Java: Capture Web Page

Recently I was investigating how to capture web page as image.
I recently realized that there is a nicer nice component for navigating web page in Java FX.

• javafx.scene.web.WebView
• javafx.embed.swing.JFXPanel

I referred this stackoverflow post and wrote a Java class for capturing web page and save it as image.

Please note that you should include jfxrt.jar, which was in JDK1.7.0_XX/jre/lib, should be added to your classpath.

Code

Here is an actual code. Please feel free to use it...
I am very happy if you give me some feedback for this code.

package com.dukesoftware.javafx;

import static com.dukesoftware.utils.io.IOUtils.getExtension;

import java.awt.image.BufferedImage;
import java.io.File;
import java.io.IOException;
import java.util.Map;
import java.util.Set;
import java.util.concurrent.Executors;
import java.util.concurrent.LinkedBlockingQueue;

import javafx.application.Platform;
import javafx.beans.value.ChangeListener;
import javafx.beans.value.ObservableValue;
import javafx.collections.ListChangeListener;
import javafx.concurrent.Worker.State;
import javafx.embed.swing.JFXPanel;
import javafx.scene.Node;
import javafx.scene.Scene;
import javafx.scene.web.WebEngine;
import javafx.scene.web.WebView;

import javax.imageio.ImageIO;
import javax.swing.JFrame;

import com.dukesoftware.utils.awt.AwtUtils;
import com.dukesoftware.utils.io.IOUtils;
import com.dukesoftware.utils.swing.jframe.JFrameFactory;
import com.google.common.collect.ImmutableMap;

public class WebCapture extends JFXPanel{
    
    private static final String PATH = "path";
    private static final String URL = "url";
    private static final String COMMAND = "command";

    private static final String SHUTDOWN = "shutdown";
    
    public static void main(String[] args) throws InterruptedException {
       WebCapture capture = new WebCapture(1024, 768, 1000);

       // you can see the WebView behavior if you want
       // JFrame frame = JFrameFactory.createExitJFrame(capture, "", 1024, 768);
       // frame.pack();
       // frame.setVisible(true);

       capture
       .load("http://www.google.com", "c:/temp/google.png")
       .load("http://www.yahoo.com", "c:/temp/yahoo.png")
       .load("http://www.bing.com", "c:/temp/bing.png")
       .shutdownAfterAllRequestsProcessed();
    }
    
    private WebView webView;
    
    private final long waitMills;
    private final LinkedBlockingQueue<Map<String, String>> queue = new LinkedBlockingQueue<>();
    private volatile Map<String, String> current = null;
    
    public WebCapture(int width, int height, long waitMills){
        setSize(width, height);
        this.waitMills = waitMills;
        Platform.runLater(new Runnable() {
            @Override
            public void run() {
                initWebViewFX();
            }
        });
    }

    // This method is invoked on JavaFX thread
    private void initWebViewFX() {
        if(webView != null) {
            throw new IllegalStateException();
        }
        webView = new WebView();
        webView.getStyleClass().add(".scroll-bar{ -fx-scale-x: 0;-fx-scale-y: 0;-fx-scale-z: 0; }");
        System.out.println(webView.getStylesheets());
        
        // hide webview scrollbars whenever they appear.
//        hideScrollBar();
        
        // get web engine
        final WebEngine engine = webView.getEngine();
        
        // add listener for capture image when loading page is completed
        engine.getLoadWorker().stateProperty().addListener(new ChangeListener<State>() {

            @Override
            public void changed(ObservableValue<? extends State> observable,
                    State oldValue, State newValue) {
                if (State.SUCCEEDED == newValue) {
                    Executors.newSingleThreadExecutor().execute(new Runnable() {
                        @Override
                        public void run() {
                            try {
                                // need this sleep because sometimes the capturing is started before the rendering is done 
                                // if anyone knows better solution, please tell me!!
                                Thread.sleep(waitMills);
                                BufferedImage image = AwtUtils.captureView(WebCapture.this);
                                String path = current.get(PATH);
                                ImageIO.write(image, IOUtils.getExtension(path), new File(path));

                            } catch (IOException e) {
                                e.printStackTrace();
                            } catch (InterruptedException e) {
                                e.printStackTrace();
                            }
                            fireNextOnJavaFXThread();

                        }
                    });
                }
                else if(State.FAILED == newValue){
                    fireNextOnJavaFXThread();
                }
            }
        });
        
        setScene(new Scene(webView, getWidth(), getHeight()));
    }


    private void fireNext() {
        if(queue.isEmpty()){
            current = null;
            return;
        }
        try {
            current = queue.take();
            if(current.get(COMMAND).equals(SHUTDOWN)){
                System.out.println("all done");
                System.exit(0);
            }
            else{
                System.out.println("loading " + current.get(URL));
                webView.getEngine().load(current.get(URL));
            }
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
    }

    private void hideScrollBar() {
        webView.getChildrenUnmodifiable().addListener(new ListChangeListener<Node>() {
          @Override public void onChanged(Change<? extends Node> change) {
            Set<Node> deadSeaScrolls = webView.lookupAll(".scroll-bar");
            for (Node scroll : deadSeaScrolls) {
              scroll.setVisible(false);
            }
          }
        });
    }
    
    
    public WebCapture load(final String url, final String path) throws InterruptedException {
        queue.put(ImmutableMap.of(COMMAND, "load", URL, url, PATH, path));
        Platform.runLater(new Runnable() {
            @Override
            public void run() {
                if(current == null){
                    fireNext();
                }
            }
        });
        return this;
    }

    private void shutdownAfterAllRequestsProcessed() throws InterruptedException {
        queue.put(ImmutableMap.of(COMMAND, SHUTDOWN));
        if(queue.isEmpty()){
            Platform.exit();
            // sorry sorry anti-pattern but for standalone program this should be fine...
            System.exit(0);
        }
    }

    private void fireNextOnJavaFXThread() {
        Platform.runLater(new Runnable() {
            @Override
            public void run() {
                fireNext();
            }
        });
    }

    public final static String getExtension(final String fname) {
        int dotNum = fname.lastIndexOf('.');
        if (dotNum < 0) {
            return "";
        } else {
            return fname.substring(dotNum + 1);
        }
    }

}

Here is some trivial utility methods used in above web page capture class.

    // AwtUtils
    public static BufferedImage captureView(Component component) throws IOException {
        BufferedImage image = new BufferedImage(component.getWidth(), component.getHeight(), BufferedImage.TYPE_INT_ARGB);
        Graphics graphics = image.createGraphics();
        component.paint(graphics);
        graphics.dispose();
        image.flush();
        return image;
    }

    // JFrameFactory
    public static JFrame createExitJFrame(Component component, String title, int width, int height){
        JFrame frame = new JFrame(title);
        frame.getContentPane().add(component, BorderLayout.CENTER);
        frame.setPreferredSize(new Dimension(width, height));
        frame.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
        frame.pack();
        return frame;
    }

    // IOUtils
    public final static String getExtension(final String fname) {
        int dotNum = fname.lastIndexOf('.');
        if (dotNum < 0) {
            return "";
        } else {
            return fname.substring(dotNum + 1);
        }
    }

Pending Task

• Hide scroll bar
I read this stackoverflow post and calling setVisible(false) method.
it works very well but only one thing I would not prefer is white space corresponding to an original scroll bar place rendered.
Looks the best way to avoid this is defining custom css for the scroll bar.
• Is there any better way to trigger capturing page?
Sometimes capturing is started before rendering is fully done.
So I added silly sleep call before capturing the web page.