<html>
<head>यूनिकोड से Richa फ़ोण्ट परिवर्तित्र (28-04-2008 23:43) </title>


<script type="text/javascript">

function convert_Unicode_to_Richa()
{

var chunksize = document.getElementById("chunksize").value;
var array_one = new Array( 

// ignore all nuktas except in ड़ and ढ़
"‘",   "’",   "“",   "”",   "(",    ")",   "{",    "}",   "=", "।",  "?",  "-",  "µ", "॰", ",", ".", "् ", 
";","\"","'",
"०",  "१",  "२",  "३",  "४",   "५",  "६",   "७",   "८",   "९", "x", 
"् ","्,",

"फ़्",  "क़",  "ख़",  "ग़", "ज़्", "ज़",  "ड़",  "ढ़",   "फ़",  "य़",  "ऱ",  "ऩ",    // one-byte nukta varNas

"त्त्",   "त्त",     "क्त",  "दृ",  "कृ",

"ह्न","ह्व", "ह्य",  "हृ",  "ह्म",  "ह्र",  "ह्",   "द्द",  "क्ष्", "क्ष", "त्र्", "त्र","ज्ञ",
"छ्य",  "ट्य",  "ठ्य",  "ड्य",  "ढ्य", "द्य", "द्व","द्घ", "द्ध",  
"श्र",  "ट्र",    "ड्र",    "ढ्र",    "छ्र",   "क्र",  "फ्र",  "द्र",   "प्र", "भ्र",  "ग्र", "रु",  "रू",
"्र",
"ओ",  "औ",  "आ",   "अ",   "ई",   "इ",  "उ",   "ऊ",  "ऐ",  "ए", "ऋ","ॠ","ऌ",

"क्",  "क",  "क्क",  "ख्",   "ख",    "ग्",   "ग",  "घ्",  "घ",    "ङ",
"च्",   "च",   "छ",  "ज्", "ज",   "झ्",  "झ",  "ञ्", "ञ",

"ट्ट",   "ट्ठ",   "ट",   "ठ",   "ड्ड",   "ड्ढ",  "ड",   "ढ",  "ण्", "ण",  
"त्",  "त",  "थ्", "थ", "द", "ध्", "ध",  "न्",  "न",  

"प्",  "प",  "फ्", "फ",  "ब्",  "ब", "भ्",  "भ",  "म्",  "म",
"य्",  "य",  "र",  "ल्", "ल",  "ळ",  "व्",  "व", 
"श्", "श",  "ष्", "ष",  "स्",   "स",   "ह",     

"ीं","ैं","ौं",
"ऑ",   "ॉ",  "ो",   "ौ",   "ा",   "ी",   "ु",   "ू",   "ृ", "ॄ",  "े",   "ै",
"ं",   "ँ",   "ः",   "ॅ",    "ऽ", "्र",  "्","॰" )

var array_two = new Array( 

"^", "*",  "ß", "Þ", "¼", "½", "¿", "À", "¾", "A", "\\", "&", "&", "Œ", "]","-","~ ", 
"(","“","‘",
"0",  "ƒ",  "„",   "…",   "†",   "‡",   "ˆ",   "‰",   "Š",   "•","Û",  //"•","‹",
"~ ","~]",
"¶",   "d",    "[k",  "x",  "T",  "t",   "M+", "<+", "Q",  ";",    "j",   "u",//One byte nuqtas
"Ù",   "Ùk",   "ä",    "n`",  "Ñ",       

"Ð","à",   "á",    "â",   "ã",   "gª",  "º",   "í", "{", "{k",  "«", "=","K", //"ºz",
"Nî",   "Vî",    "Bî",   "Mî",   "<î", "|","}","ð", ")",
"J",   "Vª",   "Mª",  "<ªª",  "Nª",   "Ø",  "Ý",   "æ", "ç","Hkz", "xz", "#", ":",
"z",
"vks",  "vkS",  "vk",    "v",   "bZ",  "b",  "m",  "Å",  ",s",  ",",   "_","Ç","È",

"D",  "d",    "ô",     "[",     "[k",    "X",   "x",  "?",    "?k",   "³", 
"P",    "p",  "N",   "T",    "t",   "÷",  ">",  "×", "´",

"ê", "ë",  "V",  "B",   "ì",  "ï",     "M",  "<",  ".", "Æ",   
"R",  "r",  "F", "Fk", "n", "/",  "/k",  "U", "u",   

"I",  "i",   "¶", "Q",   "C",  "c",  "H",  "Hk", "E",   "e",
"¸",   ";",    "j",  "Y",   "y",  "G",  "O",  "o",
"'", "'k",  "\"", "\"k", "L",   "l",   "g",      

"ö","®","k®",
"vkW",    "kW",    "ks",   "kS",   "k",     "h",    "q",   "w",   "`", "¤",   "s",    "S",
"a",    "¡",    "%",     "W",   "·",  "~j",  "~","-")

// The following two characters are to be replaced through proper checking of locations:

// "र्" (reph) 
// "Z" )

// "ि"  
// "f" )


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 font to convert into Richa. 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 }

     var modified_substring = document.getElementById("legacy_text").value.substring ( sthiti1, sthiti2 )  ;

    Replace_Symbols( ) ;

    processed_text += modified_substring ;

//****************************************************************************************
//  Breaking part code over
//****************************************************************************************
//  processed_text = processed_text.replace( /mangal/g , "Richa" ) ;   

    document.getElementById("unicode_text").value = processed_text  ;
   }

}
// if input is HTML then

else    
   {
    document.getElementById("unicode_text").value = "You have chosen HTML TEXT in Richa font to convert into Unicode. 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( "Richa" , 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 "Richa"  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( "Richa" ) ;
       } //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  'Richa' .
 
    processed_text = processed_text.replace( /Mangal/g , "Richa" ) ;   

    // 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 != "" )  
       {


// first replace the two-byte nukta_varNa with corresponding one-byte nukta varNas.

modified_substring = modified_substring.replace ( /क़/g , "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" )  ;
//modified_substring = modified_substring.replace ( /(/g , ";" )  ; 

        // code for replacing "ि" (chhotee ee kii maatraa) with "f"  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 + "ि" ,  "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 + "f", "f" + 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  Z  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 + "Z" ) ;
            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_Richa 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/><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="5"></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 Richa font >> " onClick="convert_Unicode_to_Richa();" accesskey="c" title="शॉर्टकट alt+c"> 
</div>  
<div id="hiddenLayer"></div><br>
The above text converted to Richa font is shown in the box below : <br/>
<textarea name="ConvertedText" id="unicode_text" cols="80" rows="10"></textarea>
<br />

</form>
</body>

</html>