Tokenização de palavras em arquivos txt, doc e pdf

// Java core packages

import java.awt.<em>;

import java.awt.event.</em>;

import <a href="http://java.io">java.io</a>.<em>;

import java.text.</em>;

import java.util.*;

private JTextField enterField;

private JTextArea outputArea;

private BufferedWriter out;

private String word;

// set up GUI

public ConstrutorDeTemplate1()

{

super( “Testing class File” );

enterField = new JTextField(
     "Enter file or directory name here" );
  enterField.addActionListener( this );
  outputArea = new JTextArea();

  ScrollPane scrollPane = new ScrollPane();
  scrollPane.add( outputArea );

  Container container = getContentPane();
  container.add( enterField, BorderLayout.NORTH );
  container.add( scrollPane, BorderLayout.CENTER );

  setSize( 400, 400 );
  show();

// display information about file user specifies

public void actionPerformed( ActionEvent actionEvent )

{

File name = new File( actionEvent.getActionCommand() );

// if name exists, output information about it
  if ( name.exists() ) {
     outputArea.setText(
        name.getName() + " exists\n" +
        ( name.isFile() ? 
           "is a file\n" : "is not a file\n" ) +
        ( name.isDirectory() ? 
           "is a directory\n" : "is not a directory\n" ) +
        ( name.isAbsolute() ? "is absolute path\n" : 
           "is not absolute path\n" ) +
        "Last modified: " + name.lastModified() +
        "\nLength: " + name.length() +
        "\nPath: " + name.getPath() +
        "\nAbsolute path: " + name.getAbsolutePath() +
        "\nParent: " + name.getParent() );
     
     // output information if name is a file
     if ( name.isFile() ) {

        // append contents of file to outputArea

        try {
            // Create the tokenizer to read from a file
            FileReader rd = new FileReader(name);
            StreamTokenizer st = new StreamTokenizer(rd);

            // Prepare the tokenizer for Java-style tokenizing rules
            st.parseNumbers();
            st.wordChars('_', '_');
            st.eolIsSignificant(true);

            // If whitespace is not to be discarded, make this call
            st.ordinaryChars(0, ' ');

            // These calls caused comments to be discarded
            st.slashSlashComments(true);
            st.slashStarComments(true);

            // Parse the file
            int token = st.nextToken();
            String word_ant = "";
            outputArea.append( " \n" );
            out = new BufferedWriter(new FileWriter(name, true));
    
            while (token != StreamTokenizer.TT_EOF) {
                token = st.nextToken();
                if (token == StreamTokenizer.TT_EOL){
                    out.flush();
                    out = new BufferedWriter(new FileWriter(name, true));
                }
        
                switch (token) {
                case StreamTokenizer.TT_NUMBER:
                    // A number was found; the value is in nval
                    double num = st.nval;
                    break;
                case StreamTokenizer.TT_WORD:
                    // A word was found; the value is in sval

                    word = st.sval;

		outputArea.append( word.toString() + " \n " );

                    break;

                case '"':
                    // A double-quoted string was found; sval contains the contents
                    String dquoteVal = st.sval;
                    break;
                case '\'':
                    // A single-quoted string was found; sval contains the contents
                    String squoteVal = st.sval;
                    break;
                case StreamTokenizer.TT_EOL:
                    // End of line character found
                    break;
                case StreamTokenizer.TT_EOF:
                    // End of file has been reached
                    break;
                default:
                    // A regular character was found; the value is the token itself
                    char ch = (char)st.ttype;
                    break;
                } // fim do switch
            } // fim do while
            rd.close();
            out.close(); 
        } // fim do try 

         
        // process file processing problems
        catch( IOException ioException ) {
           JOptionPane.showMessageDialog( this,
              "FILE ERROR",
              "FILE ERROR", JOptionPane.ERROR_MESSAGE );
        }
     } // fim do if da linha 68

     // output directory listing
     else if ( name.isDirectory() ) {
        String directory[] = name.list();

        outputArea.append( "\n\nDirectory contents:\n");

        for ( int i = 0; i < directory.length; i++ )
           outputArea.append( directory[ i ] + "\n" );
     }
  } // fim do if da linha 52

  // not file or directory, output error message
  else {
     JOptionPane.showMessageDialog( this,
        actionEvent.getActionCommand() + " Does Not Exist",
        "ERROR", JOptionPane.ERROR_MESSAGE );
  }

// executa a aplicação

public static void main( String args[] )

{

ConstrutorDeTemplate1 application = new ConstrutorDeTemplate1();

application.setDefaultCloseOperation(
     JFrame.EXIT_ON_CLOSE );