Definition Extractor

/*
 * ShowSource.java
 *
 * Created on Dec 24, 2011, 5:08:47 PM
 */
package web.page.downloader;
import java.net.*;
import java.io.*;
import java.util.regex.*;
 
/**
 *
 * @author Blackhat11907
 */
 
public class ShowSource extends javax.swing.JFrame {
 
    /** Creates new form ShowSource */
    public ShowSource() {
        initComponents();
    }
 
    /** This method is called from within the constructor to
     * initialize the form.
     * WARNING: Do NOT modify this code. The content of this method is
     * always regenerated by the Form Editor.
     */
    @SuppressWarnings("unchecked")
    // <editor-fold defaultstate="collapsed" desc="Generated Code">                          
    private void initComponents() {
 
        jScrollPane1 = new javax.swing.JScrollPane();
        SourceTxtArea = new javax.swing.JTextArea();
        URLTxtField = new javax.swing.JTextField();
        GrabBtn = new javax.swing.JButton();
        jLabel1 = new javax.swing.JLabel();
        ClearBtn = new javax.swing.JButton();
 
        setDefaultCloseOperation(javax.swing.WindowConstants.EXIT_ON_CLOSE);
        setTitle("Definition Grabber");
        setResizable(false);
 
        SourceTxtArea.setColumns(20);
        SourceTxtArea.setEditable(false);
        SourceTxtArea.setLineWrap(true);
        SourceTxtArea.setRows(5);
        jScrollPane1.setViewportView(SourceTxtArea);
 
        GrabBtn.setText("Grab Def");
        GrabBtn.addActionListener(new java.awt.event.ActionListener() {
            public void actionPerformed(java.awt.event.ActionEvent evt) {
                GrabBtnActionPerformed(evt);
            }
        });
 
        jLabel1.setText("Word");
 
        ClearBtn.setText("Clear");
        ClearBtn.addActionListener(new java.awt.event.ActionListener() {
            public void actionPerformed(java.awt.event.ActionEvent evt) {
                ClearBtnActionPerformed(evt);
            }
        });
 
        javax.swing.GroupLayout layout = new javax.swing.GroupLayout(getContentPane());
        getContentPane().setLayout(layout);
        layout.setHorizontalGroup(
            layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING)
            .addGroup(layout.createSequentialGroup()
                .addContainerGap()
                .addGroup(layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING)
                    .addGroup(layout.createSequentialGroup()
                        .addComponent(jLabel1)
                        .addGap(18, 18, 18)
                        .addComponent(URLTxtField, javax.swing.GroupLayout.PREFERRED_SIZE, 245, javax.swing.GroupLayout.PREFERRED_SIZE)
                        .addGap(18, 18, 18)
                        .addComponent(GrabBtn, javax.swing.GroupLayout.PREFERRED_SIZE, 110, javax.swing.GroupLayout.PREFERRED_SIZE)
                        .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED, 20, Short.MAX_VALUE)
                        .addComponent(ClearBtn))
                    .addComponent(jScrollPane1, javax.swing.GroupLayout.DEFAULT_SIZE, 494, Short.MAX_VALUE))
                .addContainerGap())
        );
        layout.setVerticalGroup(
            layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING)
            .addGroup(layout.createSequentialGroup()
                .addContainerGap()
                .addComponent(jScrollPane1, javax.swing.GroupLayout.PREFERRED_SIZE, 270, javax.swing.GroupLayout.PREFERRED_SIZE)
                .addGap(18, 18, 18)
                .addGroup(layout.createParallelGroup(javax.swing.GroupLayout.Alignment.BASELINE)
                    .addComponent(jLabel1)
                    .addComponent(GrabBtn)
                    .addComponent(URLTxtField, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE)
                    .addComponent(ClearBtn))
                .addContainerGap(18, Short.MAX_VALUE))
        );
 
        pack();
    }// </editor-fold>                        
 
    public void parseHtmlCleaner(String content, String word) {
 
        // <style.*? means everything from that to after the ?
        Pattern style = Pattern.compile("<style.*?>.*?</style>"); 
        Matcher mstyle = style.matcher(content);
        while(mstyle.find()) content = mstyle.replaceAll("");
 
        // removes everything between script tags
        Pattern script = Pattern.compile("<script.*?>.*?</script>");
        Matcher mscript = script.matcher(content);
        while(mscript.find()) content = mscript.replaceAll("");
 
        // removes primary html tags
        Pattern tag = Pattern.compile("<.*?>");
        Matcher mtag = tag.matcher(content);
        while(mtag.find()) content = mtag.replaceAll("");
 
        //remove special chars like &67
        Pattern sChar = Pattern.compile("&.*?;");
        Matcher msChar = sChar.matcher(content);
        while(msChar.find()) content = msChar.replaceAll("");
 
        //remove tab chars and replace w newline chars
        Pattern tab = Pattern.compile("<t.*?>");
        Matcher mtab = tab.matcher(content);
        while(mtab.find()) content = mtab.replaceAll("/n");
 
        // erase the first part
        String infoReg = word + ".*?World English Dictionary";
        Pattern info = Pattern.compile(infoReg);
        Matcher minfo = info.matcher(content);
        while(minfo.find()) content = minfo.replaceAll("");
 
        // erase the last bottom
        Pattern info1 = Pattern.compile("Cite This Source.*?FIND OUT.*?Related Words.*?FIND OUT");
        Matcher minfo1 = info1.matcher(content);
        while(minfo1.find()) content = minfo1.replaceAll("");
 
        //erase final
        Pattern info2 = Pattern.compile("Collins.*?2009");
        Matcher minfo2 = info2.matcher(content);
        while(minfo2.find()) content = minfo2.replaceAll("");
 
        // erase white space
        Pattern wSpace = Pattern.compile("^[\t]+|[\t]+$");
        Matcher mwSpace = wSpace.matcher(content);
        while(mwSpace.find()) content = mwSpace.replaceAll("");
 
        // tab all lines
        Pattern tLine = Pattern.compile("[\t]+.*?\\.");
        Matcher mtLine = tLine.matcher(content);
        while(mtLine.find()) content = mtLine.replaceAll("\n");
 
 
        SourceTxtArea.append(content);      
    }
 
    public String setInfoAndGrab() throws IOException {
        String urlAddress = "http://www.dictionary.reference.com/browse/" + URLTxtField.getText();
 
        URL url = new URL(urlAddress);
        String line = null;
        String content = "";
 
        URLConnection con = url.openConnection();
        InputStream is = con.getInputStream();
 
        BufferedReader bufferR = new BufferedReader(new InputStreamReader(is));
 
        while ((line = bufferR.readLine()) != null){
 
            content += line + "\t";
 
        }
        return content;
    }
 
    public String addCap(String word){
 
        String result = "";
        for (int i = 0; i < word.length(); i++){
            String firstL = word.substring(i, i +1);
            if (i == 0){
                result += firstL.toUpperCase();
            } else {
                result += firstL.toLowerCase();
            }
        }
 
        return result;
    }
 
 
private void GrabBtnActionPerformed(java.awt.event.ActionEvent evt) {                                        
 
    String content = "";
    String word = URLTxtField.getText();
    try{
 
        content = setInfoAndGrab();
        word = addCap(word);  
        parseHtmlCleaner(content, word);
 
    } catch (Exception ex) {
        SourceTxtArea.append("Failed to connect to server");
    }
 
 
}                                       
 
private void ClearBtnActionPerformed(java.awt.event.ActionEvent evt) {                                         
 
    SourceTxtArea.setText("");
    URLTxtField.setText("");
    URLTxtField.requestFocus();
 
}                                        
 
    /**
     * @param args the command line arguments
     */
    public static void main(String args[]) throws IOException {
        /* Set the Nimbus look and feel */
        //<editor-fold defaultstate="collapsed" desc=" Look and feel setting code (optional) ">
        /* If Nimbus (introduced in Java SE 6) is not available, stay with the default look and feel.
         * For details see http://download.oracle.com/javase/tutorial/uiswing/lookandfeel/plaf.html 
         */
        try {
            for (javax.swing.UIManager.LookAndFeelInfo info : javax.swing.UIManager.getInstalledLookAndFeels()) {
                if ("Nimbus".equals(info.getName())) {
                    javax.swing.UIManager.setLookAndFeel(info.getClassName());
                    break;
                }
            }
        } catch (ClassNotFoundException ex) {
            java.util.logging.Logger.getLogger(ShowSource.class.getName()).log(java.util.logging.Level.SEVERE, null, ex);
        } catch (InstantiationException ex) {
            java.util.logging.Logger.getLogger(ShowSource.class.getName()).log(java.util.logging.Level.SEVERE, null, ex);
        } catch (IllegalAccessException ex) {
            java.util.logging.Logger.getLogger(ShowSource.class.getName()).log(java.util.logging.Level.SEVERE, null, ex);
        } catch (javax.swing.UnsupportedLookAndFeelException ex) {
            java.util.logging.Logger.getLogger(ShowSource.class.getName()).log(java.util.logging.Level.SEVERE, null, ex);
        }
        //</editor-fold>
 
        /* Create and display the form */
        java.awt.EventQueue.invokeLater(new Runnable() {
 
            public void run() {
                new ShowSource().setVisible(true);
            }
        });
 
      }
    // Variables declaration - do not modify                     
    private javax.swing.JButton ClearBtn;
    private javax.swing.JButton GrabBtn;
    private javax.swing.JTextArea SourceTxtArea;
    private javax.swing.JTextField URLTxtField;
    private javax.swing.JLabel jLabel1;
    private javax.swing.JScrollPane jScrollPane1;
    // End of variables declaration                   
}