post_parser.php

来自「sabreipb 2.1.6 utf-8中文版本！」· PHP 代码 · 共 1,500 行 · 第 1/4 页

<?php

/*
+--------------------------------------------------------------------------
|   Invision Power Board v2.1.5
|   =============================================
|   by Matthew Mecham
|   (c) 2001 - 2005 Invision Power Services, Inc.
|   
|   =============================================
|   Web: 
|   Time: Wed, 01 Mar 2006 19:11:29 GMT
|   Release: 
|   Licence Info: 
+---------------------------------------------------------------------------
|   > $Date: 2005-10-10 14:08:54 +0100 (Mon, 10 Oct 2005) $
|   > $Revision: 23 $
|   > $Author: matt $
+---------------------------------------------------------------------------
|
|   > Text processor module
|   > Module written by Matt Mecham
|   > Official Version: 2.0 - Number of changes to date 3 billion (estimated)
|   > DBA Checked: Mon 24th May 2004
|
+--------------------------------------------------------------------------
*/


class post_parser {

	var $error            = "";
	var $image_count      = 0;
	var $emoticon_count   = 0;
	var $quote_html       = array();
	var $quote_open       = 0;
	var $quote_closed     = 0;
	var $quote_error      = 0;
	var $emoticons        = "";
	var $badwords         = "";
	var $strip_quotes     = "";
	var $in_sig           = "";
	var $allow_unicode    = 1;
	var $bypass_badwords  = 0;
	var $load_custom_tags = 0;
	var $pp_do_html       = 0;
	var $pp_nl2br         = 1;
	var $pp_wordwrap      = 0;
	var $max_embed_quotes = 10;
	
	/*-------------------------------------------------------------------------*/
	// CONSTRUCTOR
	/*-------------------------------------------------------------------------*/
	
	function post_parser($load=0)
	{
		$this->strip_quotes = $ibforums->vars['strip_quotes'];
		
		if ( $load )
		{
			$this->check_caches($load);
		}
	}
	
	/*-------------------------------------------------------------------------*/
	// CHECK (AND LOAD) CACHES
	/*-------------------------------------------------------------------------*/
	
	function check_caches($load=0)
	{
		global $ibforums, $std, $DB;
		$load=0;
		if ( ! is_array( $ibforums->cache['emoticons'] ) )
		{
			$ibforums->cache['emoticons'] = array();
			
			$DB->simple_construct( array( 'select' => 'typed,image,clickable,emo_set', 'from' => 'emoticons' ) );
			$DB->simple_exec();
		
			while ( $r = $DB->fetch_row() )
			{
				$ibforums->cache['emoticons'][] = $r;
			}
			
			if ( $load )
			{
				$std->update_cache( array( 'name' => 'emoticons', 'array' => 1, 'deletefirst' => 1 ) );
			}
		}
		
		if ( ! is_array( $ibforums->cache['bbcode'] ) )
		{
			$ibforums->cache['bbcode'] = array();
			
			$DB->simple_construct( array( 'select' => 'bbcode_id, bbcode_tag, bbcode_replace, bbcode_useoption', 'from' => 'custom_bbcode' ) );
			$bbcode = $DB->simple_exec();
		
			while ( $r = $DB->fetch_row($bbcode) )
			{
				$ibforums->cache['bbcode'][] = $r;
			}
			
			if ( $load )
			{
				$std->update_cache( array( 'name' => 'bbcode', 'array' => 1, 'deletefirst' => 1 ) );
			}
		}
		
		if ( ! is_array( $ibforums->cache['badwords'] ) )
		{
			$ibforums->cache['badwords'] = array();
			
			$DB->simple_construct( array( 'select' => 'type,swop,m_exact', 'from' => 'badwords' ) );
			$bbcode = $DB->simple_exec();
		
			while ( $r = $DB->fetch_row($bbcode) )
			{
				$ibforums->cache['badwords'][] = $r;
			}
			
			if ( $load )
			{
				$std->update_cache( array( 'name' => 'badwords', 'array' => 1, 'deletefirst' => 1 ) );
			}
		}
	}
	
	/*-------------------------------------------------------------------------*/
	// Strip quote tags
	/*-------------------------------------------------------------------------*/
	
	function strip_quote_tags( $txt="" )
	{
		return preg_replace( "#\[QUOTE(=.+?,.+?)?\].+?\[/QUOTE\]#is", "", $txt );
	}
	
	/*-------------------------------------------------------------------------*/
	// strip all tags
	/*-------------------------------------------------------------------------*/
	
	function strip_all_tags( $txt="" )
	{
		$txt = $this->strip_quote_tags( $this->unconvert( $txt ) );
		
		$txt = preg_replace( "#\[.+?\](.+?)\[/.+?\]#is", "\\1", $txt );
		
		return $txt;
	}
	
	/*-------------------------------------------------------------------------*/
	// strip all tags to formatted HTML
	/*-------------------------------------------------------------------------*/
	
	function strip_all_tags_to_formatted( $txt="" )
	{
		//$txt = $this->strip_quote_tags( $this->unconvert( $txt ) );
		
		//$txt = preg_replace( "#\[CODE\](.+?)\[/CODE\]#is", "<pre>\\1</pre>", $txt );
		
		//$txt = preg_replace( "#\[LIST\](.+?)\[/LIST\]#eis", "'<ul>' .str_replace( '[*]', '<li>', nl2br('\\1') ).'</ul>';", $txt );
		
		//$txt = preg_replace( "#\[LIST=.+?\](.+?)\[/LIST\]#eis", "'<ul>' .str_replace( '[*]', '<li>', nl2br('\\1') ).'</ul>';", $txt );
		
		//$txt = preg_replace( "#\[.+?\](.+?)\[/.+?\]#is", "\\1", $txt );
		
		return $txt;
	}
	
	/*-------------------------------------------------------------------------*/
	// PARSE POLL TAGS
	// Converts certain code tags for polling
	/*-------------------------------------------------------------------------*/
	
	function parse_poll_tags($txt)
	{
	
		// if you want to parse more tags for polls, simply cut n' paste from the "convert" routine
		// anywhere here.
	
		$txt = preg_replace( "#\[img\](.+?)\[/img\]#ie" , "\$this->regex_check_image('\\1')", $txt );
		
		$txt = preg_replace( "#\[url\](\S+?)\[/url\]#ie"                                       , "\$this->regex_build_url(array('html' => '\\1', 'show' => '\\1'))", $txt );
		$txt = preg_replace( "#\[url\s*=\s*\&quot\;\s*(\S+?)\s*\&quot\;\s*\](.*?)\[\/url\]#ie" , "\$this->regex_build_url(array('html' => '\\1', 'show' => '\\2'))", $txt );
		$txt = preg_replace( "#\[url\s*=\s*(\S+?)\s*\](.*?)\[\/url\]#ie"                       , "\$this->regex_build_url(array('html' => '\\1', 'show' => '\\2'))", $txt );
	
		return $txt;
	}
	
	/*-------------------------------------------------------------------------*/
	// convert:
	// Parses raw text into smilies, HTML and iB CODE
	/*-------------------------------------------------------------------------*/

	function convert($in=array( 'TEXT' => "", 'SMILIES' => 0, 'CODE' => 0, 'SIGNATURE' => 0, 'HTML' => 0))
	{
		global $ibforums, $DB;
		
		$this->check_caches();
		
		$this->in_sig = $in['SIGNATURE'];
		
		$txt = $in['TEXT'];
		
		//-----------------------------------------
		// Returns any errors as $this->error
		//-----------------------------------------
		
		// Remove session id's from any post
		
		$txt = preg_replace( "#(\?|&amp;|;|&)s=([0-9a-zA-Z]){32}(&amp;|;|&|$)?#e", "\$this->regex_bash_session('\\1', '\\3')", $txt );
		
		//-----------------------------------------
		// convert <br> to \n
		//-----------------------------------------
		
		$txt = preg_replace( "/<br>|<br \/>/", "\n", $txt );
		
		//-----------------------------------------
		// Are we parsing iB_CODE and do we have either '[' or ']' in the
		// text we are processing?
		//-----------------------------------------
		
		if ( $in['CODE'] == 1 )
		{
			//-----------------------------------------
			// Do [CODE] tag
			//-----------------------------------------
			
			$txt = preg_replace( "#\[code\](.+?)\[/code\]#ies", "\$this->regex_code_tag( '\\1' )", $txt );
		
			//-----------------------------------------
			// Auto parse URLs
			//-----------------------------------------
			
			$txt = preg_replace( "#(^|\s)((http|https|news|ftp)://\w+[^\s\[\]]+)#ie"  , "\$this->regex_build_url(array('html' => '\\2', 'show' => '\\2', 'st' => '\\1'))", $txt );
		
			//-----------------------------------------
			// Do [QUOTE(name,date)] tags
			//-----------------------------------------
			
			// Find the first, and last quote tag (greedy match)...
			
			$txt = preg_replace( "#(\[quote(.+?)?\].*\[/quote\])#ies" , "\$this->regex_parse_quotes('\\1')"  , $txt );
			
			/*-------------------------------------------------------------------------*/
			// If we are not parsing a siggie, lets have a bash
			// at the [PHP] [SQL] and [HTML] tags.
			/*-------------------------------------------------------------------------*/
			
			if ($in['SIGNATURE'] != 1) {
				
				$txt = preg_replace( "#\[sql\](.+?)\[/sql\]#ies"    , "\$this->regex_sql_tag('\\1')"    , $txt );
				$txt = preg_replace( "#\[html\](.+?)\[/html\]#ies"  , "\$this->regex_html_tag('\\1')"   , $txt );
				
				//-----------------------------------------
				// [LIST]    [*]    [/LIST]
				//-----------------------------------------
				
				while( preg_match( "#\n?\[list\](.+?)\[/list\]\n?#ies" , $txt ) )
				{
					$txt = preg_replace( "#\n?\[list\](.+?)\[/list\]\n?#ies", "\$this->regex_list('\\1')" , $txt );
				}
				
				while( preg_match( "#\n?\[list=(a|A|i|I|1)\](.+?)\[/list\]\n?#ies" , $txt ) )
				{
					$txt = preg_replace( "#\n?\[list=(a|A|i|I|1)\](.+?)\[/list\]\n?#ies", "\$this->regex_list('\\2','\\1')" , $txt );
				}
			}
			 
			//-----------------------------------------
			// Do [IMG] [FLASH] tags
			//-----------------------------------------
			
			if ($ibforums->vars['allow_images'])
			{
				$txt = preg_replace( "#\[img\](.+?)\[/img\]#ie"                             , "\$this->regex_check_image('\\1')"          , $txt );
				$txt = preg_replace( "#(\[flash=)(\S+?)(\,)(\S+?)(\])(\S+?)(\[\/flash\])#ie", "\$this->regex_check_flash('\\2','\\4','\\6')", $txt );
			}
		
		
			// Start off with the easy stuff
			
			$txt = preg_replace( "#\[b\](.+?)\[/b\]#is", "<b>\\1</b>", $txt );
			$txt = preg_replace( "#\[i\](.+?)\[/i\]#is", "<i>\\1</i>", $txt );
			$txt = preg_replace( "#\[u\](.+?)\[/u\]#is", "<u>\\1</u>", $txt );
			$txt = preg_replace( "#\[s\](.+?)\[/s\]#is", "<s>\\1</s>", $txt );
			
			// (c) (r) and (tm)
			
			$txt = preg_replace( "#\(c\)#i"     , "&copy;" , $txt );
			$txt = preg_replace( "#\(tm\)#i"    , "&#153;" , $txt );
			$txt = preg_replace( "#\(r\)#i"     , "&reg;"  , $txt );
			
			// email tags
			// [email][e-mail][/email]   [email=[e-mail]]Email me[/email]
			
			$txt = preg_replace( "#\[email\](\S+?)\[/email\]#i"                                                                , "<a href='mailto:\\1'>\\1</a>", $txt );
			$txt = preg_replace( "#\[email\s*=\s*\&quot\;([\.\w\-]+\@[\.\w\-]+\.[\.\w\-]+)\s*\&quot\;\s*\](.*?)\[\/email\]#i"  , "<a href='mailto:\\1'>\\2</a>", $txt );
			$txt = preg_replace( "#\[email\s*=\s*([\.\w\-]+\@[\.\w\-]+\.[\w\-]+)\s*\](.*?)\[\/email\]#i"                       , "<a href='mailto:\\1'>\\2</a>", $txt );
			
			// url tags
			// [url]http://www.index.com[/url]   [url=http://www.index.com]ibforums![/url]
			
			$txt = preg_replace( "#\[url\](\S+?)\[/url\]#ie"                                       , "\$this->regex_build_url(array('html' => '\\1', 'show' => '\\1'))", $txt );
			$txt = preg_replace( "#\[url\s*=\s*\&quot\;\s*(\S+?)\s*\&quot\;\s*\](.*?)\[\/url\]#ie" , "\$this->regex_build_url(array('html' => '\\1', 'show' => '\\2'))", $txt );
			$txt = preg_replace( "#\[url\s*=\s*(\S+?)\s*\](.*?)\[\/url\]#ie"                       , "\$this->regex_build_url(array('html' => '\\1', 'show' => '\\2'))", $txt );
			
			
			// font size, colour and font style
			// [font=courier]Text here[/font]  [size=6]Text here[/size]  [color=red]Text here[/color]
			
			while ( preg_match( "#\[size=([^\]]+)\](.+?)\[/size\]#ies", $txt ) )
			{
				$txt = preg_replace( "#\[size=([^\]]+)\](.+?)\[/size\]#ies"    , "\$this->regex_font_attr(array('s'=>'size','1'=>'\\1','2'=>'\\2'))", $txt );
			}
			
			while ( preg_match( "#\[font=([^\]]+)\](.*?)\[/font\]#ies", $txt ) )
			{
				$txt = preg_replace( "#\[font=([^\]]+)\](.*?)\[/font\]#ies"    , "\$this->regex_font_attr(array('s'=>'font','1'=>'\\1','2'=>'\\2'))", $txt );
			}
			
			while( preg_match( "#\[color=([^\]]+)\](.+?)\[/color\]#ies", $txt ) )
			{
				$txt = preg_replace( "#\[color=([^\]]+)\](.+?)\[/color\]#ies"  , "\$this->regex_font_attr(array('s'=>'col' ,'1'=>'\\1','2'=>'\\2'))", $txt );
			}
			
			
		}
		
		// Swop \n back to <br>
		
		$txt = preg_replace( "/\n/", "<br />", $txt );
		
		// Unicode?
		
		if ( $this->allow_unicode )
		{
			$txt = preg_replace("/&amp;#([0-9]+);/s", "&#\\1;", $txt );
		}
		
		//-----------------------------------------
		// Parse smilies (disallow smilies in siggies, or we'll have to query the DB for each post
		// and each signature when viewing a topic, not something that we really want to do.
		//-----------------------------------------
		
		if ($in['SMILIES'] != 0 and $in['SIGNATURE'] == 0)
		{
			$txt = ' '.$txt.' ';
		
			usort( $ibforums->cache['emoticons'] , array( 'post_parser', 'smilie_length_sort' ) );
			
			if ( count( $ibforums->cache['emoticons'] ) > 0 )
			{
				foreach( $ibforums->cache['emoticons']  as $a_id => $row)
				{
					if ( $row['emo_set'] != $this->ipsclass->skin['_emodir'] )
					{
						continue;
					}
					
					$code  = $row['typed'];
					$image = $row['image'];
					
					//-----------------------------------------
					// Make safe for regex
					//-----------------------------------------
					
					$code = preg_quote($code, "/");
					
					$txt = preg_replace( "!(?<=[^\w&;/])$code(?=.\W|\W.|\W$)!ei", "\$this->convert_emoticon('$code', '$image')", $txt );
				}
			}
			
			$txt = trim($txt);