'zh-hant:<span style="font-size:120%;">yyy</span>'
            //    [2] => ''
            //  ]
            $pat = '/;\s*(?=';
            foreach ( $this->mVariants as $variant ) {
                // zh-hans:xxx;zh-hant:yyy
                $pat .= $variant . '\s*:|';
                // xxx=>zh-hans:yyy; xxx=>zh-hant:zzz
                $pat .= '[^;]*?=>\s*' . $variant . '\s*:|';
            }
            $pat .= '\s*$)/';
            $this->mVarSeparatorPattern = $pat;
        }
        return $this->mVarSeparatorPattern;
    }
}