<html>
<head>यूनिकोड से DVBW-TTYogeshEN फ़ोण्ट परिवर्तित्र (28-2-2008 10:00)</title>

<script type="text/javascript">

function convert_to_YogeshEN()
{

var chunksize = document.getElementById("chunksize").value;

var array_one = new Array( 
"कृ","के","ङे","छे","टे","ठे","डे","ढे","दे","रे","ळे","हे",
"क़्","ख़्","ग़्","ज़्","फ़्","क़","ख़","ग़","ज़","फ़","ड़","ढ़", // two-byte varnas

"ॐ","ऽ","ः","।","् ", "फ़्र", "ऑ","ॉ","क्ष्","क्ष","त्र्","त्र","ज्ञ्","ज्ञ","श्र्","श्र","द्ध","द्य",

"ओं","औं","ओ","औ","आ","अ","ई","इ","ऊ","उ","ऋ","ऐ","ए",

"क्र","ट्र","ड्र","ढ्र", "्र",

"क़्","क्","क़","क","ख़्","ख्","ग़्","ग्","ग्र्","घ्","ङ",
"ख़","ख","ग़","ग","ग्र","घ",

"च्","छ","ज़्","ज्","झ्","ञ्",
"च","ज़","ज","झ","ञ",

"ट","ठ","ड","ढ","ण्","ण","ड़","ढ़",

"त्त्","त्त","त्","थ्","द","दृ","द्र","द्द","द्म","द्व","न्","ध्",
"त","थ","न","ध",

//"न्","ध्",

"प्","फ़्","फ़","फ्","फ","ब्","भ्","म्",
"प","ब","भ","म",

"य्","रु","रू","र","ल्","ळ्","ळ","व्",
"य","ल","व",

"श्","ष्","स्",
"श","ष","स",

"ट्ट","ट्ठ",

"ह्","ह","हृ","ह्र","ह्म","ह्य","्य",

"॰","्","़","़","ँ","ं",   //"्र","्र","्र",

// "Ô","Õ","×","Ø","Ù","Ú",

// "र्","र्ं","ि","िं","र्ि","र्िं",
//
//"à","á","â","ä","ç","ì","í","ð","ñ",
//"ीं","र्ी","र्ीं","्रु","्रू","र्े","र्ें","र्ै","र्ैं",

"ट्ठ","ड्ड","ड्ढ",

//"","","","","",

"ो","ौ","े","ै","ा","ी","ु","ू","ृ","ें","ैं","ॅ")

var array_two = new Array(

"Ûéú","Ûêú","’êû","”êû","™êü","šêü","›êü","œêü","¤êü","¸êü","ôêû","Æêü",
"क़्","ख़्","ग़्","ज़्","फ़्","क़","ख़","ग़","ज़","फ़","ड़","ढ़", // one-byte varnas

"‚","ƒ","„","…","Ë ","±ÏÌ","†Öò","Öò","õ","õÖ","¡","¡Ö","–","–Ö","Á","ÁÖ","¨","ª",

"†ÖêÓ","†ÖîÓ","†Öê","†Öî","†Ö","†","‡Ô","‡","‰ú","ˆ","Š","‹ê","‹",

"Îú","™Ò","›Ò","œÒ","Î",

"ŒÌ","Œ","Ìú","Ûú","Ì","Ü","Ì","Ý","Î","‘","’û",   // "Îú","Ðú","Ï"
"ÌÖ","ÜÖ","ÌÖ","ÝÖ","ÎÖ","‘Ö",

"“","”û","•Ì","•","—","˜",
"“Ö","•ÌÖ","•Ö","—Ö","˜Ö",

"™ü","šü","›ü","œü","Þ","ÞÖ","›Í","œÍ",

"¢","¢Ö","Ÿ","£","¤ü","¥","¦","§","©","«","®","¬",
"ŸÖ","£Ö","®Ö","¬Ö",

//"­","¬",

"¯","°Ì","±Ìú","°","±ú","²","³","´",
"¯Ö","²Ö","³Ö","´Ö",

"µ","¹ú","ºþ","¸ü","»","ó","ôû","¾",
"µÖ","»Ö","¾Ö",

"¿","Â","Ã",
"¿Ö","ÂÖ","ÃÖ",


"¼","½",

"Å","Æü","Ç","È","É","Ê","•",

".","Ë","Ì","Í","Ñ","Ó",    //"Ï","Ð","Ò",


"÷","ø","ù",

//"ú","û","ü","ý","þ",

"Öê","Öî","ê","î","Ö","ß","ã","æ","é","ë","ï","ò")

//**************************************************************************************
//"ú","û","ü","ý","þ",
//**************************************************************************************
//  To be edited above this line
//**************************************************************************************
// The following two characters are to be replaced through proper checking of locations:
//**************************************************************************************
// "र्" (reph) 
// "Ô" 
//
// "ि"  
// "×" 
//**************************************************************************************

var array_one_length = array_one.length ;


// if the input is plain text
if ((document.getElementById("text_or_html")).selectedIndex == 0 )  
   {
    document.getElementById("unicode_text").value = "You have chosen SIMPLE TEXT in Unicode to convert into DVBW-TTYogeshEN. Conversion in progress.."  ;  

    var modified_substring = document.getElementById("legacy_text").value  ;

//****************************************************************************************
//  Break the long text into small bunches of max. max_text_size  characters each.
//****************************************************************************************
    var text_size = document.getElementById("legacy_text").value.length ;

    var processed_text = '' ;  //blank

    var sthiti1 = 0 ;  var sthiti2 = 0 ;  var chale_chalo = 1 ;
 
    //var max_text_size = 6000;
 
    var max_text_size = chunksize;	

//  alert(max_text_size);

    while ( chale_chalo == 1 ) 
    {
     sthiti1 = sthiti2 ;

     if ( sthiti2 < ( text_size - max_text_size ) )  
     { 
      sthiti2 +=  max_text_size ;
      while (document.getElementById("legacy_text").value.charAt ( sthiti2 ) != ' ') {sthiti2--;}
     } 
     else  { sthiti2 = text_size  ;  chale_chalo = 0 }
   
 //    alert(" sthiti 1 = "+sthiti1); alert(" sthit 2 = "+sthiti2) 

     var modified_substring = document.getElementById("legacy_text").value.substring ( sthiti1, sthiti2 )  ;

    Replace_Symbols( ) ;

  //  alert(" modified substring = "+modified_substring)

    processed_text += modified_substring ;

 //   alert(" modified substring2 = "+modified_substring)

//****************************************************************************************
//  Breaking part code over
//****************************************************************************************

    document.getElementById("unicode_text").value = processed_text  ;
   }

}
// if input is HTML then

else    
   {
    document.getElementById("unicode_text").value = "You have chosen HTML TEXT in Unicode to convert into DVBW-TTYogeshEN. Conversion in progress.."  ;  

    var remaining_text = document.getElementById("legacy_text").value ;
    var processed_text = "" ;  //blank initially

    var position_of_current_opening_bracket = 0 ;
    var position_of_next_closing_bracket = 1 ;

    // search starting from paragraphs. then search for sanskrit 99.
    var idx = remaining_text.indexOf( "<p " )  
    idx = remaining_text.indexOf( "Krutidev010" , idx )
    var idx2 = 0  //  position_of_current_opening_bracket  ">"
    var idx3 = 0  //  position of "/span"
    var idx4 = 0  //  postion of "span" only , ie span without "/"

    while ( idx != -1 )    // while-01 loop     ;  while "Krutidev010"  is found..
       {  
        idx2 = remaining_text.indexOf( ">" , idx )  
        idx3 = remaining_text.indexOf( "/span" , idx2 )
        idx4 = remaining_text.indexOf( "span" , idx2 )

        while ( idx4 < idx3 )     // this loop to take care of  nested span.
           {
            idx4 = remaining_text.indexOf( "span" , idx3+4 ) ;
            idx3 = remaining_text.indexOf( "/span" , idx3+4 ) ;
           }

        var modified_substring = remaining_text.substring( idx2 , idx3 ) ;
        modified_substring = modified_substring.replace( />/g , ">>" ) ; 
        processed_text = processed_text + remaining_text.substring( 0 , idx2 ) + modified_substring + "/span" ;   

        remaining_text = remaining_text.substring( idx3 + 5 ) ;  //remaining_text excludes "/span"
        idx = remaining_text.indexOf( "Krutidev010" ) ;
       } //end of outermost while-01

    processed_text = processed_text + remaining_text ;

    //-----------------------------

    //This section for taking care of paragraphs marked class = MsoBodyText  or class = MsoBodyText

    remaining_text = processed_text ;
    processed_text = "" ;  //blank initially

    var position_of_start_of_paragraph = 0 ;
    var position_of_end_of_paragraph = 1 ;

    position_of_start_of_paragraph = remaining_text.indexOf("<p ")   ;

    while ( position_of_start_of_paragraph != -1 )  //search for <p  in the remaining_text
       {
        position_of_start_of_paragraph = remaining_text.indexOf("<p ")   ;
        position_of_end_of_paragraph = remaining_text.indexOf("/p>")   ;

        modified_substring = remaining_text.substring( position_of_start_of_paragraph + 3 ,  position_of_end_of_paragraph ) ;

        if (modified_substring.indexOf( "MsoBodyText" ) != -1 )
           { 
            // repace all ">" with ">>" in this paragraph
            modified_substring = modified_substring.replace( />/g , ">>" )  ;    

            // in Mybodytext, wherever font-family is found, it means it is roman text.
            idx = modified_substring.indexOf("font-family")  ;  

            idx2 = 0 ;  // position of ">>"
            idx3 = 0 ;  // position of "/span"
            idx4 = 0 ;  // position of "span" only without "/"

            // again change ">>" to ">" only those which occur immediately after font-family:"Times New Roman"
            while ( idx != -1 )     
               {
                idx2 = modified_substring.indexOf( ">>" , idx ) ;
                idx3 = modified_substring.indexOf( "/span" , idx2 ) ;
                idx4 = modified_substring.indexOf( "span" , idx2 ) ;

                while ( idx4 < idx3 )
                   {
                    idx4 = modified_substring.indexOf( "span" , idx3+4 )  ;
                    idx3 = modified_substring.indexOf( "/span" , idx3+4 )  ;
                   }

                modified_substring = modified_substring.substring( 0, idx2 ) + (modified_substring.substring( idx2 , idx3 )).replace( />>/g , ">" ) + modified_substring.substring( idx3 ) ; 
                idx = modified_substring.indexOf( "font-family" , idx3 ) ;

               } // end of while inner loop
           } // end of if statement

        processed_text = processed_text + remaining_text.substring( 0 , position_of_start_of_paragraph + 3 ) +  modified_substring + "/p>" ;   
        remaining_text = remaining_text.substring( position_of_end_of_paragraph + 3 ) ; 
        position_of_start_of_paragraph = remaining_text.indexOf("<p ")   ;
       } // end of outer while loop

    processed_text = processed_text + remaining_text ;

    //------------------------

    // Now do actual font conversion  of text occuring between  all the  pairs  >>  and  <

    remaining_text = processed_text ;     processed_text = "" ; 

    idx2 = remaining_text.indexOf( ">>" )  ;

    while ( idx2 != -1 )    // while-01 loop (checks if ">>" is still present
       {  
        position_of_current_opening_bracket = remaining_text.indexOf( ">>" , idx2 )  
        position_of_next_closing_bracket = remaining_text.indexOf( "<" , position_of_current_opening_bracket )

        modified_substring = remaining_text.substring( position_of_current_opening_bracket + 2 , position_of_next_closing_bracket ) ;

        processed_text = processed_text + remaining_text.substring( 0 , position_of_current_opening_bracket + 1 ) ;   // ">" included by using +1 here
        remaining_text = remaining_text.substring( position_of_next_closing_bracket + 1 ) ;  //remaining_text excludes the closing bracket

        // call the subroutine and replace the legacy symbols with corresponding Unicode.
     
        Replace_Symbols() ;   

        processed_text = processed_text + modified_substring + "<"  ;

        idx2 = remaining_text.indexOf( ">>" ) ;
       } //end of outermost while-01

    processed_text = processed_text + remaining_text ; 

    // change  font name from 'Mangal'  to  'Krutidev010' .
 
    processed_text = processed_text.replace( /Mangal/g , "Krutidev010" ) ;   

    // now put the processed text in the output box finally.

    document.getElementById("unicode_text").value = processed_text  

   } // end of else loop for HTML case

//**************************************************
//--------------------------------------------------
//**************************************************

function Replace_Symbols( )
   {


    // if string to be converted is non-blank then no need of any processing.
    if (modified_substring != "" )  
       {
//"क़्","ख़्","ग़्","ज़्","फ़्","क़","ख़","ग़","ज़","फ़","ड़","ढ़", // two-byte nuqta_varnas
// "क़्","ख़्","ग़्","ज़्","फ़्","क़","ख़","ग़","ज़","फ़","ड़","ढ़", // one-byte nuqta_varnas

// first replace the two-byte nuqta_varNas with corresponding one-byte nukta varNas.

modified_substring = modified_substring.replace ( /क़/ , "क़" )  ; 
modified_substring = modified_substring.replace ( /ख़‌/g , "ख़" )  ;
modified_substring = modified_substring.replace ( /ग़/g , "ग़" )  ;
modified_substring = modified_substring.replace ( /ज़/g , "ज़" )  ;
modified_substring = modified_substring.replace ( /ड़/g , "ड़" )  ;
modified_substring = modified_substring.replace ( /ढ़/g , "ढ़" )  ;
modified_substring = modified_substring.replace ( /ऩ/g , "ऩ" )  ;
modified_substring = modified_substring.replace ( /फ़/g , "फ़" )  ;
modified_substring = modified_substring.replace ( /य़/g , "य़" )  ;
modified_substring = modified_substring.replace ( /ऱ/g , "ऱ" )  ;
// "र्" (reph) 
// "Ô" 
//
// "ि"  
// "×"  

        // code for replacing "ि" (chhotee ee kii maatraa) with "×" and correcting its position too.
        
        var position_of_f = modified_substring.indexOf( "ि" )  ;
         while ( position_of_f != -1 )  //while-02
           {
            var character_left_to_f = modified_substring.charAt( position_of_f - 1 )  ;
            modified_substring = modified_substring.replace( character_left_to_f + "ि" ,  "×" + character_left_to_f )  ;

            position_of_f = position_of_f - 1  ;

            while (( modified_substring.charAt( position_of_f - 1 ) == "्" )  &  ( position_of_f != 0  ) )
               {
                var string_to_be_replaced = modified_substring.charAt( position_of_f - 2 ) + "्"  ;
                  modified_substring = modified_substring.replace( string_to_be_replaced + "×", "×" + string_to_be_replaced ) ;

                position_of_f = position_of_f - 2  ;
               }
            position_of_f = modified_substring.search( /ि/ , position_of_f + 1 ) ; // search for f ahead of the current position.

           } // end of while-02 loop
   //************************************************************     
   //     modified_substring = modified_substring.replace( /fर्/g , "£"  )  ;
   //************************************************************     
        // Eliminating "र्" and putting  "Ô"   at proper position for this.

       set_of_matras = "ािीुूृेैोौं:ँॅ" 

modified_substring += '  '    ;  // add two spaces after the string to avoid UNDEFINED char in the following code.

       var position_of_half_R = modified_substring.indexOf( "र्" ) ;
        while ( position_of_half_R > 0  )  // while-04
           {
            // "र्"  is two bytes long
            var probable_position_of_Z = position_of_half_R + 2   ;  

            var character_right_to_probable_position_of_Z = modified_substring.charAt( probable_position_of_Z + 1 )

            // trying to find non-maatra position right to probable_position_of_Z .

            while ( set_of_matras.indexOf( character_right_to_probable_position_of_Z ) != -1 )  
               {
                probable_position_of_Z = probable_position_of_Z + 1 ;
                character_right_to_probable_position_of_Z = modified_substring.charAt( probable_position_of_Z + 1 ) ;
               } // end of while-05

            string_to_be_replaced = modified_substring.substr ( position_of_half_R + 2 , ( probable_position_of_Z - position_of_half_R - 1 ))  ;
            modified_substring = modified_substring.replace( "र्" + string_to_be_replaced  ,  string_to_be_replaced + "Ô" ) ;
            position_of_half_R = modified_substring.indexOf( "र्" ) ;
           } // end of while-04


modified_substring = modified_substring.substr ( 0 , modified_substring.length - 2 )  ;



        //substitute array_two elements in place of corresponding array_one elements

        for( input_symbol_idx = 0; input_symbol_idx < array_one_length; input_symbol_idx++ )
           {
            idx = 0  ;  // index of the symbol being searched for replacement

            while (idx != -1 ) //whie-00
               {
                modified_substring = modified_substring.replace( array_one[ input_symbol_idx ] , array_two[input_symbol_idx] )
                idx = modified_substring.indexOf( array_one[input_symbol_idx] )
               } // end of while-00 loop
           } // end of for loop

        } // end of IF  statement  meant to  supress processing of  blank  string.

    } // end of the function  Replace_Symbols( )

  } // end of Convert_Unicode_to_YogeshEN  function


</script>
</head>

<!-- ----------------------------------------------------------------------------------
      body of the HTML starts here. one text box is provided each for input and output. 
-->

<body>
<form name="form1">

Put the UNICODE text (plain text or HTML text) in the box below : <br/>
Enter chunk size: <input type="text" name="chunksize" value="6000" size="7" maxsize="7" style="text-align:right"><br/><br/>

<textarea name="TextToConvert" id="legacy_text" cols="80" rows="7"></textarea> <br>

The input given in the above box is : 
<div > <select id="text_or_html">
<option>Plain TEXT</option>
<option>HTML TEXT</option>
</select> </div> 

<div align="middle">
<input  type="button" id="converter" name="converter" value=" Convert to DVBW-TTYogeshEN font >> " onClick="convert_to_YogeshEN();" accesskey="c" title="शॉर्टकट alt+c"> 
</div>  <br>

<div id="hiddenLayer"></div><br>
The above text converted to DVBW-TTYogeshEN font is shown in the box below : <br/>
<textarea name="ConvertedText" id="unicode_text" cols="100" rows="13"></textarea>
<br /><br />

</form>
</body>

</html>