BMP and ICO image formats support : integrated /haraldk/TwelveMonkeys

imageio-bmp-3.2 library.

 - better BMP format flavours support
 - handle PNG encoded icons
 - handle transparency
 
Added some javadoc url references to .classpath

.classpath

@@ -31,7 +31,11 @@
 	<classpathentry kind="lib" path="lib/jcl-over-slf4j-1.7.12.jar"/>
 	<classpathentry kind="lib" path="lib/slf4j-jdk14-1.7.12.jar"/>
 	<classpathentry kind="lib" path="lib/log4j-over-slf4j-1.7.12.jar"/>
-	<classpathentry kind="lib" path="lib/httpcore-4.4.4.jar"/>
+	<classpathentry kind="lib" path="lib/httpcore-4.4.4.jar">
+		<attributes>
+			<attribute name="javadoc_location" value="https://hc.apache.org/httpcomponents-core-ga/httpcore/apidocs"/>
+		</attributes>
+	</classpathentry>
 	<classpathentry kind="lib" path="lib/metadata-extractor-2.8.1.jar"/>
 	<classpathentry kind="lib" path="lib/xmpcore-5.1.2.jar"/>
 	<classpathentry kind="lib" path="lib/jcifs-1.3.17.jar"/>
@@ -46,10 +50,18 @@
 	<classpathentry kind="lib" path="lib/bcprov-jdk15-1.46.jar"/>
 	<classpathentry kind="lib" path="lib/poi-3.13-20150929.jar"/>
 	<classpathentry kind="lib" path="lib/poi-scratchpad-3.13-20150929.jar"/>
-	<classpathentry kind="lib" path="lib/commons-io-2.4.jar"/>
+	<classpathentry kind="lib" path="lib/commons-io-2.4.jar">
+		<attributes>
+			<attribute name="javadoc_location" value="https://commons.apache.org/proper/commons-io/javadocs/api-release/"/>
+		</attributes>
+	</classpathentry>
 	<classpathentry kind="lib" path="lib/slf4j-api-1.7.12.jar"/>
 	<classpathentry kind="lib" path="lib/chardet.jar"/>
-	<classpathentry kind="lib" path="lib/jetty-client-9.2.13.v20150730.jar"/>
+	<classpathentry kind="lib" path="lib/jetty-client-9.2.13.v20150730.jar">
+		<attributes>
+			<attribute name="javadoc_location" value="http://download.eclipse.org/jetty/stable-9/apidocs/"/>
+		</attributes>
+	</classpathentry>
 	<classpathentry kind="lib" path="lib/jetty-continuation-9.2.13.v20150730.jar"/>
 	<classpathentry kind="lib" path="lib/jetty-deploy-9.2.13.v20150730.jar"/>
 	<classpathentry kind="lib" path="lib/jetty-http-9.2.13.v20150730.jar"/>
@@ -97,6 +109,11 @@
 	<classpathentry kind="lib" path="lib/common-lang-3.2.jar"/>
 	<classpathentry kind="lib" path="lib/imageio-core-3.2.jar"/>
 	<classpathentry kind="lib" path="lib/imageio-metadata-3.2.jar"/>
-	<classpathentry kind="lib" path="lib/imageio-tiff-3.2.jar"/>
+	<classpathentry kind="lib" path="lib/imageio-tiff-3.2.jar">
+		<attributes>
+			<attribute name="javadoc_location" value="http://search.maven.org/remotecontent?filepath=com/twelvemonkeys/imageio/imageio-tiff/3.2/imageio-tiff-3.2-javadoc.jar"/>
+		</attributes>
+	</classpathentry>
+	<classpathentry kind="lib" path="lib/imageio-bmp-3.2.jar"/>
 	<classpathentry kind="output" path="gen"/>
 </classpath>

build.xml

@@ -182,6 +182,7 @@
         <pathelement location="${lib}/httpcore-4.4.4.jar" />
         <pathelement location="${lib}/httpmime-4.5.1.jar" />
         <pathelement location="${lib}/icu4j-56_1.jar" />
+    	<pathelement location="${lib}/imageio-bmp-3.2.jar" />
     	<pathelement location="${lib}/imageio-core-3.2.jar" />
     	<pathelement location="${lib}/imageio-metadata-3.2.jar" />
     	<pathelement location="${lib}/imageio-tiff-3.2.jar" />

pom.xml

@@ -430,6 +430,11 @@
             <artifactId>icu4j</artifactId>
             <version>56.1</version>
         </dependency>
+        <dependency>
+            <groupId>com.twelvemonkeys.imageio</groupId>
+            <artifactId>imageio-bmp</artifactId>
+            <version>3.2</version>
+        </dependency> 
         <dependency>
             <groupId>com.twelvemonkeys.imageio</groupId>
             <artifactId>imageio-tiff</artifactId>

source/net/yacy/document/ImageParser.java

@@ -20,53 +20,31 @@
 
 package net.yacy.document;
 
-import java.awt.Container;
 import java.awt.Image;
-import java.awt.MediaTracker;
+import java.awt.image.BufferedImage;
 import java.io.ByteArrayInputStream;
 import java.io.IOException;
 
 import javax.imageio.ImageIO;
 
 import net.yacy.cora.util.ConcurrentLog;
-import net.yacy.document.parser.images.bmpParser;
-import net.yacy.document.parser.images.icoParser;
 
 public class ImageParser {
 
+	/**
+	 * @param filename source image file url
+	 * @param source image content as bytes
+	 * @return an Image instance parsed from image content bytes, or null if no parser can handle image format or an error occured
+	 */
 	public static final Image parse(final String filename, final byte[] source) {
-        final MediaTracker mediaTracker = new MediaTracker(new Container());
+		BufferedImage image = null;
-        Image image = null;
-        if (((filename.endsWith(".ico")) || (filename.endsWith(".bmp"))) && (bmpParser.isBMP(source))) {
-            // parse image with BMP parser
-            image = bmpParser.parse(source).getImage();
-            if (image == null) {
-                if (ConcurrentLog.isFine("IMAGEPARSER")) {
-                    ConcurrentLog.fine("IMAGEPARSER", "IMAGEPARSER.parse : bmpParser failed for " + filename);
-                }
-                return null;
-            }
-        } else if ((filename.endsWith(".ico")) && (icoParser.isICO(source))) {
-            // parse image with ICO parser
-            icoParser icoparser;
-            try {
-                icoparser = new icoParser(source);
-                image = icoparser.getImage(0);
-            } catch (final Throwable e) {
-    			if (ConcurrentLog.isFine("IMAGEPARSER")) {
-    				ConcurrentLog.fine("IMAGEPARSER", "IMAGEPARSER.parse : could not parse image " + filename, e);
-    			}
-            }
-            if (image == null) {
-                if (ConcurrentLog.isFine("IMAGEPARSER")) {
-                    ConcurrentLog.fine("IMAGEPARSER", "IMAGEPARSER.parse : icoParser failed for " + filename);
-                }
-                return null;
-            }
-        } else {
 		try {
 			image = ImageIO.read(new ByteArrayInputStream(source));
-    		} catch(IOException e) {
+			/*
+			 * With ImageIO.read, image is already loaded as a complete BufferedImage, no need to wait
+			 * full loading with a MediaTracker
+			 */
+		} catch (IOException e) {
 			if (ConcurrentLog.isFine("IMAGEPARSER")) {
 				ConcurrentLog.fine("IMAGEPARSER", "IMAGEPARSER.parse : could not parse image " + filename, e);
 			}
@@ -77,23 +55,6 @@ public class ImageParser {
 			}
 			return null;
 		}
-        }
-        if (image == null) {
-        	return null;
-        }
-
-        final int handle = image.hashCode();
-        mediaTracker.addImage(image, handle);
-        try {
-            mediaTracker.waitForID(handle);
-            
-            if (mediaTracker.isErrorID(handle)) { // true if status ERRORD during loading (happens on not supported formats too)
-                mediaTracker.removeImage(image, handle);
-                image = null; // return null to indicate source not handled
-            }
-        } catch (final InterruptedException e) {
-            return null;
-        }
 
 		return image;
 	}

source/net/yacy/document/parser/images/bmpParser.java

@@ -32,8 +32,17 @@ import java.io.IOException;
 
 import javax.imageio.ImageIO;
 
+import com.twelvemonkeys.imageio.plugins.bmp.BMPImageReader;
+
 import net.yacy.cora.util.ConcurrentLog;
 
+/**
+ *
+ * @deprecated use ImageIO {@link BMPImageReader} from github.com/haraldk/TwelveMonkeys
+ *             library (imageio-bmp-3.2.jar), which as better BMP format
+ *             variants support
+ */
+@Deprecated
 public class bmpParser {
 
     // this is a implementation of http://de.wikipedia.org/wiki/Windows_Bitmap

source/net/yacy/document/parser/images/genericImageParser.java

@@ -46,6 +46,13 @@ import java.util.Set;
 
 import javax.imageio.ImageIO;
 
+import com.drew.imaging.jpeg.JpegMetadataReader;
+import com.drew.lang.GeoLocation;
+import com.drew.metadata.Directory;
+import com.drew.metadata.Metadata;
+import com.drew.metadata.Tag;
+import com.drew.metadata.exif.GpsDirectory;
+
 import net.yacy.cora.document.id.AnchorURL;
 import net.yacy.cora.document.id.DigestURL;
 import net.yacy.cora.document.id.MultiProtocolURL;
@@ -55,16 +62,8 @@ import net.yacy.document.Document;
 import net.yacy.document.Parser;
 import net.yacy.document.VocabularyScraper;
 import net.yacy.document.parser.html.ImageEntry;
-import net.yacy.document.parser.images.bmpParser.IMAGEMAP;
 import net.yacy.kelondro.util.FileUtils;
 
-import com.drew.imaging.jpeg.JpegMetadataReader;
-import com.drew.lang.GeoLocation;
-import com.drew.metadata.Directory;
-import com.drew.metadata.Metadata;
-import com.drew.metadata.Tag;
-import com.drew.metadata.exif.GpsDirectory;
-
 /**
  * Parser for images, bmp and jpeg and all supported by the Java Image I/O API
  * by default java ImageIO supports bmp, gif, jpg, jpeg, png, wbmp (tif if jai-imageio is in classpath/registered)
@@ -75,11 +74,9 @@ public class genericImageParser extends AbstractParser implements Parser {
     public genericImageParser() {
         super("Generic Image Parser");
 
-        SUPPORTED_EXTENSIONS.add("bmp");
         SUPPORTED_EXTENSIONS.add("jpe"); // not listed in ImageIO extension but sometimes uses for jpeg
         SUPPORTED_EXTENSIONS.addAll(Arrays.asList(ImageIO.getReaderFileSuffixes()));
 
-        SUPPORTED_MIME_TYPES.add("image/bmp");
         SUPPORTED_MIME_TYPES.add("image/jpg"); // this is in fact a 'wrong' mime type. We leave it here because that is a common error that occurs in the internet frequently
         SUPPORTED_MIME_TYPES.addAll(Arrays.asList(ImageIO.getReaderMIMETypes()));
     }
@@ -102,21 +99,7 @@ public class genericImageParser extends AbstractParser implements Parser {
         String ext = MultiProtocolURL.getFileExtension(filename);
         double gpslat = 0;
         double gpslon = 0;        
-        if (mimeType.equals("image/bmp") || ext.equals("bmp")) {
+        if (mimeType.equals("image/jpeg") || ext.equals("jpg") || ext.equals("jpeg") || ext.equals("jpe")) {
-            byte[] b;
-            try {
-                b = FileUtils.read(source);
-            } catch (final IOException e) {
-                ConcurrentLog.logException(e);
-                throw new Parser.Failure(e.getMessage(), location);
-            }
-            if (bmpParser.isBMP(b)) {
-                final IMAGEMAP imap = bmpParser.parse(b);
-                ii = parseJavaImage(location, imap.getImage());
-            } else {
-                throw new Parser.Failure("Not supported by bmpParser", location);
-            }
-        } else if (mimeType.equals("image/jpeg") || ext.equals("jpg") || ext.equals("jpeg") || ext.equals("jpe")) {
             // use the exif parser from
             // http://www.drewnoakes.com/drewnoakes.com/code/exif/
             // javadoc is at: http://www.drewnoakes.com/drewnoakes.com/code/exif/javadoc/

source/net/yacy/document/parser/images/icoParser.java

@@ -32,9 +32,17 @@ import java.io.IOException;
 
 import javax.imageio.ImageIO;
 
-import net.yacy.cora.util.ConcurrentLog;
+import com.twelvemonkeys.imageio.plugins.bmp.ICOImageReader;
 
+import net.yacy.cora.util.ConcurrentLog;
 
+/**
+*
+* @deprecated use ImageIO {@link ICOImageReader} from github.com/haraldk/TwelveMonkeys
+*             library (imageio-bmp-3.2.jar), which as better BMP format
+*             variants support, and support PNG encoded icons.
+*/
+@Deprecated
 public class icoParser {
 
     // this is a implementation of http://msdn2.microsoft.com/en-us/library/ms997538(d=printer).aspx