<?


/*
** Functions for vulgar and spell checking of any string
**
*/


    /*
    **  count number of words in string
    */
    function SpaceCount($THEARTICLE) {

        // 1. strip HTML tags.
        $THEARTICLE = ereg_replace("<[a-zA-Z0-9/]{1,12}[^<]*>", " ", $THEARTICLE);
        // 2. strip duplicated spaces
        $THEARTICLE = ereg_replace("[ \,\.\n\t\r]+", " ", $THEARTICLE);
        // 3. strip whitespace from the start and the end of a string
        $THEARTICLE = trim($THEARTICLE);
        // 4. count words
        $SpaceCount = explode(" ",$THEARTICLE);
        return sizeof($SpaceCount);;
    }



	function CheckForLinks($THETEXT){
	
		return "";
		
	}
	
    /*
    ** check string for vulgar words
    ** vulgar words and patterns stored in AIV_VULGAR table.
    */
    function IsArticleVulgar($THEARTICLE) {

        global $DB, $DBID;

        $SQL = "select * from AIV_VULGAR where VW_STATUS='E'";
        $result = mysql_query($SQL, $DBID);

        $BadWordList = "";

        if ($result) {

            for ($i=0;$i<mysql_numrows($result);$i++) {
                $badword = mysql_result($result, $i, "VW_MASK");
                $badwordreply = mysql_result($result, $i, "VW_DESC");
                $ishtmltag = mysql_result($result, $i, "VW_IS_HTMLTAG");
                
                if( $ishtmltag=='Y' ){
                	$badword = "<[\n\r\t ]{0,}/{0,}[\n\r\t ]{0,}" . $badword;
                }
                
                $found = eregi($badword ,$THEARTICLE);
                
                if ($found){
                
                	$badwordreply = _hs($badwordreply);
                	
                	if( !eregi($badwordreply, $BadWordList) ){
	                    if( $BadWordList != ""  ){
	                        $BadWordList .= ", ";
	                    }
	                    $BadWordList .=  "&quot;" . _hs($badwordreply) . "&quot;";
                	}
                }
            }

        }

        return $BadWordList;
    }

    /*
    ** spell check the string
    ** we use standard *NIX utilities 'spell', 'sort' and 'more' to do the work.
    */
    function SpellCheck($THEARTICLE) {
        // 1. strip HTML tags.
        $THEARTICLE = ereg_replace("<[a-zA-Z0-9/]{1,12}[^<]*>", " ", $THEARTICLE);
        $THEARTICLE = ereg_replace("&quot;", " ", $THEARTICLE);
        $THEARTICLE = ereg_replace("&amp;", " ", $THEARTICLE);
        $THEARTICLE = ereg_replace("&lt;", " ", $THEARTICLE);
        $THEARTICLE = ereg_replace("&gt;", " ", $THEARTICLE);
        // 2. strip duplicated spaces
        $THEARTICLE = ereg_replace(" +", " ", $THEARTICLE);

        $tf = tempnam("/tmp", "AIVSPL");
        $currerrlvl = error_reporting( 0 );
        $tmpfile = fopen($tf, "a+");
        if( $tmpfile ){
            fwrite($tmpfile, $THEARTICLE);
            fclose($tmpfile);
        }
        error_reporting( $currerrlvl );

        $resString = "";

        //
        // now time we don't spell check!!!!
        //
        //
        //
        //
        unlink($tf);
        return $resString;
        //
        //
        //
        //
        //
        //


        $cmd = "/usr/bin/spell < $tf | /bin/sort -f | /usr/bin/uniq";

        $datastrim = popen($cmd, "r");

        // if the string correct $datastrin shud be empty
        while( !feof($datastrim) ){
            // collect all misspelled words
            $badspell = trim(fgets($datastrim, 1024));
            if( $badspell != "" ){
                if( $resString != "" ){
                    $resString .= ", ";
                }
                $resString .=  "&quot;" . _hs($badspell) . "&quot;";
            }
        }

        pclose($datastrim);
        unlink($tf);

        return $resString;
    }



?>