From 7b8f87eb7842536e2400dbd4b555955b230ad69b Mon Sep 17 00:00:00 2001
From: Andrew Wang <wang@apache.org>
Date: Tue, 28 Mar 2017 21:40:53 -0700
Subject: [PATCH] HDFS-9705. Refine the behaviour of getFileChecksum when
 length = 0. Contributed by SammiChen and Kai Zheng.

(cherry picked from commit 82b4a9c3d058bcf54051d3075ca14f30988f5cb4)
---
 .../org/apache/hadoop/hdfs/DFSClient.java     |  6 +--
 .../hdfs/TestDistributedFileSystem.java       | 40 ++++++++++++++-----
 .../hadoop/hdfs/TestFSOutputSummer.java       |  8 +++-
 3 files changed, 40 insertions(+), 14 deletions(-)

diff --git a/hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/DFSClient.java b/hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/DFSClient.java
index 25fe029a53d..962dd7990b3 100644
--- a/hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/DFSClient.java
+++ b/hadoop-hdfs-project/hadoop-hdfs-client/src/main/java/org/apache/hadoop/hdfs/DFSClient.java
@@ -1950,12 +1950,10 @@ public class DFSClient implements java.io.Closeable, RemotePeerFactory,
       // If there is no block allocated for the file,
       // return one with the magic entry that matches what previous
       // hdfs versions return.
-      if (locatedblocks.size() == 0) {
+      if (locatedblocks.size() == 0 || length == 0) {
         return new MD5MD5CRC32GzipFileChecksum(0, 0, fileMD5);
       }
-
-      // we should never get here since the validity was checked
-      // when getCrcType() was called above.
+      // We will get here if above condition is not met.
       return null;
     }
   }
diff --git a/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestDistributedFileSystem.java b/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestDistributedFileSystem.java
index 65d6cd9fe2a..2ba09e166d8 100644
--- a/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestDistributedFileSystem.java
+++ b/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestDistributedFileSystem.java
@@ -816,13 +816,6 @@ public class TestDistributedFileSystem {
         out.close();
       }
 
-      // verify the magic val for zero byte files
-      {
-        final FileChecksum zeroChecksum = hdfs.getFileChecksum(zeroByteFile);
-        assertEquals(zeroChecksum.toString(),
-            "MD5-of-0MD5-of-0CRC32:70bc8f4b72a86921468bf8e8441dce51");
-      }
-
       //write another file
       final Path bar = new Path(dir, "bar" + n);
       {
@@ -831,8 +824,37 @@ public class TestDistributedFileSystem {
         out.write(data);
         out.close();
       }
-  
-      { //verify checksum
+
+      {
+        // verify the magic val for zero byte file
+        final FileChecksum zeroChecksum = hdfs.getFileChecksum(zeroByteFile);
+        final String magicValue =
+            "MD5-of-0MD5-of-0CRC32:70bc8f4b72a86921468bf8e8441dce51";
+        // verify the magic val for zero byte files
+        assertEquals(magicValue, zeroChecksum.toString());
+
+        // verify checksum for empty file and 0 request length
+        final FileChecksum checksumWith0 = hdfs.getFileChecksum(bar, 0);
+        assertEquals(zeroChecksum, checksumWith0);
+
+        // verify none existent file
+        try {
+          hdfs.getFileChecksum(new Path(dir, "none-existent"), 8);
+          fail();
+        } catch (Exception ioe) {
+          FileSystem.LOG.info("GOOD: getting an exception", ioe);
+        }
+
+        // verify none existent file and 0 request length
+        try {
+          final FileChecksum noneExistentChecksumWith0 =
+              hdfs.getFileChecksum(new Path(dir, "none-existent"), 0);
+          fail();
+        } catch (Exception ioe) {
+          FileSystem.LOG.info("GOOD: getting an exception", ioe);
+        }
+
+        // verify checksums
         final FileChecksum barcs = hdfs.getFileChecksum(bar);
         final int barhashcode = barcs.hashCode();
         assertEquals(hdfsfoocs.hashCode(), barhashcode);
diff --git a/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestFSOutputSummer.java b/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestFSOutputSummer.java
index 9dcd449661b..ebbf2eab66a 100644
--- a/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestFSOutputSummer.java
+++ b/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestFSOutputSummer.java
@@ -19,6 +19,7 @@ package org.apache.hadoop.hdfs;
 
 import static org.apache.hadoop.fs.CommonConfigurationKeysPublic.IO_FILE_BUFFER_SIZE_KEY;
 import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertNull;
 import static org.junit.Assert.assertTrue;
 
 import java.io.IOException;
@@ -27,6 +28,7 @@ import java.util.Random;
 import org.apache.hadoop.conf.Configuration;
 import org.apache.hadoop.fs.FSDataInputStream;
 import org.apache.hadoop.fs.FSDataOutputStream;
+import org.apache.hadoop.fs.FileChecksum;
 import org.apache.hadoop.fs.FileSystem;
 import org.apache.hadoop.fs.Path;
 import org.junit.Test;
@@ -104,7 +106,11 @@ public class TestFSOutputSummer {
     checkAndEraseData(actual, 0, expected, "Read Sanity Test");
     stm.close();
     // do a sanity check. Get the file checksum
-    fileSys.getFileChecksum(name);
+    FileChecksum fileChecksum = fileSys.getFileChecksum(name);
+    if (fileSys.getConf().get(
+        DFSConfigKeys.DFS_CHECKSUM_TYPE_KEY).toString().equals("NULL")) {
+      assertNull(fileChecksum);
+    }
   }
 
   private void cleanupFile(Path name) throws IOException {