org.xhtmlrenderer.swing.NaiveUserAgent Maven / Gradle / Ivy
/*
* NaiveUserAgent.java
* Copyright (c) 2004, 2005 Torbj�rn Gannholm
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*
*/
package org.xhtmlrenderer.swing;
import org.xhtmlrenderer.resource.CSSResource;
import org.xhtmlrenderer.resource.ImageResource;
import org.xhtmlrenderer.resource.XMLResource;
import org.xhtmlrenderer.util.XRLog;
import org.xhtmlrenderer.extend.UserAgentCallback;
import org.xhtmlrenderer.event.DocumentListener;
import javax.imageio.ImageIO;
import java.awt.image.BufferedImage;
import java.awt.*;
import java.io.ByteArrayOutputStream;
import java.io.File;
import java.io.IOException;
import java.io.InputStream;
import java.io.FileNotFoundException;
import java.net.MalformedURLException;
import java.net.URL;
import java.util.LinkedHashMap;
import java.util.Iterator;
/**
* NaiveUserAgent is a simple implementation of {@link UserAgentCallback} which places no restrictions on what
* XML, CSS or images are loaded, and reports visited links without any filtering. The most straightforward process
* available in the JDK is used to load the resources in question--either using java.io or java.net classes.
*
*
The NaiveUserAgent has a small cache for images,
* the size of which (number of images) can be passed as a constructor argument. There is no automatic cleaning of
* the cache; call {@link #shrinkImageCache()} to remove the least-accessed elements--for example, you might do this
* when a new document is about to be loaded. The NaiveUserAgent is also a DocumentListener; if registered with a
* source of document events (like the panel hierarchy), it will respond to the
* {@link org.xhtmlrenderer.event.DocumentListener#documentStarted()} call and attempt to shrink its cache.
*
*
This class is meant as a starting point--it will work out of the box, but you should really implement your
* own, tuned to your application's needs.
*
* @author Torbj�rn Gannholm
*/
public class NaiveUserAgent implements UserAgentCallback, DocumentListener {
private static final int DEFAULT_IMAGE_CACHE_SIZE = 16;
/**
* a (simple) LRU cache
*/
protected LinkedHashMap _imageCache;
private int _imageCacheCapacity;
private String _baseURL;
/**
* Creates a new instance of NaiveUserAgent with a max image cache of 16 images.
*/
public NaiveUserAgent() {
this(DEFAULT_IMAGE_CACHE_SIZE);
}
/**
* Creates a new NaiveUserAgent with a cache of a specific size.
*
* @param imgCacheSize Number of images to hold in cache before LRU images are released.
*/
public NaiveUserAgent(final int imgCacheSize) {
this._imageCacheCapacity = imgCacheSize;
// note we do *not* override removeEldestEntry() here--users of this class must call shrinkImageCache().
// that's because we don't know when is a good time to flush the cache
this._imageCache = new java.util.LinkedHashMap(_imageCacheCapacity, 0.75f, true);
}
/**
* If the image cache has more items than the limit specified for this class, the least-recently used will
* be dropped from cache until it reaches the desired size.
*/
public void shrinkImageCache() {
int ovr = _imageCache.size() - _imageCacheCapacity;
Iterator it = _imageCache.keySet().iterator();
while (it.hasNext() && ovr-- > 0) {
it.next();
it.remove();
}
}
/**
* Empties the image cache entirely.
*/
public void clearImageCache() {
_imageCache.clear();
}
/**
* Gets a Reader for the resource identified
*
* @param uri PARAM
* @return The stylesheet value
*/
//TOdO:implement this with nio.
protected InputStream resolveAndOpenStream(String uri) {
java.io.InputStream is = null;
uri = resolveURI(uri);
try {
is = new URL(uri).openStream();
} catch (java.net.MalformedURLException e) {
XRLog.exception("bad URL given: " + uri, e);
} catch (java.io.FileNotFoundException e) {
XRLog.exception("item at URI " + uri + " not found");
} catch (java.io.IOException e) {
XRLog.exception("IO problem for " + uri, e);
}
return is;
}
/**
* Retrieves the CSS located at the given URI. It's assumed the URI does point to a CSS file--the URI will
* be accessed (using java.io or java.net), opened, read and then passed into the CSS parser.
* The result is packed up into an CSSResource for later consumption.
*
* @param uri Location of the CSS source.
* @return A CSSResource containing the parsed CSS.
*/
public CSSResource getCSSResource(String uri) {
return new CSSResource(resolveAndOpenStream(uri));
}
/**
* Retrieves the image located at the given URI. It's assumed the URI does point to an image--the URI will
* be accessed (using java.io or java.net), opened, read and then passed into the JDK image-parsing routines.
* The result is packed up into an ImageResource for later consumption.
*
* @param uri Location of the image source.
* @return An ImageResource containing the image.
*/
public ImageResource getImageResource(String uri) {
ImageResource ir = null;
uri = resolveURI(uri);
ir = (ImageResource) _imageCache.get(uri);
//TODO: check that cached image is still valid
if (ir == null) {
InputStream is = resolveAndOpenStream(uri);
if (is != null) {
try {
BufferedImage img = ImageIO.read(is);
if (img == null) {
throw new IOException("ImageIO.read() returned null");
}
ir = createImageResource(uri, img);
_imageCache.put(uri, ir);
} catch (FileNotFoundException e) {
XRLog.exception("Can't read image file; image at URI '" + uri + "' not found");
} catch (IOException e) {
XRLog.exception("Can't read image file; unexpected problem for URI '" + uri + "'", e);
}
}
}
if (ir == null) {
ir = createImageResource(uri, null);
}
return ir;
}
/**
* Factory method to generate ImageResources from a given Image. May be overridden in subclass.
* @param uri The URI for the image, resolved to an absolute URI.
* @param img The image to package; may be null (for example, if image could not be loaded).
*
* @return An ImageResource containing the image.
*/
protected ImageResource createImageResource(String uri, Image img) {
return new ImageResource(AWTFSImage.createImage(img));
}
/**
* Retrieves the XML located at the given URI. It's assumed the URI does point to a XML--the URI will
* be accessed (using java.io or java.net), opened, read and then passed into the XML parser (XMLReader)
* configured for Flying Saucer. The result is packed up into an XMLResource for later consumption.
*
* @param uri Location of the XML source.
* @return An XMLResource containing the image.
*/
public XMLResource getXMLResource(String uri) {
InputStream inputStream = resolveAndOpenStream(uri);
XMLResource xmlResource;
try {
xmlResource = XMLResource.load(inputStream);
} finally {
if ( inputStream != null ) try {
inputStream.close();
} catch (IOException e) {
// swallow
}
}
return xmlResource;
}
public byte[] getBinaryResource(String uri) {
InputStream is = resolveAndOpenStream(uri);
try {
ByteArrayOutputStream result = new ByteArrayOutputStream();
byte[] buf = new byte[10240];
int i;
while ( (i = is.read(buf)) != -1) {
result.write(buf, 0, i);
}
is.close();
is = null;
return result.toByteArray();
} catch (IOException e) {
return null;
} finally {
if (is != null) {
try {
is.close();
} catch (IOException e) {
// ignore
}
}
}
}
/**
* Returns true if the given URI was visited, meaning it was requested at some point since initialization.
*
* @param uri A URI which might have been visited.
* @return Always false; visits are not tracked in the NaiveUserAgent.
*/
public boolean isVisited(String uri) {
return false;
}
/**
* URL relative to which URIs are resolved.
*
* @param url A URI which anchors other, possibly relative URIs.
*/
public void setBaseURL(String url) {
_baseURL = url;
}
/**
* Resolves the URI; if absolute, leaves as is, if relative, returns an absolute URI based on the baseUrl for
* the agent.
*
* @param uri A URI, possibly relative.
*
* @return A URI as String, resolved, or null if there was an exception (for example if the URI is malformed).
*/
public String resolveURI(String uri) {
if (uri == null) return null;
String ret = null;
if (_baseURL == null) {//first try to set a base URL
try {
URL result = new URL(uri);
setBaseURL(result.toExternalForm());
} catch (MalformedURLException e) {
try {
setBaseURL(new File(".").toURI().toURL().toExternalForm());
} catch (Exception e1) {
XRLog.exception("The default NaiveUserAgent doesn't know how to resolve the base URL for " + uri);
return null;
}
}
}
try {
URL result = new URL(new URL(_baseURL), uri);
ret = result.toString();
} catch (MalformedURLException e1) {
XRLog.exception("The default NaiveUserAgent cannot resolve the URL " + uri + " with base URL " + _baseURL);
}
return ret;
}
/**
* Returns the current baseUrl for this class.
*/
public String getBaseURL() {
return _baseURL;
}
public void documentStarted() {
shrinkImageCache();
}
public void documentLoaded() { /* ignore*/ }
public void onLayoutException(Throwable t) { /* ignore*/ }
public void onRenderException(Throwable t) { /* ignore*/ }
}
/*
* $Id: NaiveUserAgent.java,v 1.37 2007/11/23 07:03:30 pdoubleya Exp $
*
* $Log: NaiveUserAgent.java,v $
* Revision 1.37 2007/11/23 07:03:30 pdoubleya
* Applied patch from N. Barozzi to allow either toolkit or buffered images to be used, see https://xhtmlrenderer.dev.java.net/servlets/ReadMsg?list=dev&msgNo=3847
*
* Revision 1.36 2007/10/31 23:14:43 peterbrant
* Add rudimentary support for @font-face rules
*
* Revision 1.35 2007/06/20 12:24:31 pdoubleya
* Fix bug in shrink cache, trying to modify iterator without using safe remove().
*
* Revision 1.34 2007/06/19 21:25:41 pdoubleya
* Cleanup for caching in NUA, making it more suitable to use as a reusable UAC. NUA is also now a document listener and uses this to try and trim its cache down. PanelManager and iTextUA are now NUA subclasses.
*
* Revision 1.33 2007/05/20 23:25:33 peterbrant
* Various code cleanups (e.g. remove unused imports)
*
* Patch from Sean Bright
*
* Revision 1.32 2007/05/09 21:52:06 pdoubleya
* Fix for rendering problems introduced by removing GraphicsUtil class. Use Image instead of BufferedImage in most cases, convert to AWT image if necessary. Not complete, requires cleanup.
*
* Revision 1.31 2007/05/05 21:08:27 pdoubleya
* Changed image-related interfaces (FSImage, ImageUtil, scaling) to all use BufferedImage, since there were no Image-specific APIs we depended on, and we have more control over what we do with BIs as compared to Is.
*
* Revision 1.30 2007/05/05 18:05:21 pdoubleya
* Remove references to GraphicsUtil and the class itself, no longer needed
*
* Revision 1.29 2007/04/10 20:46:02 pdoubleya
* Fix, was not closing XML source stream when done
*
* Revision 1.28 2007/02/07 16:33:31 peterbrant
* Initial commit of rewritten table support and associated refactorings
*
* Revision 1.27 2006/06/28 13:46:59 peterbrant
* ImageIO.read() can apparently return sometimes null instead of throwing an exception when processing an invalid image
*
* Revision 1.26 2006/04/27 13:28:48 tobega
* Handle situations without base url and no file access gracefully
*
* Revision 1.25 2006/04/25 00:23:20 peterbrant
* Fixes from Mike Curtis
*
* Revision 1.23 2006/04/08 08:21:24 tobega
* relative urls and linked stylesheets
*
* Revision 1.22 2006/02/02 02:47:33 peterbrant
* Support non-AWT images
*
* Revision 1.21 2005/10/25 19:40:38 tobega
* Suggestion from user to use File.toURI.toURL instead of File.toURL because the latter is buggy
*
* Revision 1.20 2005/10/09 09:40:27 tobega
* Use current directory as default base URL
*
* Revision 1.19 2005/08/11 01:35:37 joshy
* removed debugging
* updated stylesheet to use right aligns
* Issue number:
* Obtained from:
* Submitted by:
* Reviewed by:
*
* Revision 1.17 2005/06/25 19:27:47 tobega
* UAC now supplies Resources
*
* Revision 1.16 2005/06/25 17:23:35 tobega
* first refactoring of UAC: ImageResource
*
* Revision 1.15 2005/06/21 17:52:10 joshy
* new hover code
* removed some debug statements
* Issue number:
* Obtained from:
* Submitted by:
* Reviewed by:
*
* Revision 1.14 2005/06/20 23:45:56 joshy
* hack to fix the mangled background images on osx
* Issue number:
* Obtained from:
* Submitted by:
* Reviewed by:
*
* Revision 1.13 2005/06/20 17:26:45 joshy
* debugging for image issues
* font scale stuff
*
* Issue number:
* Obtained from:
* Submitted by:
* Reviewed by:
*
* Revision 1.12 2005/06/15 11:57:18 tobega
* Making Browser a better model application with UserAgentCallback
*
* Revision 1.11 2005/06/15 11:53:47 tobega
* Changed UserAgentCallback to getInputStream instead of getReader. Fixed up some consequences of previous change.
*
* Revision 1.10 2005/06/13 06:50:16 tobega
* Fixed a bug in table content resolution.
* Various "tweaks" in other stuff.
*
* Revision 1.9 2005/06/03 00:29:49 tobega
* fixed potential bug
*
* Revision 1.8 2005/06/01 21:36:44 tobega
* Got image scaling working, and did some refactoring along the way
*
* Revision 1.7 2005/03/28 14:24:22 pdoubleya
* Remove stack trace on loading images.
*
* Revision 1.6 2005/02/02 12:14:01 pdoubleya
* Clean, format, buffer reader.
*
*
*/