class_editor_rte.php

来自「sabreipb 2.1.6 utf-8中文版本！」· PHP 代码 · 共 1,216 行 · 第 1/3 页

<?php

/*
+--------------------------------------------------------------------------
|   Invision Power Board v2.1.5
|   =============================================
|   by Matthew Mecham
|   (c) 2001 - 2005 Invision Power Services, Inc.
|   
|   =============================================
|   Web: 
|   Time: Wed, 01 Mar 2006 19:11:28 GMT
|   Release: 
|   Licence Info: 
+---------------------------------------------------------------------------
|   > $Date: 2006-01-30 18:53:19 +0000 (Mon, 30 Jan 2006) $
|   > $Revision: 127 $
|   > $Author: bfarber $
+---------------------------------------------------------------------------
|
|   > Posting RTE Editor
|   > Module written by Matt Mecham
|   > Date started: Thursday 10th March 2005 13:47
|
+--------------------------------------------------------------------------
*/

/**
* Text Editor: RTE (WYSIWYG) Class
*
* Class for parsing RTE
*
* @package		InvisionPowerBoard
* @subpackage	TextEditor
* @author  	 	Matt Mecham
* @version		2.1
* @since		2.1.0
*/

/**
*
*/

/**
* Text Editor: RTE (WYSIWYG) Class
*
* Class for parsing RTE
*
* @package		InvisionPowerBoard
* @subpackage	TextEditor
* @author  	 	Matt Mecham
* @version		2.1
* @since		2.1.0
*/
class class_editor_module extends class_editor
{
	/**
	* Main IPS class object
	*
	* @var	object
	*/
	var $ipsclass;
	
	/**
	* Clean up HTML on save
	*
	* @var	integer
	*/
	var $clean_on_save = 1;
	
	/**
	* Allow HTML
	*
	* @var	integer
	*/
	var $allow_html = 0;
	
	/**
	* Debug level
	*
	* @var	integer
	*/
	var $debug      = 0;
	
	/**
	* Parsing array
	*
	* @var	array
	*/
	var $delimiters     = array( "'", '"' );
	
	/**
	* Parsing array
	*
	* @var	array
	*/
	var $non_delimiters = array( "=", ' ' );
	
	/**
	* Start tags
	*
	* @var	string
	*/
	var $start_tags;
	
	/**
	* End tags
	*
	* @var	string
	*/
	var $end_tags;
	
	/**
	* Dunno, forgotten
	*
	* @var	integer
	*/
	var $_called = 0;
	
	/*-------------------------------------------------------------------------*/
	// Process the raw post with BBCode before showing in the form
	/*-------------------------------------------------------------------------*/
	
	/**
	* Process the raw post with BBCode before showing in the form
	*
	* @param	string	Raw text
	* @return	string	Converted text
	*/
	function process_before_form( $t )
	{
		//-----------------------------------------
		// Opening style attributes
		//-----------------------------------------
		
		$t = preg_replace( "#<!--sizeo:(.+?)-->(.+?)<!--/sizeo-->#"               , '<font size="\\1">'       , $t );
		$t = preg_replace( "#<!--coloro:(.+?)-->(.+?)<!--/coloro-->#"             , '<font color="\\1">'      , $t );
		$t = preg_replace( "#<!--fonto:(.+?)-->(.+?)<!--/fonto-->#"               , '<font face="\\1">'       , $t );
		$t = preg_replace( "#<!--backgroundo:(.+?)-->(.+?)<!--/backgroundo-->#"   , '<font background="\\1">' , $t );
		
		//-----------------------------------------
		// Closing style attributes
		//-----------------------------------------
		
		$t = preg_replace( "#<!--sizec-->(.+?)<!--/sizec-->#"            , "</font>" , $t );
		$t = preg_replace( "#<!--colorc-->(.+?)<!--/colorc-->#"          , "</font>" , $t );
		$t = preg_replace( "#<!--fontc-->(.+?)<!--/fontc-->#"            , "</font>" , $t );
		$t = preg_replace( "#<!--backgroundc-->(.+?)<!--/backgroundc-->#", "</font>" , $t );
			
		//-----------------------------------------
		// Remove comments
		//-----------------------------------------
		
		$t = preg_replace( "#\<\!\-\-(.+?)\-\-\>#is", "", $t );
		
		//-----------------------------------------
		// Return
		//-----------------------------------------
		
		$t = $this->_make_rtf_safe( $t );
		
		//-----------------------------------------
		// Clean up nbsp
		//-----------------------------------------
		
		# Doing so knocks out tabs
		//$t = str_replace( '&nbsp;&nbsp;&nbsp;&nbsp;', "\t", $t );
		//$t = str_replace( '&nbsp;&nbsp;'            , "  ", $t );
		
		return $t;
	}
	
	/*-------------------------------------------------------------------------*/
	// Process the raw post with BBCode before saving
	/*-------------------------------------------------------------------------*/
	
	/**
	* Process the raw post with BBCode before saving
	*
	* @param	string	Raw text
	* @return	string	Converted text
	*/
	function process_after_form( $form_field )
	{
		return $this->_rte_html_to_bbcode( $this->ipsclass->txt_stripslashes($_POST[ $form_field ]) );
	}
	
	/*-------------------------------------------------------------------------*/
	// Covert RTE-HTML to BBCode
	/*-------------------------------------------------------------------------*/
	
	/**
	* Covert RTE-HTML to BBCode
	*
	* @param	string	Raw text
	* @return	string	Converted text
	*/
	function _rte_html_to_bbcode( $t )
	{
		if ( $this->debug )
		{
			$ot = $t;
		}
		
		//-----------------------------------------
		// Fix up spaces
		//-----------------------------------------
		
		$t = str_replace( '&nbsp;', ' ', $t );
		
		//-----------------------------------------
		// Gecko engine seems to put \r\n at edge
		// of iframe when wrapping? If so, add a 
		// space or it'll get weird later
		//-----------------------------------------
		
		if ( $this->ipsclass->browser['browser'] == 'mozilla' OR $this->ipsclass->browser['browser'] == 'gecko' )
		{
			$t = str_replace( "\r\n", " ", $t );
		}
		else
		{
			$t = str_replace( "\r\n", "", $t );
		}

		//-----------------------------------------
		// Clean up already encoded HTML
		//-----------------------------------------
		
		$t = str_replace( '&quot;', '"', $t );
		$t = str_replace( '&apos;', "'", $t );
		
		//-----------------------------------------
		// Fix up incorrectly nested urls / BBcode
		//-----------------------------------------
		
		$t = preg_replace( '#<a\s+?href=[\'"]([^>]+?)\[(.+?)[\'"](.+?)'.'>(.+?)\[\\2</a>#is', '<a href="\\1"\\3>\\4</a>[\\2', $t );
		
		//-----------------------------------------
		// Make URLs safe (prevent tag stripping)
		//-----------------------------------------
		
		$t = preg_replace( '#<(a href|img src)=([\'"])([^>]+?)(\\3)#ise', "\$this->unhtml_url( '\\1', '\\3' )", $t );
		
		//-----------------------------------------
		// Make EMOids safe (prevent tag stripping)
		//-----------------------------------------
		
		$t = preg_replace( '#(emoid=\")(.+?)(\")#ise', "'emoid=\"'.\$this->unhtml_emoid('\\2').'\"';", $t );
		
		//-----------------------------------------
		// Remove tags we're not bothering with
		// with PHPs wonderful strip tags func
		//-----------------------------------------
		
		if ( ! $this->allow_html )
		{
			$t = strip_tags( $t, '<h1><h2><h3><h4><h5><h6><font><span><div><br><p><img><a><li><ol><ul><b><strong><em><i><u><strike><blockquote>' );
		}
		
		//-----------------------------------------
		// WYSI-Weirdness #0: BR tags to \n
		//-----------------------------------------
		
		$t = preg_replace( "#<br.*>#isU", "\n", $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #1: Named anchors
		//-----------------------------------------
		
		$t = preg_replace( "#<a\s+?name=.+?".">(.+?)</a>#is", "\\1", $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #1.5: Empty a hrefs
		//-----------------------------------------
		
		$t = preg_replace( "#<a\s+?href([^>]+)></a>#is"         , ""   , $t );
		$t = preg_replace( "#<a\s+?href=(['\"])>\\1(.+?)</a>#is", "\\1", $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #1.6: Double linked links
		//-----------------------------------------
		
		$t = preg_replace( "#href=[\"']\w+://(%27|'|\"|&quot;)(.+?)\\1[\"']#is", "href=\"\\2\"", $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #2: Headline tags
		//-----------------------------------------
		
		$t = preg_replace( "#<(h[0-9])>(.+?)</\\1>#is", "\n[b]\\2[/b]\n", $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #2.5: Fix up smilies
		//-----------------------------------------
		
		# Legacy Fix: Old smilies
		
		$t = str_replace( 'emoid=":""' , 'emoid="{{:&amp;quot;}}"', $t );
		$t = str_replace( 'emoid=":-""', 'emoid="{{:-&amp;quot;}}"', $t );
		
		# Make EMO_DIR safe so the ^> regex works
		$t = str_replace( "<#EMO_DIR#>", "&lt;#EMO_DIR&gt;", $t );
		
		# Parse...
		$t = preg_replace( "#[ ]?<([^>]+?)emoid=\"(.+?)\"([^>]+?)?".">[ ]?#ise", "' '.\$this->html_emoid('\\2').' '", $t );
		
		# And convert it back again...
		$t = str_replace( "&lt;#EMO_DIR&gt;", "<#EMO_DIR#>", $t );
		
		# Convert back
		$t = str_replace( "{{:&amp;quot;}}" , ':"' , $t );
		$t = str_replace( "{{:-&amp;quot;}}", ':-"', $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #3: Image tags
		//-----------------------------------------
		
		$t = preg_replace( "#<img.+?src=[\"'](.+?)[\"']([^>]+?)?".">#is", "[img]\\1[/img]", $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #4: Linked URL tags
		//-----------------------------------------
		
		$t = preg_replace( "#\[url=(\"|'|&quot;)<a\s+?href=[\"'](.*)/??['\"]\\2/??</a>#is", "[url=\\1\\2", $t );
	
		//-----------------------------------------
		// Now, recursively parse the other tags
		// to make sure we get the nested ones
		//-----------------------------------------
		
		$t = $this->_recurse_and_parse( 'b'          , $t, "_parse_simple_tag", 'b' );
		$t = $this->_recurse_and_parse( 'u'          , $t, "_parse_simple_tag", 'u' );
		$t = $this->_recurse_and_parse( 'strong'     , $t, "_parse_simple_tag", 'b' );
		$t = $this->_recurse_and_parse( 'i'          , $t, "_parse_simple_tag", 'i' );
		$t = $this->_recurse_and_parse( 'em'         , $t, "_parse_simple_tag", 'i' );
		$t = $this->_recurse_and_parse( 'strike'     , $t, "_parse_simple_tag", 's' );
		$t = $this->_recurse_and_parse( 'blockquote' , $t, "_parse_simple_tag", 'indent' );
		
		//-----------------------------------------
		// More complex tags
		//-----------------------------------------
		
		$t = $this->_recurse_and_parse( 'a'          , $t, "_parse_anchor_tag" );
		$t = $this->_recurse_and_parse( 'font'       , $t, "_parse_font_tag" );
		$t = $this->_recurse_and_parse( 'div'        , $t, "_parse_div_tag" );
		$t = $this->_recurse_and_parse( 'span'       , $t, "_parse_span_tag" );
		$t = $this->_recurse_and_parse( 'p'          , $t, "_parse_paragraph_tag" );
		
		//-----------------------------------------
		// Lists
		//-----------------------------------------
		
		$t = $this->_recurse_and_parse( 'ol'         , $t, "_parse_list_tag" );
		$t = $this->_recurse_and_parse( 'ul'         , $t, "_parse_list_tag" );
		
		//-----------------------------------------
		// WYSI-Weirdness #6: Fix up para tags
		//-----------------------------------------
		
		$t = preg_replace( "#<p.*>#isU", "\n\n", $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #7: Random junk
		//-----------------------------------------
		
		$t = preg_replace( "#(<a>|</a>|</li>)#is", "", $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #8: Fix up list stuff
		//-----------------------------------------
		
		$t = preg_replace( '#<li>(.*)((?=<li>)|</li>)#is', '\\1', $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #9: Convert rest to HTML
		//-----------------------------------------
		
		$t = str_replace(  '&lt;' , '<', $t );
		$t = str_replace(  '&gt;' , '>', $t );
		$t = str_replace(  '&amp;', '&', $t );
		$t = preg_replace( '#&amp;(quot|lt|gt);#', '&\\1;', $t );
		
		//-----------------------------------------
		// WYSI-Weirdness #10: Remove useless tags
		//-----------------------------------------
		
		while( preg_match( "#\[(url|img|b|u|i|s|email|list|indent|right|left|center)\]\[/\\1\]#is", $t ) )
		{
			$t = preg_replace( "#\[(url|img|b|u|i|s|email|list|indent|right|left|center)\]\[/\\1\]#is", "", $t );
		}
		
		//-----------------------------------------
		// Now call the santize routine to make
		// html and nasties safe. VITAL!!
		//-----------------------------------------
		
		$t = $this->_clean_post( $t );
		
		//-----------------------------------------
		// Debug?
		//-----------------------------------------
		
		if ( $this->debug )
		{
			print "<hr>";
			print nl2br(htmlspecialchars($ot));