Fix for Bug 19860: Modified Base64 to remedy non-compliance with RFC 2045. Non-Base64 characters were not being discarded during the decode. RFC 2045 explicitly states that all characters outside of the base64 alphabet are to be ignored. A new function discardNonBase64 is called prior to a decode. Note that isArrayBase64 only discards whitespace before testing the contents of an encoded byte array. Unit tests for Base64 were updated to reflect the changes related to discarding non-Base64 characters.

git-svn-id: https://svn.apache.org/repos/asf/jakarta/commons/proper/codec/trunk@130141 13f79535-47bb-0310-9956-ffa450edef68

Author: Timothy O'Brien

src/java/org/apache/commons/codec/binary/Base64.java

@@ -1,7 +1,7 @@
 /*
- * $Header: /home/jerenkrantz/tmp/commons/commons-convert/cvs/home/cvs/jakarta-commons//codec/src/java/org/apache/commons/codec/binary/Base64.java,v 1.2 2003/05/06 20:52:18 ggregory Exp $
- * $Revision: 1.2 $
- * $Date: 2003/05/06 20:52:18 $
+ * $Header: /home/jerenkrantz/tmp/commons/commons-convert/cvs/home/cvs/jakarta-commons//codec/src/java/org/apache/commons/codec/binary/Base64.java,v 1.3 2003/05/14 02:40:18 tobrien Exp $
+ * $Revision: 1.3 $
+ * $Date: 2003/05/14 02:40:18 $
   *
   * ====================================================================
   *
@@ -196,19 +196,8 @@ public Object decode(Object pObject) throws DecoderException {
     }
 
     public byte[] decode(byte[] pArray) throws DecoderException {
-
         byte[] result;
-
-        if (!isArrayByteBase64(pArray)) {
-            throw new DecoderException(
-                "Parameter supplied to "
-                    + "Base64 "
-                    + "decode is not a valid base64 data.");
-        } 
-        else {
-            result = decodeBase64((byte[]) pArray);
-        }
-
+        result = decodeBase64((byte[]) pArray);
         return (result);
     }
 
@@ -375,9 +364,8 @@ else if (fewerThan24bits == SIXTEENBIT) {
      * @return Array containing decoded data.
      */
     public static byte[] decodeBase64(byte[] base64Data) {
-        // RFC 2045 suggests line wrapping at (no more than) 76
-        // characters -- we may have embedded whitespace.
-        base64Data = discardWhitespace(base64Data);
+        // RFC 2045 requires that we discard ALL non-Base64 characters
+        base64Data = discardNonBase64(base64Data);
 
         // handle the edge case, so we don't have to worry about it later
         if (base64Data.length == 0) {
@@ -468,6 +456,33 @@ static byte[] discardWhitespace(byte[] data) {
 
         return packedData;
     }
+    
+	/**
+	 * Discards any characters outside of the base64 alphabet, per
+	 * the requirements on page 25 of RFC 2045 - "Any characters
+	 * outside of the base64 alphabet are to be ignored in base64
+	 * encoded data."
+	 *
+	 * @param data The base-64 encoded data to groom
+	 * @return The data, less non-base64 characters (see RFC 2045).
+	 */
+	static byte[] discardNonBase64(byte[] data) {
+		byte groomedData[] = new byte[data.length];
+		int bytesCopied = 0;
+
+		for (int i = 0; i < data.length; i++) {
+			if( isBase64(data[i]) ) {
+			  groomedData[bytesCopied++] = data[i];
+			}
+		}
+
+		byte packedData[] = new byte[bytesCopied];
+
+		System.arraycopy(groomedData, 0, packedData, 0, bytesCopied);
+
+		return packedData;
+	}
+
 
     // Implementation of the Encoder Interface
 

src/test/org/apache/commons/codec/binary/Base64Test.java

@@ -1,7 +1,7 @@
 /*
- * $Header: /home/jerenkrantz/tmp/commons/commons-convert/cvs/home/cvs/jakarta-commons//codec/src/test/org/apache/commons/codec/binary/Base64Test.java,v 1.2 2003/05/06 20:45:16 ggregory Exp $
- * $Revision: 1.2 $
- * $Date: 2003/05/06 20:45:16 $
+ * $Header: /home/jerenkrantz/tmp/commons/commons-convert/cvs/home/cvs/jakarta-commons//codec/src/test/org/apache/commons/codec/binary/Base64Test.java,v 1.3 2003/05/14 02:40:18 tobrien Exp $
+ * $Revision: 1.3 $
+ * $Date: 2003/05/14 02:40:18 $
  *
  * ====================================================================
  *
@@ -67,7 +67,7 @@
 import junit.framework.TestCase;
 
 /**
- * @version $Revision: 1.2 $ $Date: 2003/05/06 20:45:16 $
+ * @version $Revision: 1.3 $ $Date: 2003/05/14 02:40:18 $
  * @author <a href="mailto:sanders@apache.org">Scott Sanders</a>
  * @author <a href="mailto:rwaldhoff@apache.org">Rodney Waldhoff</a>
  * @author <a href="mailto:tobrien@apache.org">Tim O'Brien</a>
@@ -313,28 +313,38 @@ public void testKnownDecodings() {
         assertEquals("xyzzy!",new String(Base64.decodeBase64("eHl6enkh".getBytes())));
         } 
 
-    public void testNonBase64() throws Exception {
+	public void testNonBase64Test() throws Exception {
 
-        byte[] bArray = { '%' };
+		byte[] bArray = { '%' };
 
-        assertFalse( "Invalid Base64 array was incorrectly validated as " +
-                     "an array of Base64 encoded data", 
-                     Base64.isArrayByteBase64( bArray ) );
+		assertFalse( "Invalid Base64 array was incorrectly validated as " +
+					 "an array of Base64 encoded data", 
+					 Base64.isArrayByteBase64( bArray ) );
 
-        boolean exceptionThrown = false;
-
-        try {
-            Base64 b64 = new Base64();
-            b64.decode( bArray );
-        } 
-        catch( Exception e ) {
-            exceptionThrown = true;
-        }
-
-        assertTrue( "Exception wasn't thrown when trying to decode " +
-                    "invalid base64 encoded data", exceptionThrown );
-
-    }
+		boolean exceptionThrown = false;
+
+		try {
+			Base64 b64 = new Base64();
+			byte[] result = b64.decode( bArray );
+            
+			assertTrue( "The result should be empty as the test encoded content did " +
+				"not contain any valid base 64 characters", result.length == 0);
+		} 
+		catch( Exception e ) {
+			exceptionThrown = true;
+		}
+
+		assertFalse( "Exception was thrown when trying to decode " +
+					"invalid base64 encoded data - RFC 2045 requires that all " +
+					"non base64 character be discarded, an exception should not" +
+					" have been thrown", exceptionThrown );
+          
+
+	}
+    
+	public void testIgnoringNonBase64InDecode() throws Exception {
+		assertEquals("The quick brown fox jumped over the lazy dogs.",new String(Base64.decodeBase64("VGhlIH@$#$@%F1aWN@#@#@@rIGJyb3duIGZve\n\r\t%#%#%#%CBqd##$#$W1wZWQgb3ZlciB0aGUgbGF6eSBkb2dzLg==".getBytes())));
+	}
 
     public void testObjectDecodeWithInvalidParameter() throws Exception {
         boolean exceptionThrown = false;