Parser - first cut of restructuring of control logic

2025-08-01 20:30:39 +02:00 · 2009-01-03 22:32:54 +00:00
parent 5732bc1f4c
commit adce648182
1 changed files with 695 additions and 622 deletions
--- a/e107_handlers/e_parse_class.php
+++ b/e107_handlers/e_parse_class.php
@@ -1,20 +1,18 @@
 <?php
 /*
-+ ----------------------------------------------------------------------------+
-|     e107 website system
-|
-|      Steve Dunstan 2001-2002
-|     http://e107.org
-|     jalist@e107.org
-|
-|     Released under the terms and conditions of the
-|     GNU General Public License (http://gnu.org).
-|
-|     $Source: /cvs_backup/e107_0.8/e107_handlers/e_parse_class.php,v $
-|     $Revision: 1.47 $
-|     $Date: 2008-12-30 13:51:41 $
-|     $Author: secretr $
-+----------------------------------------------------------------------------+
+* e107 website system
+*
+* Copyright (C) 2001-2008 e107 Inc (e107.org)
+* Released under the terms and conditions of the
+* GNU General Public License (http://www.gnu.org/licenses/gpl.txt)
+*
+* Text processing and parsing functions
+*
+* $Source: /cvs_backup/e107_0.8/e107_handlers/e_parse_class.php,v $
+* $Revision: 1.48 $
+* $Date: 2009-01-03 22:32:54 $
+* $Author: e107steved $
+*
 */
 if (!defined('e107_INIT')) { exit; }

@@ -23,91 +21,130 @@ define('E_UTF8_PACK',e_HANDLER.'utf8/');		// Directory for the hard-coded utf-8
 define ("E_NL", chr(2));
 class e_parse
 {
+	var	$isutf8 = FALSE;	// Flag for global use indicates whether utf-8 character set
 	var	$utfAction;			// Determine how to handle utf-8.   0 = 'do nothing'    1 = 'use mb_string'    2 = emulation
-	var $e_sc;
-	var $e_bb;
-	var $e_pf;
-	var $e_emote;
-	var $e_hook;
+	var $e_sc;				// Shortcode processor
+	var $e_bb;				// BBCode processor
+	var $e_pf;				// Profanity filter
+	var $e_emote;			// Emote filter
+	var $e_hook;			// 'Hooked' parsers (array)
 	var $search = array('&#39;', '&#039;', '&quot;', 'onerror', '&gt;', '&amp;#039;', '&amp;quot;', ' & ');
 	var $replace = array("'", "'", '"', 'one<i></i>rror', '>', "'", '"', ' &amp; ');
 	var $e_highlighting;	// Set to TRUE or FALSE once it has been calculated
 	var $e_query;			// Highlight query

-		// toHTML Action defaults. For now these match existing convention.
-		// Let's reverse the logic on the first set ASAP; too confusing!
+		// Set up the defaults
 	var $e_optDefault = array(
-		'context' 		=> 'olddefault',	// default context: all "opt-out" conversions :(
+		'context' 		=> 'OLDDEFAULT',	// default context: reflects legacy settings (many items enabled)
 		'fromadmin' 	=> FALSE,
-
-		// Enabled by Default
+		'emotes'		=> TRUE,			// Enable emote display
+		'defs' 			=> FALSE,			// Convert defines(constants) within text.
+		'constants' 	=> FALSE,			// replace all {e_XXX} constants with their e107 value - 'rel' or 'abs'
+		'hook'			=> TRUE,			// Enable hooked parsers
+		'scripts'		=> TRUE,			// Allow scripts through (new for 0.8)
+		'link_click'	=> TRUE,			// Make links clickable
+		'link_replace'	=> TRUE,			// Substitute on clickable links (only if link_click == TRUE)
+		'parse_sc' 		=> FALSE,			// Parse shortcodes - TRUE enables parsing
+		'no_tags' 		=> FALSE,			// remove HTML tags.
 		'value'			=> FALSE,			// Restore entity form of quotes and such to single characters - TRUE disables

 		'nobreak' 		=> FALSE,			// Line break compression - TRUE removes multiple line breaks
-		'retain_nl' 	=> FALSE,			// Retain newlines - wraps to \n instead of <br /> if TRUE
-
-		'no_make_clickable' => FALSE,		// URLs etc are clickable - TRUE disables
-		'no_replace' 	=> FALSE,			// Replace clickable links - TRUE disables (only if no_make_clickable not set)
-
-	  	'emotes_off' 	=> FALSE,			// Convert emoticons to graphical icons - TRUE disables conversion
-		'emotes_on'  	=> FALSE,			// FORCE conversion to emotes, even if syspref is disabled
-
-		'no_hook' 		=> FALSE,			// Hooked parsers (TRUE disables completely) (deprecated)
-
-		// Disabled by Default
-		'defs' 			=> FALSE,			// Convert defines(constants) within text.
-		'constants' 	=> FALSE,			// replace all {e_XXX} constants with their e107 value
-		'abs_links' 	=> FALSE,			// Convert constants to absolute paths if TRUE
-		'parse_sc' 		=> FALSE,			// Parse shortcodes - TRUE enables parsing
-		'no_tags' 		=> FALSE			// remove HTML tags.
+		'retain_nl' 	=> FALSE			// Retain newlines - wraps to \n instead of <br /> if TRUE
 		);

-		// Super modifiers adjust default option values
-		// First line of adjustments change default-ON options
-		// Second line changes default-OFF options
+		// Super modifiers override default option values
 	var	$e_SuperMods = array(
 				'TITLE' =>				//text is part of a title (e.g. news title)
 					array(
-						'nobreak'=>TRUE, 'retain_nl'=>TRUE, 'no_make_clickable'=>TRUE,'emotes_off'=>TRUE,
-						'defs'=>TRUE,'parse_sc'=>TRUE),
+						'nobreak'=>TRUE, 'retain_nl'=>TRUE, 'link_click' => FALSE, 'emotes'=>FALSE, 'defs'=>TRUE, 'parse_sc'=>TRUE
+						),

 				'USER_TITLE' =>				//text is user-entered (i.e. untrusted) and part of a title (e.g. forum title)
 					array(
-						'nobreak'=>TRUE, 'retain_nl'=>TRUE, 'no_make_clickable'=>TRUE,'emotes_off'=>TRUE,'no_hook'=>TRUE
+						'nobreak'=>TRUE, 'retain_nl'=>TRUE, 'link_click' => FALSE, 'scripts' => FALSE, 'emotes'=>FALSE, 'hook'=>FALSE
+						),
+
+				'E_TITLE' =>					// text is 'body' of email or similar - being sent 'off-site' so don't rely on server availability
+					array(
+						'nobreak'=>TRUE, 'retain_nl'=>TRUE, 'defs'=>TRUE, 'parse_sc'=>TRUE, 'emotes'=>FALSE, 'scripts' => FALSE, 'link_click' => FALSE
 						),

 				'SUMMARY' =>			// text is part of the summary of a longer item (e.g. content summary)
 					array(
-						// no changes to default-on items
-						'defs'=>TRUE, 'constants'=>TRUE, 'parse_sc'=>TRUE),
+						'defs'=>TRUE, 'constants'=>'rel', 'parse_sc'=>TRUE
+						),

 				'DESCRIPTION' =>	// text is the description of an item (e.g. download, link)
 					array(
-						// no changes to default-on items
-						'defs'=>TRUE, 'constants'=>TRUE, 'parse_sc'=>TRUE),
+						'defs'=>TRUE, 'constants'=>'rel', 'parse_sc'=>TRUE
+						),

 				'BODY' =>					// text is 'body' or 'bulk' text (e.g. custom page body, content body)
 					array(
-						// no changes to default-on items
-						'defs'=>TRUE, 'constants'=>TRUE, 'parse_sc'=>TRUE),
+						'defs'=>TRUE, 'constants'=>'rel', 'parse_sc'=>TRUE
+						),

 				'USER_BODY' =>					// text is user-entered (i.e. untrusted)'body' or 'bulk' text (e.g. custom page body, content body)
 					array(
-						'constants'=>TRUE
+						'constants'=>TRUE, 'scripts' => FALSE
+						),
+
+				'E_BODY' =>					// text is 'body' of email or similar - being sent 'off-site' so don't rely on server availability
+					array(
+						'defs'=>TRUE, 'constants'=>'abs', 'parse_sc'=>TRUE, 'emotes'=>FALSE, 'scripts' => FALSE, 'link_click' => FALSE
 						),

 				'LINKTEXT' =>			// text is the 'content' of a link (A tag, etc)
 					array(
-						'nobreak'=>TRUE, 'retain_nl'=>TRUE, 'no_make_clickable'=>TRUE,'emotes_off'=>TRUE,'no_hook'=>TRUE,
-						'defs'=>TRUE,'parse_sc'=>TRUE),
+						'nobreak'=>TRUE, 'retain_nl'=>TRUE, 'link_click' => FALSE, 'emotes_on'=>FALSE, 'hook'=>FALSE, 'defs'=>TRUE, 'parse_sc'=>TRUE
+						),

 				'RAWTEXT' =>			// text is used (for admin edit) without fancy conversions or html.
 					array(
-						'nobreak'=>TRUE, 'retain_nl'=>TRUE, 'no_make_clickable'=>TRUE,'emotes_off'=>TRUE,'no_hook'=>TRUE,'no_tags'=>TRUE
-						// leave opt-in options off
+						'nobreak'=>TRUE, 'retain_nl'=>TRUE, 'link_click' => FALSE, 'emotes'=>FALSE, 'hook'=>FALSE, 'no_tags'=>TRUE
 						)
 		);

+	// Individual modifiers change the current context
+	var $e_Modifiers = array(
+				'emotes_off'	=> array('emotes_on' => FALSE),	
+				'emotes_on'		=> array('emotes_on' => TRUE),	
+				'no_hook'		=> array('hook' => FALSE),	
+				'do_hook'		=> array('hook' => TRUE),	
+				'scripts_off'	=> array('scripts' => FALSE),			// New for 0.8
+				'scripts_on'	=> array('scripts' => TRUE),			// New for 0.8
+				'no_make_clickable' => array('link_click' => FALSE),
+				'make_clickable' => array('link_click' => TRUE),
+				'no_replace' 	=> array('link_replace' => FALSE),
+				'replace' 		=> array('link_replace' => TRUE),		// Replace text of clickable links (only if make_clickable option set)
+				'consts_off'	=> array('constants' => FALSE),			// No path replacement
+				'consts_rel'	=> array('constants' => 'rel'),			// Relative path replacement
+				'consts_abs'	=> array('constants' => 'abs'),			// Absolute path replacement
+				'scparse_off'	=> array('parse_sc'	=> FALSE),			// No shortcode parsing
+				'scparse_on'	=> array('parse_sc'	=> TRUE),
+				'no_tags' 		=> array('no_tags' 	=> TRUE),			// Strip tags
+				'do_tags' 		=> array('no_tags' 	=> FALSE),			// Leave tags
+				'fromadmin'		=> array('fromadmin' => TRUE),
+				'notadmin'		=> array('fromadmin' => FALSE),
+				'er_off'		=> array('value' => FALSE),				// entity replacement
+				'er_on'			=> array('value' => TRUE),
+				'defs_off'		=> array('defs' => FALSE),				// Decode constant if exists
+				'defs_on'		=> array('defs' => TRUE),
+
+				'dobreak'		=> array('nobreak' => TRUE),	
+				'nobreak'		=> array('nobreak' => FALSE),
+				'lb_nl'			=> array('retain_nl' => TRUE),				// Line break using \n
+				'lb_br'			=> array('retain_nl' => FALSE),				// Line break using <br />
+
+				// Legacy option names below here - discontinue later
+				'retain_nl'		=> array('retain_nl' => TRUE),
+				'defs'			=> array('defs' => TRUE),
+				'parse_sc'		=> array('parse_sc'	=> TRUE),
+				'constants'		=> array('constants' => 'rel'),
+				'value'			=> array('value' => TRUE)
+		);
+
+
 	function e_parse()
 	{
 		// Preprocess the supermods to be useful default arrays with all values
@@ -125,7 +162,10 @@ class e_parse
 	{
 		// Start by working out what, if anything, we do about utf-8 handling.
 		$this->utfAction = 0;			// 'Do nothing' is the simple option
-		if ((strtolower(CHARSET) == 'utf-8') && (version_compare(PHP_VERSION, '6.0.0') < 1))
+		if (strtolower(CHARSET) == 'utf-8')
+		{
+			$this->isutf8 = TRUE;
+			if (version_compare(PHP_VERSION, '6.0.0') < 1)
 			{	// Need to do something here
 				if(extension_loaded('mbstring'))
 				{
@@ -144,6 +184,7 @@ class e_parse
 				}
 			}
 		}
+	}


 	function uStrLen($str)
@@ -170,6 +211,7 @@ class e_parse
 		return utf8_strtolower($str);
 	}

+
 	function uStrToUpper($str)
 	{
 		switch ($this->utfAction)
@@ -193,6 +235,7 @@ class e_parse
 		utf8_strpos($haystack, $needle, $offset);
 	}

+
 	function uStrrPos($haystack, $needle, $offset = 0)
 	{
 		switch ($this->utfAction)
@@ -248,12 +291,17 @@ class e_parse
 		* $mod: the 'no_html' and 'no_php' modifiers blanket prevent html and php posting regardless of posting permissions. (used in logging)
 		*/
 		global $pref;
-		if (is_array($data)) {
-			foreach ($data as $key => $var) {
+		if (is_array($data)) 
+		{
+			foreach ($data as $key => $var) 
+			{
 				$ret[$key] = $this -> toDB($var, $nostrip, $no_encode, $mod, $original_author);
 			}
-		} else {
-			if (MAGIC_QUOTES_GPC == true && $nostrip == false) {
+		} 
+		else 
+		{
+			if (MAGIC_QUOTES_GPC == true && $nostrip == false) 
+			{
 				$data = stripslashes($data);
 			}
 			if (isset($pref['post_html']) && check_class($pref['post_html']))
@@ -269,7 +317,9 @@ class e_parse
 				$search = array('$', '"', "'", '\\', '<?');
 				$replace = array('&#036;','&quot;','&#039;', '&#092;', '&lt;?');
 				$ret = str_replace($search, $replace, $data);
-			} else {
+			} 
+			else 
+			{
 				$data = htmlspecialchars($data, ENT_QUOTES, CHARSET);
 				$data = str_replace('\\', '&#092;', $data);
 				$ret = preg_replace("/&amp;#(\d*?);/", "&#\\1;", $data);
@@ -278,9 +328,7 @@ class e_parse
 			{
 				$ret = str_replace(array("[php]", "[/php]"), array("&#91;php&#93;", "&#91;/php&#93;"), $ret);
 			}
-
 		}
-
 		return $ret;
 	}

@@ -291,28 +339,33 @@ class e_parse
 		$search = array('&#036;', '&quot;', '<', '>');
 		$replace = array('$', '"', '&lt;', '&gt;');
 		$text = str_replace($search, $replace, $text);
-		if (e_WYSIWYG !== true){
+		if (e_WYSIWYG !== true)
+		{
 	   	  	$text = str_replace("&nbsp;", " ", $text); // fix for utf-8 issue with html_entity_decode();
 		}
 		return $text;
 	}


-	function post_toForm($text) {
-		if (MAGIC_QUOTES_GPC == true) {
+	function post_toForm($text) 
+	{
+		if (MAGIC_QUOTES_GPC == true) 
+		{
 			$text = stripslashes($text);
 		}
 		return str_replace(array( "'", '"', "<", ">"), array("&#039;", "&quot;", "&lt;", "&gt;"), $text);
 	}


-	function post_toHTML($text, $original_author = false, $extra = '', $mod = false) {
+	function post_toHTML($text, $original_author = false, $extra = '', $mod = false) 
+	{
 		$text = $this -> toDB($text, false, false, $mod, $original_author);
 		return $this -> toHTML($text, true, $extra);
 	}


-	function parseTemplate($text, $parseSCFiles = TRUE, $extraCodes = "") {
+	function parseTemplate($text, $parseSCFiles = TRUE, $extraCodes = "") 
+	{
 		$this->sch_load();
 		return $this->e_sc->parseCodes($text, $parseSCFiles, $extraCodes);
 	}
@@ -328,6 +381,8 @@ class e_parse
 		$nobreak is a list of tags within which word wrap is to be inactive
 		*/

+		return $str;
+
 		if (!ctype_digit($width)) return $str;		// Don't wrap if non-numeric width
 		if ($width < 6) return $str;					// Trap stupid wrap counts, as well
 	
@@ -593,19 +648,20 @@ class e_parse
 	}


+
 	function textclean ($text, $wrap=100)
 	{
 		$text = str_replace ("\n\n\n", "\n\n", $text);
 		$text = $this->htmlwrap($text, $wrap);
-		$text = str_replace (array ("<br /> ", " <br />", " <br /> "), "<br />", $text);
+		$text = str_replace (array ('<br /> ', ' <br />', ' <br /> '), '<br />', $text);
 		/* we can remove any linebreaks added by htmlwrap function as any \n's will be converted later anyway */
 		return $text;
 	}

-	//
+
+
 	// Test for text highlighting, and determine the text highlighting transformation
 	// Returns TRUE if highlighting is active for this page display
-	//
 	function checkHighlighting()
 	{
 		global $pref;
@@ -633,6 +689,8 @@ class e_parse
 	}


+
+
 	function toHTML($text, $parseBB = FALSE, $modifiers = "", $postID = "", $wrap=FALSE)
 	{
 		if ($text == '') return $text;
@@ -643,30 +701,38 @@ class e_parse
 		$opts = $this->e_optDefault;

 		// Now process any modifiers that are specified
-	  if (strlen($modifiers))
+		if ($modifiers)
 		{
 			$aMods = explode(',',$modifiers);

 			// If there's a supermodifier, it must be first, and in uppercase
 			$psm = trim($aMods[0]);
 			if (isset($this->e_SuperMods[$psm]))
-		{
-	  	  $opts = array_merge($this->e_optDefault,$this->e_SuperMods[$psm]);
+			{	// Supermodifier found - it simply overrides the default
+				$opts = $this->e_SuperMods[$psm];
 				$opts['context'] = $psm;
 				unset($aMods[0]);
 			}
-		else
-		{
-		// Set default modifiers
-		$opts = $this->e_optDefault;
-		}

-		// Now find any regular mods (could check each exists, but unnecessary processing really)
+
+			// Now find any regular modifiers; use them to modify the context (there should only be one or two out of the list of possibles)
 			foreach ($aMods as $mod)
 			{
-		  $opts[trim($mod)] = TRUE;  // Change mods as spec'd
+				$mod = trim($mod);					// Slight concession to varying coding styles
+				if (isset($this->e_Modifiers[$mod]))
+				{
+					foreach ($this->e_Modifiers[$mod] as $k => $v)		// This is probably quicker than array_merge - especially as usually only one or two loops
+					{
+						$opts[$k] = $v;				// Update our context-specific options
 					}
 				}
+			}
+		}
+
+		// Turn off a few things if not enabled in options
+		if (!varsettrue($pref['smiley_activate'])) $opts['emotes'] = FALSE;
+		if (!varsettrue($pref['make_clickable'])) $opts['link_click'] = FALSE;
+		if (!varsettrue($pref['link_replace'])) $opts['link_replace'] = FALSE;

 		$fromadmin = $opts['fromadmin'];

@@ -678,6 +744,7 @@ class e_parse
 		}


+
 		if ($opts['no_tags'])
 		{
 			$text = strip_tags($text);
@@ -685,8 +752,11 @@ class e_parse


 		// Make sure we have a valid count for word wrapping
-	  if(!$wrap && $pref['main_wordwrap']) $wrap = $pref['main_wordwrap'];
-	  $text = " ".$text;
+		if (!$wrap && $pref['main_wordwrap'])
+		{
+			$wrap = $pref['main_wordwrap'];
+		}
+//		$text = " ".$text;



@@ -781,14 +851,16 @@ class e_parse
 				$subcon = preg_split('#((?:<s)(?:cript[^>]+>.*?</script>|tyle[^>]+>.*?</style>))#mis', $full_text, -1, PREG_SPLIT_NO_EMPTY | PREG_SPLIT_DELIM_CAPTURE );
 				foreach ($subcon as $sub_blk)
 				{
-//			if (preg_match('#^<(script|style)[^>]+>#',$sub_blk))		//
-			if ((substr($sub_blk,0,7) == '<script') || (substr($sub_blk,0,6) == '<style'))
-			{  // Its a script/style block - just pass it through unaltered - except, do we need the line break stuff? - QUERY XXX-01
-			  if (DB_INF_SHOW) echo "Processing script: {$sub_blk}<br />";
-			  if (!$opts['nobreak'])
+					if (substr($sub_blk,0,7) == '<script')
 					{
-				$sub_blk = preg_replace("#>\s*[\r]*\n[\r]*#", ">", $sub_blk);
+						if ($opts['scripts'])
+						{
+							$ret_parser .= $sub_blk;	// Strip scripts unless permitted
 						}
+					}
+					elseif (substr($sub_blk,0,6) == '<style')
+					{  // Its a style block - just pass it through unaltered - except, do we need the line break stuff? - QUERY XXX-01
+						if (DB_INF_SHOW) echo "Processing stylesheet: {$sub_blk}<br />";
 						$ret_parser .= $sub_blk;
 					}
 					else
@@ -798,50 +870,44 @@ class e_parse
 	
 						// Could put tag stripping in here
 	
-
-				//	Line break compression (why?)
-				// Prepare for line-break compression. Avoid compressing newlines in embedded scripts and CSS
+/*
+						//	Line break compression - filter white space after HTML tags
 						if (!$opts['nobreak'])
 						{
 							$sub_blk = preg_replace("#>\s*[\r]*\n[\r]*#", ">", $sub_blk);
 						}
-
+*/	
 	
 						//	Link substitution
 						// Convert URL's to clickable links, unless modifiers or prefs override
-			  if ($pref['make_clickable'] && !$opts['no_make_clickable'])
+						if ($opts['link_click'])
 						{
-				if ($pref['link_replace'] && !$opts['no_replace'])
+							if ($opts['link_replace'])
 							{
 								$_ext = ($pref['links_new_window'] ? " rel=\"external\"" : "");
-//				  $sub_blk = preg_replace("#(^|[\n ])([\w]+?://[^ \"\n\r\t<,]*)#is", "\\1<a href=\"\\2\" {$_ext}>".$pref['link_text']."</a>", $sub_blk);
 								$sub_blk = preg_replace("#(^|[\s])([\w]+?://(?:[\w-%]+?)(?:\.[\w-%]+?)+.*?)(?=$|[\s()[\]<]|\.\s|\.$|,\s|,$)#is", "\\1<a href=\"\\2\" {$_ext}>".$pref['link_text']."</a>", $sub_blk);
-//				  $sub_blk = preg_replace("#(^|[\n \]])((www|ftp)\.[\w+-]+?\.[\w+\-.]*(?(?=/)(/.+?(?=\s|,\s))|(?=\W)))#is", "\\1<a href=\"http://\\2\" {$_ext}>".$pref['link_text']."</a>", $sub_blk);
 								$sub_blk = preg_replace("#(^|[\s])((?:www|ftp)(?:\.[\w-%]+?){2}.*?)(?=$|[\s()[\]<]|\.\s|\.$|,\s|,$)#is", "\\1<a href=\"http://\\2\" {$_ext}>".$pref['link_text']."</a>", $sub_blk);
-				  if(CHARSET != "utf-8" && CHARSET != "UTF-8")
+								if ($this->isutf8)
 								{
-					$email_text = ($pref['email_text']) ? $this->replaceConstants($pref['email_text']) : "\\1\\2&copy;\\3";
+									$email_text = ($pref['email_text']) ? $this->replaceConstants($pref['email_text']) : "\\1\\2©\\3";
 								}
 								else
 								{
-					$email_text = ($pref['email_text']) ? $this->replaceConstants($pref['email_text']) : "\\1\\2©\\3";
+									$email_text = ($pref['email_text']) ? $this->replaceConstants($pref['email_text']) : "\\1\\2&copy;\\3";
 								}
 								$sub_blk = preg_replace("#([\n ])([a-z0-9\-_.]+?)@([\w\-]+\.([\w\-\.]+\.)*[\w]+)#i", "\\1<a rel='external' href='javascript:window.location=\"mai\"+\"lto:\"+\"\\2\"+\"@\"+\"\\3\";self.close();' onmouseover='window.status=\"mai\"+\"lto:\"+\"\\2\"+\"@\"+\"\\3\"; return true;' onmouseout='window.status=\"\";return true;'>".$email_text."</a>", $sub_blk);
 							}
 							else
 							{
-//				  $sub_blk = preg_replace("#(^|[\n ])([\w]+?://[^ \"\n\r\t<,]*)#is", "\\1<a href=\"\\2\" rel=\"external\">\\2</a>", $sub_blk);
 								$sub_blk = preg_replace("#(^|[\s])([\w]+?://(?:[\w-%]+?)(?:\.[\w-%]+?)+.*?)(?=$|[\s()[\]<]|\.\s|\.$|,\s|,$)#is", "\\1<a href=\"\\2\" rel=\"external\">\\2</a>", $sub_blk);
-//				  $sub_blk = preg_replace("#(^|[\n \]])((www|ftp)\.[\w+-]+?\.[\w+\-.]*(?(?=/)(/.+?(?=\s|,\s))|(?=\W)))#is", "\\1<a href=\"http://\\2\" rel=\"external\">\\2</a>", $sub_blk);
 								$sub_blk = preg_replace("#(^|[\s])((?:www|ftp)(?:\.[\w-%]+?){2}.*?)(?=$|[\s()[\]<]|\.\s|\.$|,\s|,$)#is", "\\1<a href=\"http://\\2\" rel=\"external\">\\2</a>", $sub_blk);
 								$sub_blk = preg_replace("#([\n ])([a-z0-9\-_.]+?)@([\w\-]+\.([\w\-\.]+\.)*[\w]+)#i", "\\1<a rel='external' href='javascript:window.location=\"mai\"+\"lto:\"+\"\\2\"+\"@\"+\"\\3\";self.close();' onmouseover='window.status=\"mai\"+\"lto:\"+\"\\2\"+\"@\"+\"\\3\"; return true;' onmouseout='window.status=\"\";return true;'>".LAN_EMAIL_SUBS."</a>", $sub_blk);
 							}
 						}
 		
 		
-			   //	Emoticons
-				// Convert emoticons to graphical icons, unless modifiers override
-			  if (!$opts['emotes_off'] && ($pref['smiley_activate'] || $opts['emotes_on']))
+						// Convert emoticons to graphical icons, if enabled
+						if ($opts['emotes'])
 						{
 							if (!is_object($this->e_emote))
 							{
@@ -853,8 +919,7 @@ class e_parse



-			   //	Newline processing (more)
-			// Reduce multiple newlines in all forms to a single newline character, except for embedded scripts and CSS
+						// Reduce newlines in all forms to a single newline character (finds '\n', '\r\n', '\n\r')
 						if (!$opts['nobreak'])
 						{
 							$sub_blk = preg_replace("#[\r]*\n[\r]*#", E_NL, $sub_blk);
@@ -864,18 +929,18 @@ class e_parse
 		
 						//	Entity conversion
 						// Restore entity form of quotes and such to single characters, except for text destined for tag attributes or JS.
-			  if (!$opts['value'])
-			  { // output not used for attribute values.
-				$sub_blk = str_replace($this -> search, $this -> replace, $sub_blk);
-			  }
-			  else
+						if ($opts['value'])
 						{   									// output used for attribute values.
 							$sub_blk = str_replace($this -> replace, $this -> search, $sub_blk);
 						}
+						else
+						{ // output not used for attribute values.
+							$sub_blk = str_replace($this -> search, $this -> replace, $sub_blk);
+						}
+


 						//   BBCode processing (other than the four already done, which shouldn't appear at all in the text)
-				// Start parse [bb][/bb] codes
 						if ($parseBB !== FALSE)
 						{
 							if (!is_object($this->e_bb))
@@ -896,7 +961,6 @@ class e_parse
 								$sub_blk = $this->e_bb->parseBBCodes($sub_blk, $postID, 'default', $parseBB);		// Need to strip just some BBCodes
 							}
 						}
-			  // End parse [bb][/bb] codes



@@ -904,7 +968,7 @@ class e_parse
 						// (Moved to after bbcode processing by Cameron)
 						if ($opts['constants'])
 						{
-				$sub_blk = $this->replaceConstants($sub_blk, ($opts['abs_links'] ? 'full' : ''));
+							$sub_blk = $this->replaceConstants($sub_blk, ($opts['constants'] == 'abs' ? 'full' : ''));
 						}
 		
 		
@@ -931,7 +995,7 @@ class e_parse


 					  //Run any hooked in parsers
-			  if (!$opts['no_hook'])
+						if ($opts['hook'])
 						{
 							if ( varset($pref['tohtml_hook']))
 							{	//Process the older tohtml_hook pref (deprecated)
@@ -963,9 +1027,8 @@ class e_parse
 						}


-
 						// 	Word wrap
-			  if (!$opts['nobreak'])
+						if ($wrap && !$opts['nobreak'])
 						{
 							$sub_blk = $this -> textclean($sub_blk, $wrap);
 						}
@@ -973,8 +1036,7 @@ class e_parse
 		
 		
 						//	Search highlighting
-			  // Search Highlight
-			  if (!$opts['emotes_off'])
+						if ($opts['emotes'])			// Why??
 						{
 							if ($this->checkHighlighting())
 							{
@@ -983,8 +1045,7 @@ class e_parse
 						}
 		

-			// Purpose of this block?
-			  $nl_replace = "<br />";
+						$nl_replace = '<br />';			// Default replaces all \n with <br /> for HTML display
 						if ($opts['nobreak'])
 						{
 							$nl_replace = '';
@@ -1001,7 +1062,6 @@ class e_parse

 				}		// End of 'foreach() on each block of non-script text

-
 			}		// End of 'normal' parsing (non-script text)
 			else
 			{
@@ -1012,19 +1072,26 @@ class e_parse
 	}


-	function toAttribute($text) {
+
+	function toAttribute($text) 
+	{
 		$text = str_replace("&amp;","&",$text); // URLs posted without HTML access may have an &amp; in them.
 		$text = htmlspecialchars($text, ENT_QUOTES, CHARSET); // Xhtml compliance.
 		if (!preg_match('/&#|\'|"|\(|\)|<|>/s', $text))
 		{
 			$text = $this->replaceConstants($text);
 			return $text;
-		} else {
+		} 
+		else 
+		{
 			return '';
 		}
 	}

-	function toJS($stringarray) {
+
+	// Convert text blocks which are to be embedded within JS
+	function toJS($stringarray) 
+	{
 		$search = array("\r\n","\r","<br />","'");
 		$replace = array("\\n","","\\n","\'");
 		$stringarray = str_replace($search, $replace, $stringarray);
@@ -1036,9 +1103,10 @@ class e_parse
 		return strtr ($stringarray, $trans_tbl);
 	}

+
+
 	function toRss($text,$tags=FALSE)
 	{
-
 		if($tags != TRUE)
 		{
 			$text = $this -> toHTML($text,TRUE);
@@ -1101,12 +1169,14 @@ class e_parse
 									SITEURL.$THEMES_DIRECTORY,
 									SITEURL.$DOWNLOADS_DIRECTORY);
 			$search = array("{e_BASE}","{e_IMAGE_ABS}","{e_THEME_ABS}","{e_IMAGE}","{e_PLUGIN}","{e_FILE}","{e_THEME}","{e_DOWNLOAD}");
-			if (ADMIN) {
+			if (ADMIN) 
+			{
 				$replace_relative[] = $ADMIN_DIRECTORY;
 				$replace_absolute[] = SITEURL.$ADMIN_DIRECTORY;
 				$search[] = "{e_ADMIN}";
 			}
-			if ($all) {
+			if ($all) 
+			{
 				if (USER)
 				{  // Can only replace with valid number for logged in users
 					$replace_relative[] = USERID;
@@ -1131,6 +1201,7 @@ class e_parse
 		return $text;
 	}

+
 	function doReplace($matches)
 	{
 		if(defined($matches[1]) && ($matches[1] != 'e_ADMIN' || ADMIN))
@@ -1178,11 +1249,13 @@ class e_parse
    }


-	function e_highlight($text, $match) {
+	function e_highlight($text, $match) 
+	{
 		preg_match_all("#<[^>]+>#", $text, $tags);
 		$text = preg_replace("#<[^>]+>#", "<|>", $text);
 		$text = preg_replace("#(\b".$match."\b)#i", "<span class='searchhighlight'>\\1</span>", $text);
-		foreach ($tags[0] as $tag) {
+		foreach ($tags[0] as $tag) 
+		{
 			$text = preg_replace("#<\|>#", $tag, $text, 1);
 		}
 		return $text;