parser.php

来自「php 开发的内容管理系统」· PHP 代码 · 共 2,095 行 · 第 1/5 页

<?php
/**
 * File for Parser and related classes
 *
 * @package MediaWiki
 * @subpackage Parser
 */

/**
 * Update this version number when the ParserOutput format
 * changes in an incompatible way, so the parser cache
 * can automatically discard old data.
 */
define( 'MW_PARSER_VERSION', '1.6.1' );

/**
 * Variable substitution O(N^2) attack
 *
 * Without countermeasures, it would be possible to attack the parser by saving
 * a page filled with a large number of inclusions of large pages. The size of
 * the generated page would be proportional to the square of the input size.
 * Hence, we limit the number of inclusions of any given page, thus bringing any
 * attack back to O(N).
 */

define( 'MAX_INCLUDE_REPEAT', 100 );
define( 'MAX_INCLUDE_SIZE', 1000000 ); // 1 Million

define( 'RLH_FOR_UPDATE', 1 );

# Allowed values for $mOutputType
define( 'OT_HTML', 1 );
define( 'OT_WIKI', 2 );
define( 'OT_MSG' , 3 );

# Flags for setFunctionHook
define( 'SFH_NO_HASH', 1 );

# string parameter for extractTags which will cause it
# to strip HTML comments in addition to regular
# <XML>-style tags. This should not be anything we
# may want to use in wikisyntax
define( 'STRIP_COMMENTS', 'HTMLCommentStrip' );

# Constants needed for external link processing
define( 'HTTP_PROTOCOLS', 'http:\/\/|https:\/\/' );
# Everything except bracket, space, or control characters
define( 'EXT_LINK_URL_CLASS', '[^][<>"\\x00-\\x20\\x7F]' );
# Including space, but excluding newlines
define( 'EXT_LINK_TEXT_CLASS', '[^\]\\x0a\\x0d]' );
define( 'EXT_IMAGE_FNAME_CLASS', '[A-Za-z0-9_.,~%\\-+&;#*?!=()@\\x80-\\xFF]' );
define( 'EXT_IMAGE_EXTENSIONS', 'gif|png|jpg|jpeg' );
define( 'EXT_LINK_BRACKETED',  '/\[(\b(' . wfUrlProtocols() . ')'.
	EXT_LINK_URL_CLASS.'+) *('.EXT_LINK_TEXT_CLASS.'*?)\]/S' );
define( 'EXT_IMAGE_REGEX',
	'/^('.HTTP_PROTOCOLS.')'.  # Protocol
	'('.EXT_LINK_URL_CLASS.'+)\\/'.  # Hostname and path
	'('.EXT_IMAGE_FNAME_CLASS.'+)\\.((?i)'.EXT_IMAGE_EXTENSIONS.')$/S' # Filename
);

// State constants for the definition list colon extraction
define( 'MW_COLON_STATE_TEXT', 0 );
define( 'MW_COLON_STATE_TAG', 1 );
define( 'MW_COLON_STATE_TAGSTART', 2 );
define( 'MW_COLON_STATE_CLOSETAG', 3 );
define( 'MW_COLON_STATE_TAGSLASH', 4 );
define( 'MW_COLON_STATE_COMMENT', 5 );
define( 'MW_COLON_STATE_COMMENTDASH', 6 );
define( 'MW_COLON_STATE_COMMENTDASHDASH', 7 );

/**
 * PHP Parser
 *
 * Processes wiki markup
 *
 * <pre>
 * There are three main entry points into the Parser class:
 * parse()
 *   produces HTML output
 * preSaveTransform().
 *   produces altered wiki markup.
 * transformMsg()
 *   performs brace substitution on MediaWiki messages
 *
 * Globals used:
 *    objects:   $wgLang, $wgContLang
 *
 * NOT $wgArticle, $wgUser or $wgTitle. Keep them away!
 *
 * settings:
 *  $wgUseTex*, $wgUseDynamicDates*, $wgInterwikiMagic*,
 *  $wgNamespacesWithSubpages, $wgAllowExternalImages*,
 *  $wgLocaltimezone, $wgAllowSpecialInclusion*
 *
 *  * only within ParserOptions
 * </pre>
 *
 * @package MediaWiki
 */
class Parser
{
	/**#@+
	 * @private
	 */
	# Persistent:
	var $mTagHooks, $mFunctionHooks, $mFunctionSynonyms, $mVariables;

	# Cleared with clearState():
	var $mOutput, $mAutonumber, $mDTopen, $mStripState = array();
	var $mIncludeCount, $mArgStack, $mLastSection, $mInPre;
	var $mInterwikiLinkHolders, $mLinkHolders, $mUniqPrefix;
	var $mTemplates,	// cache of already loaded templates, avoids
		                // multiple SQL queries for the same string
	    $mTemplatePath;	// stores an unsorted hash of all the templates already loaded
		                // in this path. Used for loop detection.

	# Temporary
	# These are variables reset at least once per parse regardless of $clearState
	var $mOptions,      // ParserOptions object
		$mTitle,        // Title context, used for self-link rendering and similar things
		$mOutputType,   // Output type, one of the OT_xxx constants
		$mRevisionId;   // ID to display in {{REVISIONID}} tags

	/**#@-*/

	/**
	 * Constructor
	 *
	 * @public
	 */
	function Parser() {
		$this->mTagHooks = array();
		$this->mFunctionHooks = array();
		$this->mFunctionSynonyms = array( 0 => array(), 1 => array() );
		$this->mFirstCall = true;
	}

	/**
	 * Do various kinds of initialisation on the first call of the parser
	 */
	function firstCallInit() {
		if ( !$this->mFirstCall ) {
			return;
		}

		wfProfileIn( __METHOD__ );
		global $wgAllowDisplayTitle, $wgAllowSlowParserFunctions;

		$this->setHook( 'pre', array( $this, 'renderPreTag' ) );

		$this->setFunctionHook( MAG_NS, array( 'CoreParserFunctions', 'ns' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_URLENCODE, array( 'CoreParserFunctions', 'urlencode' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_LCFIRST, array( 'CoreParserFunctions', 'lcfirst' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_UCFIRST, array( 'CoreParserFunctions', 'ucfirst' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_LC, array( 'CoreParserFunctions', 'lc' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_UC, array( 'CoreParserFunctions', 'uc' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_LOCALURL, array( 'CoreParserFunctions', 'localurl' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_LOCALURLE, array( 'CoreParserFunctions', 'localurle' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_FULLURL, array( 'CoreParserFunctions', 'fullurl' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_FULLURLE, array( 'CoreParserFunctions', 'fullurle' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_FORMATNUM, array( 'CoreParserFunctions', 'formatnum' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_GRAMMAR, array( 'CoreParserFunctions', 'grammar' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_PLURAL, array( 'CoreParserFunctions', 'plural' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_NUMBEROFPAGES, array( 'CoreParserFunctions', 'numberofpages' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_NUMBEROFUSERS, array( 'CoreParserFunctions', 'numberofusers' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_NUMBEROFARTICLES, array( 'CoreParserFunctions', 'numberofarticles' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_NUMBEROFFILES, array( 'CoreParserFunctions', 'numberoffiles' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_NUMBEROFADMINS, array( 'CoreParserFunctions', 'numberofadmins' ), SFH_NO_HASH );
		$this->setFunctionHook( MAG_LANGUAGE, array( 'CoreParserFunctions', 'language' ), SFH_NO_HASH );

		if ( $wgAllowDisplayTitle ) {
			$this->setFunctionHook( MAG_DISPLAYTITLE, array( 'CoreParserFunctions', 'displaytitle' ), SFH_NO_HASH );
		}
		if ( $wgAllowSlowParserFunctions ) {
			$this->setFunctionHook( MAG_PAGESINNAMESPACE, array( 'CoreParserFunctions', 'pagesinnamespace' ), SFH_NO_HASH );
		}
		
		$this->initialiseVariables();

		$this->mFirstCall = false;
		wfProfileOut( __METHOD__ );
	}		

	/**
	 * Clear Parser state
	 *
	 * @private
	 */
	function clearState() {
		if ( $this->mFirstCall ) {
			$this->firstCallInit();
		}
		$this->mOutput = new ParserOutput;
		$this->mAutonumber = 0;
		$this->mLastSection = '';
		$this->mDTopen = false;
		$this->mIncludeCount = array();
		$this->mStripState = array();
		$this->mArgStack = array();
		$this->mInPre = false;
		$this->mInterwikiLinkHolders = array(
			'texts' => array(),
			'titles' => array()
		);
		$this->mLinkHolders = array(
			'namespaces' => array(),
			'dbkeys' => array(),
			'queries' => array(),
			'texts' => array(),
			'titles' => array()
		);
		$this->mRevisionId = null;
		
		/**
		 * Prefix for temporary replacement strings for the multipass parser.
		 * \x07 should never appear in input as it's disallowed in XML.
		 * Using it at the front also gives us a little extra robustness
		 * since it shouldn't match when butted up against identifier-like
		 * string constructs.
		 */
		$this->mUniqPrefix = "\x07UNIQ" . Parser::getRandomString();

		# Clear these on every parse, bug 4549
 		$this->mTemplates = array();
 		$this->mTemplatePath = array();

		$this->mShowToc = true;
		$this->mForceTocPosition = false;

		wfRunHooks( 'ParserClearState', array( &$this ) );
	}

	/**
	 * Accessor for mUniqPrefix.
	 *
	 * @public
	 */
	function UniqPrefix() {
		return $this->mUniqPrefix;
	}

	/**
	 * Convert wikitext to HTML
	 * Do not call this function recursively.
	 *
	 * @private
	 * @param string $text Text we want to parse
	 * @param Title &$title A title object
	 * @param array $options
	 * @param boolean $linestart
	 * @param boolean $clearState
	 * @param int $revid number to pass in {{REVISIONID}}
	 * @return ParserOutput a ParserOutput
	 */
	function parse( $text, &$title, $options, $linestart = true, $clearState = true, $revid = null ) {
		/**
		 * First pass--just handle <nowiki> sections, pass the rest off
		 * to internalParse() which does all the real work.
		 */

		global $wgUseTidy, $wgAlwaysUseTidy, $wgContLang;
		$fname = 'Parser::parse';
		wfProfileIn( $fname );

		if ( $clearState ) {
			$this->clearState();
		}

		$this->mOptions = $options;
		$this->mTitle =& $title;
		$this->mRevisionId = $revid;
		$this->mOutputType = OT_HTML;

		//$text = $this->strip( $text, $this->mStripState );
		// VOODOO MAGIC FIX! Sometimes the above segfaults in PHP5.
		$x =& $this->mStripState;

		wfRunHooks( 'ParserBeforeStrip', array( &$this, &$text, &$x ) );
		$text = $this->strip( $text, $x );
		wfRunHooks( 'ParserAfterStrip', array( &$this, &$text, &$x ) );

		# Hook to suspend the parser in this state
		if ( !wfRunHooks( 'ParserBeforeInternalParse', array( &$this, &$text, &$x ) ) ) {
			wfProfileOut( $fname );
			return $text ;
		}

		$text = $this->internalParse( $text );

		$text = $this->unstrip( $text, $this->mStripState );

		# Clean up special characters, only run once, next-to-last before doBlockLevels
		$fixtags = array(
			# french spaces, last one Guillemet-left
			# only if there is something before the space
			'/(.) (?=\\?|:|;|!|\\302\\273)/' => '\\1&nbsp;\\2',
			# french spaces, Guillemet-right
			'/(\\302\\253) /' => '\\1&nbsp;',
		);
		$text = preg_replace( array_keys($fixtags), array_values($fixtags), $text );

		# only once and last
		$text = $this->doBlockLevels( $text, $linestart );

		$this->replaceLinkHolders( $text );

		# the position of the parserConvert() call should not be changed. it
		# assumes that the links are all replaced and the only thing left
		# is the <nowiki> mark.
		# Side-effects: this calls $this->mOutput->setTitleText()
		$text = $wgContLang->parserConvert( $text, $this );

		$text = $this->unstripNoWiki( $text, $this->mStripState );

		wfRunHooks( 'ParserBeforeTidy', array( &$this, &$text ) );

		$text = Sanitizer::normalizeCharReferences( $text );

		if (($wgUseTidy and $this->mOptions->mTidy) or $wgAlwaysUseTidy) {
			$text = Parser::tidy($text);
		} else {
			# attempt to sanitize at least some nesting problems
			# (bug #2702 and quite a few others)
			$tidyregs = array(	
				# ''Something [http://www.cool.com cool''] --> 
				# <i>Something</i><a href="http://www.cool.com"..><i>cool></i></a>
				'/(<([bi])>)(<([bi])>)?([^<]*)(<\/?a[^<]*>)([^<]*)(<\/\\4>)?(<\/\\2>)/' =>
				'\\1\\3\\5\\8\\9\\6\\1\\3\\7\\8\\9',
				# fix up an anchor inside another anchor, only
				# at least for a single single nested link (bug 3695)
				'/(<a[^>]+>)([^<]*)(<a[^>]+>[^<]*)<\/a>(.*)<\/a>/' =>
				'\\1\\2</a>\\3</a>\\1\\4</a>',
				# fix div inside inline elements- doBlockLevels won't wrap a line which
				# contains a div, so fix it up here; replace
				# div with escaped text
				'/(<([aib]) [^>]+>)([^<]*)(<div([^>]*)>)(.*)(<\/div>)([^<]*)(<\/\\2>)/' =>
				'\\1\\3&lt;div\\5&gt;\\6&lt;/div&gt;\\8\\9',
				# remove empty italic or bold tag pairs, some
				# introduced by rules above
				'/<([bi])><\/\\1>/' => '' 
			);

			$text = preg_replace( 
				array_keys( $tidyregs ),
				array_values( $tidyregs ),
				$text );
		}

		wfRunHooks( 'ParserAfterTidy', array( &$this, &$text ) );

		$this->mOutput->setText( $text );
		wfProfileOut( $fname );

		return $this->mOutput;
	}

	/**
	 * Get a random string
	 *
	 * @private
	 * @static
	 */
	function getRandomString() {
		return dechex(mt_rand(0, 0x7fffffff)) . dechex(mt_rand(0, 0x7fffffff));
	}

	function &getTitle() { return $this->mTitle; }
	function getOptions() { return $this->mOptions; }

	function getFunctionLang() {
		global $wgLang, $wgContLang;
		return $this->mOptions->getInterfaceMessage() ? $wgLang : $wgContLang;
	}

	/**
	 * Replaces all occurrences of HTML-style comments and the given tags
	 * in the text with a random marker and returns teh next text. The output
	 * parameter $matches will be an associative array filled with data in
	 * the form:
	 *   'UNIQ-xxxxx' => array(
	 *     'element',
	 *     'tag content',
	 *     array( 'param' => 'x' ),
	 *     '<element param="x">tag content</element>' ) )
	 *
	 * @param $elements list of element names. Comments are always extracted.
	 * @param $text Source text string.
	 * @param $uniq_prefix
	 *
	 * @private
	 * @static
	 */
	function extractTagsAndParams($elements, $text, &$matches, $uniq_prefix = ''){
		$rand = Parser::getRandomString();
		$n = 1;
		$stripped = '';
		$matches = array();

		$taglist = implode( '|', $elements );
		$start = "/<($taglist)(\\s+[^>]*?|\\s*?)(\/?>)|<(!--)/i";

		while ( '' != $text ) {
			$p = preg_split( $start, $text, 2, PREG_SPLIT_DELIM_CAPTURE );
			$stripped .= $p[0];
			if( count( $p ) < 5 ) {
				break;
			}
			if( count( $p ) > 5 ) {
				// comment
				$element    = $p[4];
				$attributes = '';
				$close      = '';
				$inside     = $p[5];
			} else {
				// tag
				$element    = $p[1];
				$attributes = $p[2];
				$close      = $p[3];
				$inside     = $p[4];