From f4ed787669bc2d9337f503f0a5188167b490a961 Mon Sep 17 00:00:00 2001
From: cnauroth <cnauroth@apache.org>
Date: Tue, 4 Nov 2014 10:27:41 -0800
Subject: [PATCH] HADOOP-11165. TestUTF8 fails when run against java 8.
 Contributed by Stephen Chu.

(cherry picked from commit 85da71c2d3c565a8920e47fe3925e8e0bef353a5)
---
 .../hadoop-common/CHANGES.txt                 |  3 +++
 .../java/org/apache/hadoop/io/TestUTF8.java   | 23 +++++++++++++------
 2 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/hadoop-common-project/hadoop-common/CHANGES.txt b/hadoop-common-project/hadoop-common/CHANGES.txt
index bcafb120af8..9d73cf079da 100644
--- a/hadoop-common-project/hadoop-common/CHANGES.txt
+++ b/hadoop-common-project/hadoop-common/CHANGES.txt
@@ -35,6 +35,9 @@ Release 2.7.0 - UNRELEASED
     HADOOP-11186. documentation should talk about
     hadoop.htrace.spanreceiver.classes, not hadoop.trace.spanreceiver.classes (cmccabe)
 
+    HADOOP-11165. TestUTF8 fails when run against java 8.
+    (Stephen Chu via cnauroth)
+
 Release 2.6.0 - UNRELEASED
 
   INCOMPATIBLE CHANGES
diff --git a/hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/io/TestUTF8.java b/hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/io/TestUTF8.java
index b3872248327..ede59406768 100644
--- a/hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/io/TestUTF8.java
+++ b/hadoop-common-project/hadoop-common/src/test/java/org/apache/hadoop/io/TestUTF8.java
@@ -19,8 +19,11 @@
 package org.apache.hadoop.io;
 
 import junit.framework.TestCase;
+import java.io.ByteArrayInputStream;
+import java.io.DataInputStream;
 import java.io.IOException;
 import java.io.UTFDataFormatException;
+import java.nio.ByteBuffer;
 import java.util.Random;
 
 import org.apache.hadoop.test.GenericTestUtils;
@@ -54,11 +57,22 @@ public void testGetBytes() throws Exception {
       // generate a random string
       String before = getTestString();
 
-      // check its utf8
-      assertEquals(before, new String(UTF8.getBytes(before), "UTF-8"));
+      // Check that the bytes are stored correctly in Modified-UTF8 format.
+      // Note that the DataInput and DataOutput interfaces convert between
+      // bytes and Strings using the Modified-UTF8 format.
+      assertEquals(before, readModifiedUTF(UTF8.getBytes(before)));
     }
   }
 
+  private String readModifiedUTF(byte[] bytes) throws IOException {
+    final short lengthBytes = (short)2;
+    ByteBuffer bb = ByteBuffer.allocate(bytes.length + lengthBytes);
+    bb.putShort((short)bytes.length).put(bytes);
+    ByteArrayInputStream bis = new ByteArrayInputStream(bb.array());
+    DataInputStream dis = new DataInputStream(bis);
+    return dis.readUTF();
+  }
+
   public void testIO() throws Exception {
     DataOutputBuffer out = new DataOutputBuffer();
     DataInputBuffer in = new DataInputBuffer();
@@ -80,11 +94,6 @@ public void testIO() throws Exception {
       in.reset(out.getData(), out.getLength());
       String after2 = in.readUTF();
       assertEquals(before, after2);
-
-      // test that it is compatible with Java's other decoder
-      String after3 = new String(out.getData(), 2, out.getLength()-2, "UTF-8");
-      assertEquals(before, after3);
-
     }
 
   }