<?php
/**
* @version	$Id: pdf_styles.php 14244 2011-03-16 20:53:41Z alex $
* @package	In-Portal
* @copyright	Copyright (C) 1997 - 2009 Intechnic. All rights reserved.
* @license      GNU/GPL
* In-Portal is Open Source software.
* This means that this software may have been modified pursuant
* the GNU General Public License, and as distributed it includes
* or is derivative of works licensed under the GNU General Public License
* or other free or open source software licenses.
* See http://www.in-portal.org/license for copyright notices and details.
*/

defined('FULL_PATH') or die('restricted access!');

require_once(FULL_PATH.'/core/units/pdf/css_defaults.php');

class kPDFStylesheet {

	protected $Selectors;
	protected $Macros;

	protected $Buffer = array();
	protected $Openings = array();
	protected $Level = 0;

	public $Rules = array();
	public $SelectorOrder = 0;
	public $HTMLVisualPropsSelectorOrder = 0;

	const STYLE_ORIGIN_AGENT_NORMAL = 0;
	const STYLE_ORIGIN_USER_NORMAL = 1;
	const STYLE_ORIGIN_AUTHOR_NORMAL = 2;
	const STYLE_ORIGIN_AUTHOR_IMPORTANT = 3;
	const STYLE_ORIGIN_USER_IMPORTANT = 4;

	function __construct($nodefaults = false)
	{
		$this->Prepare();
		if (!$nodefaults) {
			$tokens = $this->GetTokens(kCSSDefaults::$DEFAULT_STYLE);
			$this->ParseTokens($tokens, kPDFStylesheet::STYLE_ORIGIN_AGENT_NORMAL);
			$this->HTMLVisualPropsSelectorOrder = $this->SelectorOrder;
			$this->SelectorOrder += 1000;
		}
	}

	public function ParseStyle($style)
	{
		$res = array();
		$pairs = explode(';', $style);
		foreach ($pairs as $property) {
			$property = trim($property);
			list($name, $value) = explode(':', $property);
			$res[trim($name)] = trim($value);
		}
	}

	/*
		stylesheet  : [ CDO | CDC | S | statement ]*;
		statement   : ruleset | at-rule;
		at-rule     : ATKEYWORD S* any* [ block | ';' S* ];
		block       : '{' S* [ any | block | ATKEYWORD S* | ';' S* ]* '}' S*;
		ruleset     : selector? '{' S* declaration? [ ';' S* declaration? ]* '}' S*;
		selector    : any+;
		declaration : DELIM? property S* ':' S* value;
		property    : IDENT;
		value       : [ any | block | ATKEYWORD S* ]+;
		any         : [ IDENT | NUMBER | PERCENTAGE | DIMENSION | STRING
		              | DELIM | URI | HASH | UNICODE-RANGE | INCLUDES
		              | DASHMATCH | FUNCTION S* any* ')'
		              | '(' S* any* ')' | '[' S* any* ']' ] S*;
	*/

	function ParseTokens($tokens, $origin=kPDFStylesheet::STYLE_ORIGIN_AUTHOR_NORMAL )
	{
		$this->Buffer[0] = Array();
		foreach ($tokens as $token) {
			if ($token['name'] == 'LBRACE') {
				$this->Buffer[++$this->Level] = Array();
				$this->Openings[$this->Level] = 'LBRACE';
			}
			elseif ($token['name'] == 'TEXT' && $token['data'] == '}') {
				if ($this->Level == 1 && $this->Openings[$this->Level] == 'LBRACE') {
					$this->AppendRule($this->Buffer[0], $this->Buffer[$this->Level], $origin);
					$this->Buffer[0] = Array();
				}
				$this->Level--;
			}
			else {
				$this->Buffer[$this->Level][] = $token;
			}
		}
	}

	protected function ConcatTokensData($tokens)
	{
		$res = '';
		foreach ($tokens as $token) {$res .= $token['data'];}
		return $res;
	}

	public function ParseDefinitionTokens($tokens)
	{
		$mode = 'property';
		$properties = array();
		$value = '';
		foreach ($tokens as $token) {
			if ($mode == 'property') {
				if ($token['name'] == 'IDENT') {
					$property = $token['data'];
					$mode = 'colon';
				}
			}
			elseif ($mode == 'colon') {
				if ($token['name'] == 'TEXT' && $token['data'] == ':') {
					$mode = 'value';
				}
			}
			elseif ($mode == 'value') {
				if ($token['name'] == 'TEXT' && $token['data'] == ';') {
					$properties[strtoupper($property)] = trim($value);
					$value = '';
					$mode = 'property';
				}
				else {
					$value .= $token['data'];
				}
			}
		}
		if ($mode == 'value') {
			$properties[strtoupper($property)] = trim($value);
		}
		if ($mode == 'colon') {
			trigger_error('Error parsing CSS definition, no colon and/or value after property '.$property, E_USER_WARNING);
		}

		$properties = $this->ProcessShortHands($properties);

		return $properties;
	}

	public function ProcessShortHands($properties)
	{
		$res = array();
		foreach ($properties as $property => $value)
		{
			switch ($property) {
				case 'MARGIN':
					if (preg_match('/^([.0-9]+(?:px|pt|em|ex|%)?|auto)$/i', $value, $regs)) {
						$res['MARGIN-TOP'] = $regs[1];
						$res['MARGIN-RIGHT'] = $regs[1];
						$res['MARGIN-BOTTOM'] = $regs[1];
						$res['MARGIN-LEFT'] = $regs[1];
					}
					if (preg_match('/^([.0-9]+(?:px|pt|em|ex|%)?|auto) ([.0-9]+(?:px|pt|em|ex|%)?|auto)$/i', $value, $regs)) {
						$res['MARGIN-TOP'] = $regs[1];
						$res['MARGIN-RIGHT'] = $regs[2];
						$res['MARGIN-BOTTOM'] = $regs[1];
						$res['MARGIN-LEFT'] = $regs[2];
					}
					if (preg_match('/^([.0-9]+(?:px|pt|em|ex|%)?|auto) ([.0-9]+(?:px|pt|em|ex|%)?|auto) ([.0-9]+(?:px|pt|em|ex|%)?|auto) ([.0-9]+(?:px|pt|em|ex|%)?|auto)$/i', $value, $regs)) {
						$res['MARGIN-TOP'] = $regs[1];
						$res['MARGIN-RIGHt'] = $regs[2];
						$res['MARGIN-BOTTOM'] = $regs[3];
						$res['MARGIN-LEFT'] = $regs[4];
					}

					break;
				case 'BORDER-TOP':
				case 'BORDER-RIGHT':
				case 'BORDER-BOTTOM':
				case 'BORDER-LEFT':
					$parts = $this->ParseBorderShorthand($value);
					if (isset($parts['style'])) {
						$res[$property.'-STYLE'] = $parts['style'];
					}
					if (isset($parts['width'])) {
						$res[$property.'-WIDTH'] = $parts['width'];
					}
					if (isset($parts['color'])) {
						$res[$property.'-COLOR'] = $parts['color'];
					}
					break;
				case 'BORDER':
					$parts = $this->ParseBorderShorthand($value);
					if (isset($parts['style'])) {
						$res['BORDER-TOP-STYLE'] = $parts['style'];
						$res['BORDER-RIGHT-STYLE'] = $parts['style'];
						$res['BORDER-BOTTOM-STYLE'] = $parts['style'];
						$res['BORDER-LEFT-STYLE'] = $parts['style'];
					}
					if (isset($parts['width'])) {
						$res['BORDER-TOP-WIDTH'] = $parts['width'];
						$res['BORDER-RIGHT-WIDTH'] = $parts['width'];
						$res['BORDER-BOTTOM-WIDTH'] = $parts['width'];
						$res['BORDER-LEFT-WIDTH'] = $parts['width'];
					}
					if (isset($parts['color'])) {
						$res['BORDER-TOP-COLOR'] = $parts['color'];
						$res['BORDER-RIGHT-COLOR'] = $parts['color'];
						$res['BORDER-BOTTOM-COLOR'] = $parts['color'];
						$res['BORDER-LEFT-COLOR'] = $parts['color'];
					}
					break;
				case 'PADDING':
					$parts = explode(' ', $value);
					switch (count($parts)) {
						case 1:
							$res['PADDING-TOP'] = $parts[0];
							$res['PADDING-RIGHT'] = $parts[0];
							$res['PADDING-BOTTOM'] = $parts[0];
							$res['PADDING-LEFT'] = $parts[0];
							break;
						case 2:
							$res['PADDING-TOP'] = $parts[0];
							$res['PADDING-RIGHT'] = $parts[1];
							$res['PADDING-BOTTOM'] = $parts[0];
							$res['PADDING-LEFT'] = $parts[1];
							break;
						case 3:
							$res['PADDING-TOP'] = $parts[0];
							$res['PADDING-RIGHT'] = $parts[1];
							$res['PADDING-BOTTOM'] = $parts[2];
							$res['PADDING-LEFT'] = $parts[1];
							break;
						case 4:
							$res['PADDING-TOP'] = $parts[0];
							$res['PADDING-RIGHT'] = $parts[1];
							$res['PADDING-BOTTOM'] = $parts[2];
							$res['PADDING-LEFT'] = $parts[3];
							break;
					}
					break;
				default:
					$res[$property] = $value;
			}
		}
		return $res;
	}

	public function ParseBorderShorthand($definition)
	{
		$res = array();
		$parts = explode(' ', $definition);
		foreach ($parts as $part) {
			if (preg_match('/none|hidden|dotted|dashed|solid|double|groove|ridge|inset|outset/', $part)) { //style
				$res['style'] = $part;
			}
			elseif (preg_match('/^(thin|medium|thick|[.0-9]+(?:px|pt|em|ex|%)?)/', $part)) { // width
				$res['width'] = $part;
			}
			else { // color
				$res['color'] = $part;
			}
		}
		return $res;
	}

	public function ParseSelectorTokens($tokens, $origin)
	{
		$selectors = array();
		$current = '';
		foreach ($tokens as $token) {
			if ($token['name'] == 'COMMA') {
				$selectors[] = trim($current);
				$current = '';
			}
			else {
				$current .= $token['data'];
			}
		}
		if (trim($current) != '') {
			$selectors[] = trim($current);
		}
		return $this->IdentifySelectors($selectors, $origin);
	}

	/*

	'h'				=>		'[0-9a-f]',
	'nonascii'	=>	'[\\200-\\377]',
	'unicode'		=>	'(\\{h}{1,6}(\r\n|[ \t\r\n\f])?)',
	'escape'		=>	'(\\[^\r\n\f0-9a-f])',
	'nmstart'		=>	'([_a-z]|{nonascii}|{escape})',
	'nmchar'		=>	'([_a-z0-9-]|{nonascii}|{escape})',
	'string1'		=>	'("([^\n\r\f"]|{nl}|{escape})*")',
	'string2'		=>	'(\'([^\n\r\f\']|{nl}|{escape})*\')',
	'invalid1'	=>	'("([^\n\r\f"]|{nl}|{escape})*?)',
	'invalid2'	=>	'(\'([^\n\r\f\']|{nl}|{escape})*?)',

	'ident'			=>	'-?{nmstart}{nmchar}*',
	'name'			=>	'{nmchar}+',
	'num'				=>	'([0-9]+|[0-9]*\.[0-9]+)',
	'string'		=>	'({string1}|{string2})',
	'invalid'		=>	'({invalid1}|{invalid2})',
	'url'				=>	'([!#$%&*-~]|{nonascii}|{escape})*',
	's'					=>	'[ \t\r\n\f]',
	'w'					=>	'{s}*',
	'nl'				=>	'(\n|\r\n|\r|\f)',

	*/

	/*

	A simple selector is either a type selector or universal selector followed immediately by zero or more attribute selectors, ID selectors,
	or pseudo-classes, in any order. The simple selector matches if all of its components match.

	A selector is a chain of one or more simple selectors separated by combinators. Combinators are: whitespace, ">", and "+".
	Whitespace may appear between a combinator and the simple selectors around it.

	A selector's specificity is calculated as follows:

    * count 1 if the selector is a 'style' attribute rather than a selector, 0 otherwise (= a)
    	(In HTML, values of an element's "style" attribute are style sheet rules. These rules have no selectors, so a=1, b=0, c=0, and d=0.)
    * count the number of ID attributes in the selector (= b)
    * count the number of other attributes and pseudo-classes in the selector (= c)
    * count the number of element names and pseudo-elements in the selector (= d)

	The specificity is based only on the form of the selector.
	In particular, a selector of the form "[id=p33]" is counted as an attribute selector (a=0, b=0, c=1, d=0),
	even if the id attribute is defined as an "ID" in the source document's DTD.

	Concatenating the four numbers a-b-c-d (in a number system with a large base) gives the specificity.

	*/

	function IdentifySelectors($selectors, $origin)
	{
		$processed = array();
		$ident = $this->Macros['ident'];

		foreach ($selectors as $selector) {
			$parts = preg_split('/[ ]*([ >+])[ ]*/', $selector, null, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY);
			$parsed_selector = array();
			$a = 0;
			$b = 0;
			$c = 0;
			$d = 0;
			foreach ($parts as $simple_selector) {
				$parsed_part = array();
				if (preg_match('/^([ >+])$/', $simple_selector, $regs)) {
					$parsed_part['combinator'] = $regs[1];
					$parsed_selector[] = $parsed_part;
					continue;
				}
				if (preg_match('/^(\*|'.$ident.')/i', $simple_selector, $regs)) {
					$main = $regs[1];
					if ($regs[1] != '*') {
						$d++;
					}
				}
				else {
					$main = '*';
				}
				$parsed_part['main'] = strtoupper($main);
				if (preg_match_all('/\.('.$ident.')/', $simple_selector, $regs)) {
					$parsed_part['classes'] = $regs[1];
					$c += count($regs[1]);
				}
				if (preg_match_all('/\[([^\]]+)\]/', $simple_selector, $regs)) {
					$atts = $regs[1];
					$c += count($regs[1]);
					$parsed_atts = array();
					foreach ($atts as $attribute) {
						if (preg_match('/^[^=]+$/', $attribute)) {
							$parsed_atts['set'][] = strtoupper($attribute);
						}
						elseif (preg_match('/(.*)\\|=(.*)/', $attribute, $att_regs)) {
							$parsed_atts['hypen'][strtoupper($att_regs[1])] = $att_regs[2];
						}
						elseif (preg_match('/(.*)~=(.*)/', $attribute, $att_regs)) {
							$parsed_atts['space'][strtoupper($att_regs[1])] = $att_regs[2];
						}
						elseif (preg_match('/(.*)=(.*)/', $attribute, $att_regs)) {
							$parsed_atts['equals'][strtoupper($att_regs[1])] = $att_regs[2];
						}
					}
					$parsed_part['atts'] = $parsed_atts;
				}
				if (preg_match_all('/#('.$ident.')/', $simple_selector, $regs)) {
					$parsed_part['ids'] = $regs[1];
					$b += count($regs[1]);
				}
				if (preg_match_all('/:('.$ident.')/', $simple_selector, $regs)) {
					$pseudo_classes = array();
					$pseudo_elements = array();
					foreach ($regs[1] as $pseudo) {
						if (preg_match('/^(first-line|first-letter|before|after)$/i', $pseudo)) {
							$pseudo_elements[] = $pseudo;
						}
						else {
							$pseudo_classes[] = $pseudo;
						}
					}
					if ($pseudo_classes) {
						$parsed_part['pseudo_classes'] = $pseudo_classes;
					}
					if ($pseudo_elements) {
						$parsed_part['pseudo_elements'] = $pseudo_elements;
					}
					$c += count($pseudo_classes);
					$d += count($pseudo_elements);
				}
				$parsed_selector[] = $parsed_part;
			}
			$parsed_selector = array_reverse($parsed_selector);

			$main = array();
			$cur =& $main;

			foreach ($parsed_selector as $parts) {
				if (isset($parts['combinator'])) {
					switch ($parts['combinator']) {
						case ' ':
							$cur =& $cur['descendant_of'];
							break;
						case '>':
							$cur =& $cur['child_of'];
							break;
						case '+':
							$cur =& $cur['sibling_of'];
							break;
					}
					continue;
				}
				$cur['main'] = $parts['main'];
				if (isset($parts['classes'])) {
					$cur['classes'] = $parts['classes'];
				}
				if (isset($parts['ids'])) {
					$cur['ids'] = $parts['ids'];
				}
				if (isset($parts['pseudo_classes'])) {
					$cur['pseudo_classes'] = $parts['pseudo_classes'];
				}
				if (isset($parts['pseudo_elements'])) {
					$cur['pseudo_elements'] = $parts['pseudo_elements'];
				}
				if (isset($parts['atts'])) {
					$cur['atts'] = $parts['atts'];
				}
			}
			$main['specifity'] = intval(str_pad($a,2,0).str_pad($b,2,0).str_pad($c,2,0).str_pad($d,2,0));
			$main['order'] = $this->SelectorOrder++;
			$main['origin'] = $origin;
			$processed[] = $main;
		}
		return $processed;
	}

	public function AppendRule($selector_tokens, $definition_tokens, $origin)
	{
		$selectors = $this->ParseSelectorTokens($selector_tokens, $origin);
		$properties = $this->ParseDefinitionTokens($definition_tokens);
		$definition = '';
		foreach ($properties as $property => $value) {
			$definition .= "<b>$property</b>: $value<br>";
		}

		foreach ($selectors as $selector) {
			$this->Mapping[strtoupper($selector['main'])][] = array('selector' => $selector, 'properties' => $properties);
		}

		$this->Rules[] = array('selectors' => $selectors, 'properties' => $properties);
//		echo "appending rule:<br> selector: ".join(',', $selectors)."<br> definition:<br> $definition<br><br>";
	}

	public function GetTokens($css)
	{
		$patterns = array(
			'{s}+'			=>'S',

			'<!--'						=>'CDO',
			'-->'							=>'CDC',
			'~='							=>'INCLUDES',
			'\\|='							=>'DASHMATCH',

			'{w}\\{'					=>'LBRACE',
			'{w}\\+'					=>'PLUS',
			'{w}\\>'					=>'GREATER',
			'{w},'						=>'COMMA',

			'{string}'				=>'STRING',
			'{invalid}'				=>'INVALID', /* unclosed string */

			'{ident}'					=>'IDENT',

			'#{name}'					=>'HASH',

			'@import'					=>'IMPORT_SYM',
			'@page'						=>'PAGE_SYM',
			'@media'					=>'MEDIA_SYM',
			'@charset'				=>'CHARSET_SYM',

			'!{w}important'	=>'IMPORTANT_SYM',

			/*'{num}{E}{M}'				=>'EMS',
			'{num}{E}{X}'				=>'EXS',
			'{num}{P}{X}'				=>'LENGTH',
			'{num}{C}{M}'				=>'LENGTH',
			'{num}{M}{M}'				=>'LENGTH',
			'{num}{I}{N}'				=>'LENGTH',
			'{num}{P}{T}'				=>'LENGTH',
			'{num}{P}{C}'				=>'LENGTH',
			'{num}{D}{E}{G}'		=>'ANGLE',
			'{num}{R}{A}{D}'		=>'ANGLE',
			'{num}{G}{R}{A}{D}'	=>'ANGLE',
			'{num}{M}{S}'				=>'TIME',
			'{num}{S}'					=>'TIME',
			'{num}{H}{Z}'				=>'FREQ',
			'{num}{K}{H}{Z}'		=>'FREQ',
			'{num}{ident}'			=>'DIMENSION',*/

			'{num}em'				=>'EMS',
			'{num}ex'				=>'EXS',
			'{num}px'				=>'LENGTH',
			'{num}cm'				=>'LENGTH',
			'{num}mm'				=>'LENGTH',
			'{num}in'				=>'LENGTH',
			'{num}pt'				=>'LENGTH',
			'{num}pc'				=>'LENGTH',
			'{num}deg'		=>'ANGLE',
			'{num}rad'		=>'ANGLE',
			'{num}grad'	=>'ANGLE',
			'{num}ms'				=>'TIME',
			'{num}s'					=>'TIME',
			'{num}hz'				=>'FREQ',
			'{num}khz'		=>'FREQ',
			'{num}{ident}'			=>'DIMENSION',

			'{num}%'						=>'PERCENTAGE',
			'{num}'							=>'NUMBER',

			'url\({w}{string}{w}\)'	=>'URI',
			'url\({w}{url}{w}\)'		=>'URI',
			'{ident}\('							=>'FUNCTION',

			/*'.'									=>'*yytext',*/
		);

		$final_patterns = array();
		foreach ($patterns as $regexp => $token) {
			foreach ($this->Macros as $macro => $replacement) {
				$regexp = str_replace('{'.$macro.'}', $replacement, $regexp);
			}
			$final_patterns[$regexp] = $token;
		}

		$css = preg_replace('/\\/\\*[^*]*\\*+([^\\/*][^*]*\\*+)*\\//', '', $css);
		$css = preg_replace('/[ \t\r\n\f]+\\/\\*[^*]*\\*+([^\\/*][^*]*\\*+)*\\//', ' ', $css);

		$css = preg_replace('/[ \t\r\n\f]+/', ' ', $css); // remove repeated whitespace

		$matches = array();
		$token_indexes = array();
		foreach ($final_patterns as $regexp => $token) {
			if (preg_match_all('/'.$regexp.'/i', $css, $res, PREG_PATTERN_ORDER | PREG_OFFSET_CAPTURE)) {
				$matches[$token] = $res[0];
				$token_indexes[$token] = 0;
			}
		}

		$tokens = array();
		$last_token_pos = 0;
		$i = 0;
		do {
			$has_more = false;
			$max_len = 0;
			$min_pos = false;
			foreach ($matches as $token => $data)
			{
				$cur_index = $token_indexes[$token];
				do {
					$cur_match = isset($data[$cur_index]) ? $data[$cur_index++] : false;
				} while ($cur_match && $cur_match[1] < $last_token_pos);
				if ( !$cur_match ) continue;
				$token_indexes[$token] = $cur_index-1;
				if (  $min_pos === false ||
							($cur_match[1] < $min_pos
								||
							 ( $cur_match[1] == $min_pos && strlen( $cur_match[0] ) > $max_len )
							)
					 ) {
					$longest = $token;
					$max_len = strlen( $cur_match[0] );
					$min_pos = $cur_match[1];
				}
				$has_more = $has_more || isset($data[$token_indexes[$token]]);
			}
			if ($min_pos !== false) {
				$token_data = $matches[$longest][$token_indexes[$longest]];
				if ($token_data[1] > $last_token_pos) {
					$text_data = substr($css, $last_token_pos, $token_data[1] - $last_token_pos);
					$tokens[] = array('name' => 'TEXT', 'data' => $text_data);
//					echo "found token TEXT: [$text_data]<br>\n";
				}
				$tokens[] = array('name' => $longest, 'data' => $token_data[0]);
//				echo "found token $longest: {$token_data[0]} at {$token_data[1]}<br>\n";
//				flush();
				$last_token_pos = $token_data[1] + strlen($token_data[0]);
				$token_indexes[$longest]++;
			}
		} while ($has_more);
		if ($last_token_pos <= strlen($css)) {
			$text_data = substr($css, $last_token_pos);
			$tokens[] = array('name' => 'TEXT', 'data' => $text_data);
//			echo "found token FINAL TEXT: [$text_data]<br>\n";
		}

		return $tokens;
	}

	public function Prepare()
	{
		/*$macros = array(
			'h'				=>		'[0-9a-f]',
			'nonascii'	=>	'[\200-\377]',
			'unicode'		=>	'(\\{h}{1,6}(\r\n|[ \t\r\n\f])?)',
			'escape'		=>	'({unicode}|\\[^\r\n\f0-9a-f])',
			'nmstart'		=>	'([_a-z]|{nonascii}|{escape})',
			'nmchar'		=>	'([_a-z0-9-]|{nonascii}|{escape})',
			'string1'		=>	'("([^\n\r\f"]|{nl}|{escape})*")',
			'string2'		=>	'(\'([^\n\r\f\']|{nl}|{escape})*\')',
			'invalid1'	=>	'("([^\n\r\f"]|{nl}|{escape})*?)',
			'invalid2'	=>	'(\'([^\n\r\f\']|{nl}|{escape})*?)',

			'ident'			=>	'-?{nmstart}{nmchar}*',
			'name'			=>	'{nmchar}+',
			'num'				=>	'([0-9]+|[0-9]*\.[0-9]+)',
			'string'		=>	'({string1}|{string2})',
			'invalid'		=>	'({invalid1}|{invalid2})',
			'url'				=>	'([!#$%&*-~]|{nonascii}|{escape})*',
			's'					=>	'[ \t\r\n\f]',
			'w'					=>	'{s}*',
			'nl'				=>	'(\n|\r\n|\r|\f)',

			'A'					=>	'a|\\0{0,4}(41|61)(\r\n|[ \t\r\n\f])?',
			'C'					=>	'c|\\0{0,4}(43|63)(\r\n|[ \t\r\n\f])?',
			'D'					=>	'd|\\0{0,4}(44|64)(\r\n|[ \t\r\n\f])?',
			'E'					=>	'e|\\0{0,4}(45|65)(\r\n|[ \t\r\n\f])?',
			'G'					=>	'g|\\0{0,4}(47|67)(\r\n|[ \t\r\n\f])?|\\g',
			'H'					=>	'h|\\0{0,4}(48|68)(\r\n|[ \t\r\n\f])?|\\h',
			'I'					=>	'i|\\0{0,4}(49|69)(\r\n|[ \t\r\n\f])?|\\i',
			'K'					=>	'k|\\0{0,4}(4b|6b)(\r\n|[ \t\r\n\f])?|\\k',
			'M'					=>	'm|\\0{0,4}(4d|6d)(\r\n|[ \t\r\n\f])?|\\m',
			'N'					=>	'n|\\0{0,4}(4e|6e)(\r\n|[ \t\r\n\f])?|\\n',
			'P'					=>	'p|\\0{0,4}(50|70)(\r\n|[ \t\r\n\f])?|\\p',
			'R'					=>	'r|\\0{0,4}(52|72)(\r\n|[ \t\r\n\f])?|\\r',
			'S'					=>	's|\\0{0,4}(53|73)(\r\n|[ \t\r\n\f])?|\\s',
			'T'					=>	't|\\0{0,4}(54|74)(\r\n|[ \t\r\n\f])?|\\t',
			'X'					=>	'x|\\0{0,4}(58|78)(\r\n|[ \t\r\n\f])?|\\x',
			'Z'					=>	'z|\\0{0,4}(5a|7a)(\r\n|[ \t\r\n\f])?|\\z',
		);*/

		$simple = array(
			'h'				=>		'[0-9a-f]',
			'nonascii'	=>	'[\\200-\\377]',
			'unicode'		=>	'(\\{h}{1,6}(\r\n|[ \t\r\n\f])?)',
			'escape'		=>	'(\\[^\r\n\f0-9a-f])',
			'nmstart'		=>	'([_a-z]|{nonascii}|{escape})',
			'nmchar'		=>	'([_a-z0-9-]|{nonascii}|{escape})',
			'string1'		=>	'("([^\n\r\f"]|{nl}|{escape})*")',
			'string2'		=>	'(\'([^\n\r\f\']|{nl}|{escape})*\')',
			'invalid1'	=>	'("([^\n\r\f"]|{nl}|{escape})*?)',
			'invalid2'	=>	'(\'([^\n\r\f\']|{nl}|{escape})*?)',

			'ident'			=>	'-?{nmstart}{nmchar}*',
			'name'			=>	'{nmchar}+',
			'num'				=>	'([0-9]+|[0-9]*\.[0-9]+)',
			'string'		=>	'({string1}|{string2})',
			'invalid'		=>	'({invalid1}|{invalid2})',
			'url'				=>	'([!#$%&*-~]|{nonascii}|{escape})*',
			's'					=>	'[ \t\r\n\f]',
			'w'					=>	'{s}*',
			'nl'				=>	'(\n|\r\n|\r|\f)',
		);
		$replaced_macros = array();
		foreach ($simple as $key => $macro) {
			$replaced = $macro;
			foreach ($replaced_macros as $shorthand => $replacement) {
				 $replaced = str_replace('{'.$shorthand.'}', $replacement, $replaced);
			}
			$replaced_macros[$key] = $replaced;
		}

		$this->Macros = $replaced_macros;
	}


	public function GetHTMLVisualPropsSelector($node)
	{
		if (!$node->Attributes) return false;
		$non_visal_props = array(
			'ABBR', 'ACCEPT-CHARSET', 'ACCEPT', 'ACCESSKEY', 'ACTION', 'ALT', 'ARCHIVE', 'AXIS', 'CHARSET', 'CHECKED', 'CITE', 'CLASS', 'CLASSID', 'CODE', 'CODEBASE',
			'CODETYPE', 'COLSPAN', 'COORDS', 'DATA', 'DATETIME', 'DECLARE', 'DEFER', 'DIR', 'DISABLED', 'ENCTYPE', 'FOR', 'HEADERS', 'HREF', 'HREFLANG', 'HTTP-EQUIV',
			'ID', 'ISMAP', 'LABEL', 'LANG', 'LANGUAGE', 'LONGDESC', 'MAXLENGTH', 'MEDIA', 'METHOD', 'MULTIPLE', 'NAME', 'NOHREF', 'OBJECT', 'ONBLUR', 'ONCHANGE',
			'ONCLICK', 'ONDBLCLICK', 'ONFOCUS', 'ONKEYDOWN', 'ONKEYPRESS', 'ONKEYUP', 'ONLOAD', 'ONLOAD', 'ONMOUSEDOWN', 'ONMOUSEMOVE', 'ONMOUSEOUT', 'ONMOUSEOVER',
			'ONMOUSEUP', 'ONRESET', 'ONSELECT', 'ONSUBMIT', 'ONUNLOAD', 'ONUNLOAD', 'PROFILE', 'PROMPT', 'READONLY', 'REL', 'REV', 'ROWSPAN', 'SCHEME', 'SCOPE',
			'SELECTED', 'SHAPE', 'SPAN', 'SRC', 'STANDBY', 'START', 'STYLE', 'SUMMARY', 'TITLE', 'USEMAP',
			'VALUE', 'VALUETYPE', 'VERSION',
		);
		if ($node->Name != 'LI' && $node->Name != 'OL' && $node->Name != 'UL') {
			array_push($non_visal_props, 'TYPE');
		}
		$visual_attributes = array_diff_key($node->Attributes, array_combine($non_visal_props, array_fill(0, count($non_visal_props), '')));
		if ($visual_attributes) {
			$mapping = array(
				'ALIGN' => 'TEXT-ALIGN',
				'VALIGN' => 'VERTICAL-ALIGN',
				'CELLPADDING' => 'PADDING',
			);
			$mapped_attributes = array();
			foreach ($visual_attributes as $key => $val) {
				if ($key == 'CELLPADDING') {
					$processed = $this->IdentifySelectors( array( $node->Name.'[cellpadding='.$val.'] TD' ), kPDFStylesheet::STYLE_ORIGIN_AUTHOR_NORMAL );
					$processed[0]['order'] = $this->HTMLVisualPropsSelectorOrder++;
					$processed[0]['specifity'] = 0;
					$this->Mapping['TD'][] = array(
						'selector' => $processed[0],
						'properties' => $this->ProcessShortHands(array(
							'PADDING' => $val.'px',
						)));
				}
				elseif (isset($mapping[$key])) {
					$mapped_attributes[$mapping[$key]] = $val;
				}
				else {
					$mapped_attributes[$key] = $val;
				}
			}

			return array(
				'selector' => array('main' => $node->Name, 'specifity' => 0, 'order' => $this->HTMLVisualPropsSelectorOrder, 'origin' => kPDFStylesheet::STYLE_ORIGIN_AUTHOR_NORMAL ),
				'properties' => $mapped_attributes,
			);
		}
		return false;
	}

	public function GetMatchingSelectors($node)
	{
		$map = isset($this->Mapping[$node->Name]) ? $this->Mapping[$node->Name] : array();
		if (isset($this->Mapping['*'])) {
			$map = array_merge($map, $this->Mapping['*']);
		}

		$matching = array();
		$i = 0;
		foreach ($map as $selector) {
			$selector_data = $selector['selector'];
			if ($this->SelectorMatches($selector['selector'], $node)) {
				$matching[] = $selector;
			}
		}

		$html_visual_selector = $this->GetHTMLVisualPropsSelector($node);
		if ($html_visual_selector) {
			$matching[] = $html_visual_selector;
		}

		usort($matching, array($this, 'CmpSelectors'));

		if (isset($node->Attributes['STYLE'])) {
			$style_selector = array(
				'selector' => array('main' => '_STYLE_'),
				'properties' => $this->ParseDefinitionTokens ( $this->GetTokens( $node->Attributes['STYLE'] ) ),
			);
			$matching[] = $style_selector;
		}

		return $matching;
	}

	public function CmpSelectors($a, $b)
	{
    if ($a['selector']['origin'] == $b['selector']['origin']) {
			if ($a['selector']['specifity'] == $b['selector']['specifity']) {
	        return $a['selector']['order'] < $b['selector']['order'] ? -1 : 1;
	    }
	    return ($a['selector']['specifity'] < $b['selector']['specifity']) ? -1 : 1;
    }
    return $a['selector']['origin'] < $b['selector']['origin'] ? -1 : 1;
	}

	public function SelectorMatches($selector_data, $node)
	{
		if ($selector_data['main'] != '*' && $node->Name != $selector_data['main']) {
			return false;
		}

		//check classes
		if (isset($selector_data['classes'])) {
			foreach ($selector_data['classes'] as $class) {
				// (\A| )+foo( |\Z)+
				if (!isset($node->Attributes['CLASS']) || !preg_match('/(\A| )+'.preg_quote($class).'( |\Z)+/i', $node->Attributes['CLASS'])) {
					return false;
				}
			}
		}

		//check ids
		if (isset($selector_data['ids'])) {
			if (!isset($node->Attributes['ID']) || !in_array($node->Attributes['ID'], $selector_data['ids'])) {
				return false;;
			}
		}

		//check atts
		if (isset($selector_data['atts'])) {
			if (isset($selector_data['atts']['set']))	{
				foreach ($selector_data['atts']['set'] as $att) {
					if (!isset($node->Attributes[$att])) {
						return false;;
					}
				}
			}
			if (isset($selector_data['atts']['equals']))	{
				foreach ($selector_data['atts']['equals'] as $att => $value) {
					if (!isset($node->Attributes[$att]) || strtoupper($node->Attributes[$att]) != strtoupper($value)) {
						return false;;
					}
				}
			}
			if (isset($selector_data['atts']['space']))	{
				foreach ($selector_data['atts']['space'] as $att => $value) {
					if (!isset($node->Attributes[$att]) || !preg_match('/(\A| )+'.preg_quote($value).'( |\Z)+/i', $node->Attributes[$att])) {
						return false;;
					}
				}
			}
			if (isset($selector_data['atts']['hypen']))	{
				foreach ($selector_data['atts']['hypen'] as $att => $value) {
					if (!isset($node->Attributes[$att]) || !preg_match('/^'.preg_quote($value).'(-|\Z)+/i', $node->Attributes[$att])) {
						return false;;
					}
				}
			}
		}

		//check pseudo
		if (isset($selector_data['pseudo_elements'])) {
			// we are not a browser - so don't know how to handle this....
			return false;
		}

		if (isset($selector_data['pseudo_classes'])) {
			// we are not a browser - so don't know how to handle this....
			return false;
		}

		//check comibantors
		if (isset($selector_data['child_of'])) {
			if (!$this->SelectorMatches($selector_data['child_of'], $node->Parent)) {
				return false;
			}
		}

		if (isset($selector_data['sibling_of'])) {
			if (!$this->SelectorMatches($selector_data['sibling_of'], $node->PrevSibling())) {
				return false;
			}
		}

		if (isset($selector_data['descendant_of'])) {
			$ancestor = $node;
			do {
				$ancestor = $ancestor->Parent;
				$matches = $this->SelectorMatches($selector_data['descendant_of'], $ancestor);
			} while (!$matches && $ancestor->Parent);
			if (!$matches) return false;
		}

		// if we came through here, the selector matches the node
		return true;
	}

	public function GetAllProperties($node)
	{
		$selectors = $this->GetMatchingSelectors($node);
		$properties = array();
		foreach ($selectors as $the_selector) {
			$properties = array_merge($properties, $the_selector['properties']);
			/*foreach ($the_selector['properties'] as $property => $value) {
				$properties[$property] = $value; //process !important here ?? !!!
			}*/
		}
		return $properties;
	}
}