tag

   if (preg_match("/<\s*pre\s*>/i",$wk)){
		$temp="";
		while (preg_match("/^(.+?)<\s*pre\s*>(.+?)<\s*\/\s*pre\s*>(.+)$/si",$wk,$m) ){
			$beg=$m[1];$pre=$m[2];$aft=$m[3];
			$pre=preg_replace("/ /","{{s}}",$pre);
			$pre=preg_replace("\t/","{{t}}",$pre);
			$pre=preg_replace("\n/","{{n}}",$pre);
			$temp.=$beg.$pre; $wk=$aft;
		}
		$wk=$temp;
	}

   #### Ditch chunks

   #preg_match("/<\s*title\s*>\s*(.+?)\s*<\s*\/\s*title\s*>/is",$wk,$m);$title=$m[1]; # keep title
   $wk=preg_replace("/^.+<\s*\/\s*head\s*>/si","",$wk);                              # ditch: head
   $wk=preg_replace("/<\s*style.*?>.*?<\s*\/\s*style\s*>/si","",$wk);               # style defns
   $wk=preg_replace("/<\s*script.*?>.+?<\s*\/\s*script\s*>/si","",$wk);             # scripts
   $wk=preg_replace("/[\t\n ]+/"," ",$wk);                                          # newlines, spaces, tabs

   #### Transfer tag information

   $wk=preg_replace("/ *< *a +href *= *\"?([^\"> ]+)\"?.*?> */i"," $1 ",$wk);          # retain urls
   $wk=preg_replace("/ *< *(br|table).*?> */i","\n",$wk);                           # various \n triggers
   $wk=preg_replace("/ *< *\/ *(dd|dt|li|tr).*?> */i","\n",$wk);                    #      "
   $wk=preg_replace("/ *< *\/?(dl|ol|ul|hr|p|h\d|blockquote).*?> */i","\n\n",$wk);  #      "
   $wk=preg_replace("/ *< *\/td *> */i","\t",$wk);                                  # crude table coherance

   #### Map (some of the) special chars

   $a=array('nbsp','quot','lt','gt','copy','deg','sup1','sup2','sup3','frac14','frac12','frac34','pound','cent','yen');
   $b=array(" ",'<','>','©','°','¹','²','³','¼','½','¾','£','¢','¥');
   for($c=0;$c<14;$c++){ $find=$a[$c];$repl=$b[$c];$wk=preg_replace("/&$find;/",$repl,$wk); }

  	$wk=preg_replace("/&#(151);/e",ord(151),$wk);                    # change — format
   $wk=preg_replace("/&#(149);/e",ord(149),$wk);                    # change — format
   $wk=preg_replace("/&#(9);/e","    ",$wk);                    	  # change — format

  	#### Clean up

   $wk=preg_replace("/<.+?>/s","",$wk);                        # ditch all tags
   $wk=preg_replace("/(\n[\t ]+|[\t ]+\n)/","\n",$wk);         # stray whitespace
   $wk=preg_replace("/(\n[\t ]+|[\t ]+\n)/","\n",$wk);         # stray whitespace
   $wk=preg_replace("/\n\n+/","\n\n",$wk);                     # ditto
   $wk=preg_replace("/{{s}}/"," ", $wk);
   $wk=preg_replace("/{{t}}/","\t",$wk);
   $wk=preg_replace("/{{n}}/","\n",$wk);    							# pre formating
   $wk=preg_replace("/^\s+/","",$wk);
   $wk=preg_replace("/\s+$/","",$wk);                       	# whitespace top and end

   return($wk);
}




function mysafehtml($wk){
	#removes any suspect html

	$wk= preg_replace("/<\s*appl[^>]+>.+?<\s*\/appl[^>]+>/si", "", $wk);
	$wk= preg_replace("/<\s*embb[^>]+>.+?<\s*\/embb[^>]+>/si", "", $wk);
	$wk= preg_replace("/<\s*scri[^>]+>.+?<\s*\/scri[^>]+>/si", "", $wk);
	$wk= preg_replace("/<\s*obje[^>]+>.+?<\s*\/obje[^>]+>/si", "", $wk);

	$wk= preg_replace("/<\s*meta[^>]+>/si", "", $wk);
	$wk= preg_replace("/",$wk); #comments, ssi
	$wk= preg_replace("/(<[^>]+) on[\w]+ ?= ?\"[^\" >]+\"([^>]*)>!--.+?>/si","$1 tag=stripped $2",$wk); #event handlers
	$wk= preg_replace("/( src\s*=\s*\"?)http[^\" >]+/si","$1"."ext-img.gif",$wk); #external images
	$wk= preg_replace ("'<[?%]=? [^>]+?>'si", "" ,$wk); #php and asp

	return($wk);
}

function make7bit ($body){
	// removes any 8 bit chars suitable for email

	$body=preg_replace("/\r\n/","\n",$body);
	$body=preg_replace("/[\r\cM]/","",$body);
	$body=preg_replace("/[^\x20-\x7E\n\t]/","§",$body);

	return($body);
}


function tidyup($wk){
	// cleans up stray white space
	$wk =  preg_replace("/[ \t]+\n/", "\n", $wk);
	$wk =  preg_replace("/\n[ \t]+\n/", "\n\n", $wk);
	$wk =  preg_replace("/\n\n+/", "\n\n", $wk);
	$wk=trim($wk);
	return($wk);
}


function sanitize($wk){

	## One off sift of any new form data
	## this is a matched pair to htmlize()


	// delete any suspect chars ie control chars, nulls, etc
	#$wk=preg_replace("/[\x00-\x08\x0B-\x0C\x0E-\x1F]/","",$wk);  //ditch ctrl chars
	#$wk=preg_replace("/[\x7f]/","",$wk);								// ?

	//make some judicous char conversions for the sake of sanity
   $wk=preg_replace("/\t/","    ",$wk);                        # tab
  	$wk=preg_replace("/[\x93\x94]/","\"",$wk);                    # smart double quotes
   $wk=preg_replace("/[\x91\x92]/","'",$wk);                     # smart single quotes
  	$wk=preg_replace("/[\x95]/","*",$wk);                 		  # bullet
   $wk=preg_replace("/[\x85]/","...",$wk);                    	  # ellipsis
   $wk=preg_replace("/[\x86\x87]/","--",$wk);                    # m and n dashes
   $wk=preg_replace("/[\xA0]/"," ",$wk);                         # nbsp

	// manage quotes and magic quoting
	$wk= preg_replace ("/\\\([\'\"\\\])/","\\1",$wk); // ditch backslashes put there by magic quoting
	$wk= preg_replace ("/['\"]/","`",$wk);				 // solve sql single quote hassles simply by using `s
	$wk= preg_replace ("/\\\/","/",$wk);				 // convert any stray backslashes

	// once and for all time rid any dos newlines
	$wk=preg_replace("/\r\n/","\n",  $wk);
	$wk=preg_replace("/[\r\cM]/","", $wk);

	// Sift any undesirable html
	$wk= mysafehtml($wk);

	//worst case over length
	if (strlen($wk)>100000){bail("Form variable exceeds 100K");}

	//tidy
	$wk= tidyup($wk);

	return ($wk);
}


function htmlize($wk){
	## gets plain text ready for placing web page
	## adds br tags and converts <>"& etc to avoid problems

	// translate key html chars
	$wk=	htmlspecialchars($wk); 			//avoids [<>"&] stuffing up the webpage, which dismembers any html tags

	// convert plural whitespace
	$wk =  preg_replace("/  /", "  ", $wk);

	// convert newlines
	$wk=	trim ($wk);
	$wk = preg_replace ("/\n/", "
", $wk);

	return($wk);
}

?>