<?php

/*

 * Enano - an open-source CMS capable of wiki functions, Drupal-like sidebar blocks, and everything in between

 * Copyright (C) 2006-2008 Dan Fuhry

 *

 * This program is Free Software; you can redistribute 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 details.

 *

 * This script contains code originally found in MediaWiki (http://www.mediawiki.org). MediaWiki is also licensed under

 * the GPLv2; see the file GPL included with this package for details.

 *

 * We're using the MW parser because the Text_Wiki version simply refused to work under PHP 5.2.0. Porting this was

 * _not_ easy. <leaves to get cup of coffee>

 */

  global $mStripState, $wgRandomKey;

  $mStripState = Array();

  $attrib = '[a-zA-Z0-9]';

  $space = '[\x09\x0a\x0d\x20]';

  define( 'MW_CHAR_REFS_REGEX',

	'/&([A-Za-z0-9]+);

	 |&\#([0-9]+);

	 |&\#x([0-9A-Za-z]+);

	 |&\#X([0-9A-Za-z]+);

	 |(&)/x' );

  define( 'MW_ATTRIBS_REGEX',

    "/(?:^|$space)($attrib+)

      ($space*=$space*

      (?:

       # The attribute value: quoted or alone

        ".'"'."([^<".'"'."]*)".'"'."

       | '([^<']*)'

       |  ([a-zA-Z0-9!#$%&()*,\\-.\\/:;<>?@[\\]^_`{|}~]+)

       |  (\#[0-9a-fA-F]+) # Technically wrong, but lots of

                 # colors are specified like this.

                 # We'll be normalizing it.

      )

       )?(?=$space|\$)/sx" );

  /**

   * emulate mediawiki parser, including stripping, etc.

   *

   * @param string $text the text to parse

   * @return string

   * @access public

   */

  function process_tables( $text )

  {

    // include some globals, do some parser stuff that would normally be done in the parent parser function

    global $mStripState;

    $x =& $mStripState;

		//$text = mwStrip( $text, $x );

    // parse the text

    $text = doTableStuff($text);

    // Unstrip it

    // $text = unstrip( $text, $mStripState );

    // $text = unstripNoWiki( $text, $mStripState );

    //die('<pre>'.print_r($mStripState, true).'</pre>');

    return $text;

  }

  /**

	 * parse the wiki syntax used to render tables

	 *

   * @param string $t the text to parse

   * @return string

	 * @access private

	 */

	function doTableStuff( $t ) {

		$t = explode ( "\n" , $t ) ;

		$td = array () ; # Is currently a td tag open?

		$ltd = array () ; # Was it TD or TH?

		$tr = array () ; # Is currently a tr tag open?

		$ltr = array () ; # tr attributes

		$has_opened_tr = array(); # Did this table open a <tr> element?

		$indent_level = 0; # indent level of the table

		foreach ( $t AS $k => $x )

		{

			$x = trim ( $x ) ;

			$fc = substr ( $x , 0 , 1 ) ;

			if ( preg_match( '/^(:*)\{\|(.*)$/', $x, $matches ) ) {

				$indent_level = strlen( $matches[1] );

				$attributes = unstripForHTML( $matches[2] );

				$t[$k] = str_repeat( '<dl><dd>', $indent_level ) .

					'<nowiki><table' . fixTagAttributes( $attributes, 'table' ) . '></nowiki>' ;

				array_push ( $td , false ) ;

				array_push ( $ltd , '' ) ;

				array_push ( $tr , false ) ;

				array_push ( $ltr , '' ) ;

				array_push ( $has_opened_tr, false );

			}

			else if ( count ( $td ) == 0 ) { } # Don't do any of the following

			else if ( '|}' == substr ( $x , 0 , 2 ) ) {

				$z = "<nowiki></table></nowiki>" . substr ( $x , 2);

				$l = array_pop ( $ltd ) ;

				if ( !array_pop ( $has_opened_tr ) ) $z = "<nowiki><tr><td></td></tr></nowiki>" . $z ;

				if ( array_pop ( $tr ) ) $z = '<nowiki></tr></nowiki>' . $z ;

				if ( array_pop ( $td ) ) $z = '<nowiki></'.$l.'></nowiki>' . $z ;

				array_pop ( $ltr ) ;

				$t[$k] = $z . str_repeat( '<nowiki></dd></dl></nowiki>', $indent_level );

			}

			else if ( '|-' == substr ( $x , 0 , 2 ) ) { # Allows for |---------------

				$x = substr ( $x , 1 ) ;

				while ( $x != '' && substr ( $x , 0 , 1 ) == '-' ) $x = substr ( $x , 1 ) ;

				$z = '' ;

				$l = array_pop ( $ltd ) ;

				array_pop ( $has_opened_tr );

				array_push ( $has_opened_tr , true ) ;

				if ( array_pop ( $tr ) ) $z = '<nowiki></tr></nowiki>' . $z ;

				if ( array_pop ( $td ) ) $z = '<nowiki></'.$l.'></nowiki>' . $z ;

				array_pop ( $ltr ) ;

				$t[$k] = $z ;

				array_push ( $tr , false ) ;

				array_push ( $td , false ) ;

				array_push ( $ltd , '' ) ;

				$attributes = unstripForHTML( $x );

				array_push ( $ltr , fixTagAttributes( $attributes, 'tr' ) ) ;

			}

			else if ( '|' == $fc || '!' == $fc || '|+' == substr ( $x , 0 , 2 ) ) { # Caption

				# $x is a table row

				if ( '|+' == substr ( $x , 0 , 2 ) ) {

					$fc = '+' ;

					$x = substr ( $x , 1 ) ;

				}

				$after = substr ( $x , 1 ) ;

				if ( $fc == '!' ) $after = str_replace ( '!!' , '||' , $after ) ;

				// Split up multiple cells on the same line.

				// FIXME: This can result in improper nesting of tags processed

				// by earlier parser steps, but should avoid splitting up eg

				// attribute values containing literal "||".

				$after = wfExplodeMarkup( '||', $after );

				$t[$k] = '' ;

				# Loop through each table cell

				foreach ( $after AS $theline )

				{

					$z = '' ;

					if ( $fc != '+' )

					{

						$tra = array_pop ( $ltr ) ;

						if ( !array_pop ( $tr ) ) $z = '<nowiki><tr'.$tra."></nowiki>\n" ;

						array_push ( $tr , true ) ;

						array_push ( $ltr , '' ) ;

						array_pop ( $has_opened_tr );

						array_push ( $has_opened_tr , true ) ;

					}

					$l = array_pop ( $ltd ) ;

					if ( array_pop ( $td ) ) $z = '<nowiki></'.$l.'></nowiki>' . $z ;

					if ( $fc == '|' ) $l = 'td' ;

					else if ( $fc == '!' ) $l = 'th' ;

					else if ( $fc == '+' ) $l = 'caption' ;

					else $l = '' ;

					array_push ( $ltd , $l ) ;

					# Cell parameters

					$y = explode ( '|' , $theline , 2 ) ;

					# Note that a '|' inside an invalid link should not

					# be mistaken as delimiting cell parameters

					if ( strpos( $y[0], '[[' ) !== false ) {

						$y = array ($theline);

					}

					if ( count ( $y ) == 1 )

						$y = "{$z}<nowiki><{$l}></nowiki>{$y[0]}" ;

					else {

						$attributes = unstripForHTML( $y[0] );

						$y = "{$z}<nowiki><{$l}".fixTagAttributes($attributes, $l)."></nowiki>{$y[1]}" ;

					}

					$t[$k] .= $y ;

					array_push ( $td , true ) ;

				}

			}

		}

		# Closing open td, tr && table

		while ( count ( $td ) > 0 )

		{

			$l = array_pop ( $ltd ) ;

			if ( array_pop ( $td ) ) $t[] = '<nowiki></td></nowiki>' ;

			if ( array_pop ( $tr ) ) $t[] = '<nowiki></tr></nowiki>' ;

			if ( !array_pop ( $has_opened_tr ) ) $t[] = "<nowiki><tr><td></td></tr></nowiki>" ;

			$t[] = '<nowiki></table></nowiki>' ;

		}

		$t = implode ( "\n" , $t ) ;

		# special case: don't return empty table

		if($t == "<nowiki><table></nowiki>\n<nowiki><tr><td></td></tr></nowiki>\n<nowiki></table></nowiki>")

			$t = '';

		return $t ;

	}

  /**

	 * Take a tag soup fragment listing an HTML element's attributes

	 * and normalize it to well-formed XML, discarding unwanted attributes.

	 * Output is safe for further wikitext processing, with escaping of

	 * values that could trigger problems.

	 *

	 * - Normalizes attribute names to lowercase

	 * - Discards attributes not on a whitelist for the given element

	 * - Turns broken or invalid entities into plaintext

	 * - Double-quotes all attribute values

	 * - Attributes without values are given the name as attribute

	 * - Double attributes are discarded

	 * - Unsafe style attributes are discarded

	 * - Prepends space if there are attributes.

	 *

	 * @param string $text

	 * @param string $element

	 * @return string

	 */

	function fixTagAttributes( $text, $element ) {

		if( trim( $text ) == '' ) {

			return '';

		}

		$stripped = validateTagAttributes(

			decodeTagAttributes( $text ), $element );

		$attribs = array();

		foreach( $stripped as $attribute => $value ) {

			$encAttribute = htmlspecialchars( $attribute );

			$encValue = safeEncodeAttribute( $value );

			$attribs[] = "$encAttribute=".'"'."$encValue".'"'.""; // "

		}

		return count( $attribs ) ? ' ' . implode( ' ', $attribs ) : '';

	}

  /**

	 * Encode an attribute value for HTML tags, with extra armoring

	 * against further wiki processing.

	 * @param $text

	 * @return HTML-encoded text fragment

	 */

	function safeEncodeAttribute( $text ) {

		$encValue= encodeAttribute( $text );

		# Templates and links may be expanded in later parsing,

		# creating invalid or dangerous output. Suppress this.

		$encValue = strtr( $encValue, array(

			'<'    => '&lt;',   // This should never happen,

			'>'    => '>',   // we've received invalid input

			'"'    => '"', // which should have been escaped.

			'{'    => '{',

			'['    => '[',

			"''"   => '''',

			'ISBN' => 'ISBN',

			'RFC'  => 'RFC',

			'PMID' => 'PMID',

			'|'    => '|',

			'__'   => '__',

		) );

		return $encValue;

	}

  /**

	 * Encode an attribute value for HTML output.

	 * @param $text

	 * @return HTML-encoded text fragment

	 */

	function encodeAttribute( $text ) {

    // In Enano 1.0.3, added this cheapo hack to keep ampersands

    // from being double-sanitized. Thanks to markybob from #deluge.

    // htmlspecialchars() the "manual" way

    $encValue = strtr( $text, array(

      '&'  => '&',

      '"' => '"',

      '&lt;'   => '<',

      '>'   => '>',

      ''' => "'"

    ) );

    $encValue = strtr( $text, array(

      '&' => '&',

      '"' => '"',

      '<' => '&lt;',

      '>' => '>',

      "'" => '''

    ) );

		// Whitespace is normalized during attribute decoding,

		// so if we've been passed non-spaces we must encode them

		// ahead of time or they won't be preserved.

		$encValue = strtr( $encValue, array(

			"\n" => '
',

			"\r" => '
',

			"\t" => '	',

		) );

		return $encValue;

	}

  function unstripForHTML( $text ) {

    global $mStripState;

		$text = unstrip( $text, $mStripState );

		$text = unstripNoWiki( $text, $mStripState );

		return $text;

	}

  /**

	 * Always call this after unstrip() to preserve the order

	 *

	 * @private

	 */

	function unstripNoWiki( $text, &$state ) {

		if ( !isset( $state['nowiki'] ) ) {

			return $text;

		}

		# TODO: good candidate for FSS

		$text = strtr( $text, $state['nowiki'] );

		return $text;

	}

  /**

	 * Take an array of attribute names and values and normalize or discard

	 * illegal values for the given element type.

	 *

	 * - Discards attributes not on a whitelist for the given element

	 * - Unsafe style attributes are discarded

	 *

	 * @param array $attribs

	 * @param string $element

	 * @return array

	 *

	 * @todo Check for legal values where the DTD limits things.

	 * @todo Check for unique id attribute :P

	 */

	function validateTagAttributes( $attribs, $element ) {

		$whitelist = array_flip( attributeWhitelist( $element ) );

		$out = array();

		foreach( $attribs as $attribute => $value ) {

			if( !isset( $whitelist[$attribute] ) ) {

				continue;

			}

			# Strip javascript "expression" from stylesheets.

			# http://msdn.microsoft.com/workshop/author/dhtml/overview/recalc.asp

			if( $attribute == 'style' ) {

				$value = checkCss( $value );

				if( $value === false ) {

					# haxx0r

					continue;

				}

			}

			if ( $attribute === 'id' )

				$value = escapeId( $value );

			// If this attribute was previously set, override it.

			// Output should only have one attribute of each name.

			$out[$attribute] = $value;

		}

		return $out;

	}

  /**

	 * Pick apart some CSS and check it for forbidden or unsafe structures.

	 * Returns a sanitized string, or false if it was just too evil.

	 *

	 * Currently URL references, 'expression', 'tps' are forbidden.

	 *

	 * @param string $value

	 * @return mixed

	 */

	function checkCss( $value ) {

		$stripped = decodeCharReferences( $value );

		// Remove any comments; IE gets token splitting wrong

		$stripped = preg_replace( '!/\\*.*?\\*/!S', '', $stripped );

		$value = $stripped;

		// ... and continue checks

		$stripped = preg_replace( '!\\\\([0-9A-Fa-f]{1,6})[ \\n\\r\\t\\f]?!e',

			'codepointToUtf8(hexdec("$1"))', $stripped );

		$stripped = str_replace( '\\', '', $stripped );

		if( preg_match( '/(expression|tps*:\/\/|url\\s*\().*/is',

				$stripped ) ) {

			# haxx0r

			return false;

		}

		return $value;

	}

  /**

	 * Decode any character references, numeric or named entities,

	 * in the text and return a UTF-8 string.

	 *

	 * @param string $text

	 * @return string

	 * @access public

	 * @static

	 */

	function decodeCharReferences( $text ) {

		return preg_replace_callback(

			MW_CHAR_REFS_REGEX,

			'decodeCharReferencesCallback',

			$text );

	}

  /**

	 * Fetch the whitelist of acceptable attributes for a given

	 * element name.

	 *

	 * @param string $element

	 * @return array

	 */

	function attributeWhitelist( $element ) {

		static $list;

		if( !isset( $list ) ) {

			$list = setupAttributeWhitelist();

		}

		return isset( $list[$element] )

			? $list[$element]

			: array();

	}

  /**

	 * @todo Document it a bit

	 * @return array

	 */

	function setupAttributeWhitelist() {

    global $db, $session, $paths, $template, $plugins;

		$common = array( 'id', 'class', 'lang', 'dir', 'title', 'style' );

		$block = array_merge( $common, array( 'align' ) );

		$tablealign = array( 'align', 'char', 'charoff', 'valign' );

		$tablecell = array( 'abbr',

		                    'axis',

		                    'headers',

		                    'scope',

		                    'rowspan',

		                    'colspan',

		                    'nowrap', # deprecated

		                    'width',  # deprecated

		                    'height', # deprecated

		                    'bgcolor' # deprecated

		                    );

		# Numbers refer to sections in HTML 4.01 standard describing the element.

		# See: http://www.w3.org/TR/html4/

		$whitelist = array (

			# 7.5.4

			'div'        => $block,

			'center'     => $common, # deprecated

			'span'       => $block, # ??

			# 7.5.5

			'h1'         => $block,

			'h2'         => $block,

			'h3'         => $block,

			'h4'         => $block,

			'h5'         => $block,

			'h6'         => $block,

			# 7.5.6

			# address

			# 8.2.4

			# bdo

			# 9.2.1

			'em'         => $common,

			'strong'     => $common,

			'cite'       => $common,

			# dfn

			'code'       => $common,

			# samp

			# kbd

			'var'        => $common,

			# abbr

			# acronym

			# 9.2.2

			'blockquote' => array_merge( $common, array( 'cite' ) ),

			# q

			# 9.2.3

			'sub'        => $common,

			'sup'        => $common,

			# 9.3.1

			'p'          => $block,

			# 9.3.2

			'br'         => array( 'id', 'class', 'title', 'style', 'clear' ),

			# 9.3.4

			'pre'        => array_merge( $common, array( 'width' ) ),

			# 9.4

			'ins'        => array_merge( $common, array( 'cite', 'datetime' ) ),

			'del'        => array_merge( $common, array( 'cite', 'datetime' ) ),

			# 10.2

			'ul'         => array_merge( $common, array( 'type' ) ),

			'ol'         => array_merge( $common, array( 'type', 'start' ) ),

			'li'         => array_merge( $common, array( 'type', 'value' ) ),

			# 10.3

			'dl'         => $common,

			'dd'         => $common,

			'dt'         => $common,

			# 11.2.1