Fix issue #20: write uncompressed bytes if uncompressed size==compressed size

build.xml

@@ -28,7 +28,7 @@
 
   <property name="Name" value="Hadoop GPL Compression"/>
   <property name="name" value="hadoop-lzo"/>
-  <property name="version" value="0.4.10"/>
+  <property name="version" value="0.4.11"/>
   <property name="final.name" value="${name}-${version}"/>
   <property name="year" value="2008"/>
 

src/java/com/hadoop/compression/lzo/LzopOutputStream.java

@@ -177,7 +177,7 @@ protected void compress() throws IOException {
       // the LZO specification says that we should write the uncompressed bytes rather
       // than the compressed bytes.  The decompressor understands this because both sizes
       // get written to the stream.
-      if (compressor.getBytesRead() < compressor.getBytesWritten()) {
+      if (compressor.getBytesRead() <= compressor.getBytesWritten()) {
         // Compression actually increased the size of the buffer, so write the uncompressed bytes.
         byte[] uncompressed = ((LzoCompressor)compressor).uncompressedBytes();
         rawWriteInt(uncompressed.length);

src/test/com/hadoop/compression/lzo/TestLzopOutputStream.java

@@ -46,6 +46,7 @@ public class TestLzopOutputStream extends TestCase {
   private final String bigFile = "100000.txt";
   private final String mediumFile = "1000.txt";
   private final String smallFile = "100.txt";
+  private final String issue20File = "issue20-lzop.txt";
 
   @Override
   protected void setUp() throws Exception {
@@ -84,6 +85,25 @@ public void testSmallFile() throws NoSuchAlgorithmException, IOException,
     runTest(smallFile);
   }
 
+  /**
+   * The LZO specification says that we should write the uncompressed bytes
+   * rather than the compressed bytes if the compressed buffer is actually
+   * larger ('>') than the uncompressed buffer.
+   *
+   * To conform to the standard, this means we have to write the uncompressed
+   * bytes also when they have exactly the same size as the compressed bytes.
+   * (the '==' in '<=').
+   *
+   * The input data of this test is known to compress to the same size as the
+   * uncompressed data.  Hence we verify that we handle the boundary condition
+   * correctly.
+   *
+   */
+  public void testIssue20File() throws NoSuchAlgorithmException, IOException,
+  InterruptedException {
+    runTest(issue20File);
+  }
+
   /**
    * Test that reading an lzo-compressed file produces the same lines as reading the equivalent
    * flat file.  The test opens both the compressed and flat file, successively reading each

src/test/data/issue20-lzop.txt

@@ -0,0 +1,6 @@
+0.5	74	25425
+0.9	200	25384
+0.95	203	4
+0.98	211	2
+0.99	219	3
+0.995	240	5