Apache
/
hadoop
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

MAPREDUCE-5994. Simplify ByteUtils and fix failing test. Contributed by Todd Lipcon. 

git-svn-id: https://svn.apache.org/repos/asf/hadoop/common/branches/MR-2841@1613003 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Todd Lipcon 2014-07-24 06:14:27 +00:00
parent 341695e731
commit d468a92447
8 changed files with 55 additions and 161 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
hadoop-mapreduce-project/CHANGES.MAPREDUCE-2841.txt
View File

@ -2,3 +2,4 @@ Changes for Hadoop Native Map Output Collector
------------------------------------------------
MAPREDUCE-5985. native-task: Fix build on macosx. Contributed by Binglin Chang
MAPREDUCE-5994. Simplify ByteUtils and fix failing test. (todd)

10
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-nativetask/src/main/java/org/apache/hadoop/mapred/nativetask/NativeRuntime.java
View File

@ -20,20 +20,16 @@ package org.apache.hadoop.mapred.nativetask;
import java.io.IOException;
import com.google.common.base.Charsets;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataInputStream;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.DataInputBuffer;
import org.apache.hadoop.io.FloatWritable;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapred.Task.TaskReporter;
import org.apache.hadoop.mapred.nativetask.util.BytesUtil;
import org.apache.hadoop.mapred.nativetask.util.ConfigUtil;
import org.apache.hadoop.mapred.nativetask.util.SnappyUtil;
import org.apache.hadoop.util.VersionInfo;
@ -93,7 +89,7 @@ public class NativeRuntime {
*/
public synchronized static long createNativeObject(String clazz) {
assertNativeLibraryLoaded();
final long ret = JNICreateNativeObject(BytesUtil.toBytes(clazz));
final long ret = JNICreateNativeObject(clazz.getBytes(Charsets.UTF_8));
if (ret == 0) {
LOG.warn("Can't create NativeObject for class " + clazz + ", probably not exist.");
}
@ -108,7 +104,7 @@ public class NativeRuntime {
*/
public synchronized static long registerLibrary(String libraryName, String clazz) {
assertNativeLibraryLoaded();
final long ret = JNIRegisterModule(BytesUtil.toBytes(libraryName), BytesUtil.toBytes(clazz));
final long ret = JNIRegisterModule(libraryName.getBytes(Charsets.UTF_8), clazz.getBytes(Charsets.UTF_8));
if (ret != 0) {
LOG.warn("Can't create NativeObject for class " + clazz + ", probably not exist.");
}

148
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-nativetask/src/main/java/org/apache/hadoop/mapred/nativetask/util/BytesUtil.java
View File

@ -18,107 +18,38 @@
package org.apache.hadoop.mapred.nativetask.util;
import java.io.UnsupportedEncodingException;
import com.google.common.primitives.Ints;
import com.google.common.primitives.Longs;
public class BytesUtil {
public static final int SIZEOF_INT = Integer.SIZE / Byte.SIZE;
public static final int SIZEOF_LONG = Long.SIZE / Byte.SIZE;
public static byte[] toBytes(String str) {
if (str == null) {
return null;
}
try {
return str.getBytes("utf-8");
} catch (final UnsupportedEncodingException e) {
throw new RuntimeException(e.getMessage());
}
}
public static String fromBytes(byte[] data) {
if (data == null) {
return null;
}
try {
return new String(data, "utf-8");
} catch (final UnsupportedEncodingException e) {
throw new RuntimeException(e.getMessage());
}
}
/**
* Converts a byte array to an int value
* @param bytes byte array
* @return the int value
*/
public static int toInt(byte[] bytes) {
return toInt(bytes, 0, SIZEOF_INT);
}
/**
* Converts a byte array to an int value
* @param bytes byte array
* @param offset offset into array
* @param length length of int (has to be {@link #SIZEOF_INT})
* @return the int value
* @throws RuntimeException if length is not {@link #SIZEOF_INT} or
* if there's not enough room in the array at the offset indicated.
*/
public static int toInt(byte[] bytes, int offset, final int length) {
if (length != SIZEOF_INT || offset + length > bytes.length) {
throw new RuntimeException(
"toInt exception. length not equals to SIZE of Int or buffer overflow");
}
int n = 0;
for (int i = offset; i< offset + length; i++) {
n <<= 4;
n ^= bytes[i] & 0xff;
}
return n;
}
/**
* Converts a byte array to a long value.
* @param bytes array
* @return the long value
*/
public static long toLong(byte[] bytes) {
return toLong(bytes, 0, SIZEOF_LONG);
}
/**
* Converts a byte array to a long value.
* Converts a big-endian byte array to a long value.
*
* @param bytes array of bytes
* @param offset offset into array
* @return the long value
*/
public static long toLong(byte[] bytes, int offset) {
return toLong(bytes, offset, SIZEOF_LONG);
return Longs.fromBytes(bytes[offset],
bytes[offset + 1],
bytes[offset + 2],
bytes[offset + 3],
bytes[offset + 4],
bytes[offset + 5],
bytes[offset + 6],
bytes[offset + 7]);
}
/**
* Converts a byte array to a long value.
*
* @param bytes array of bytes
* @param offset offset into array
* @param length length of data (must be {@link #SIZEOF_LONG})
* @return the long value
* @throws RuntimeException if length is not {@link #SIZEOF_LONG} or
* if there's not enough room in the array at the offset indicated.
* Convert a big-endian integer from a byte array to a primitive value.
* @param bytes the array to parse from
* @param offset the offset in the array
*/
public static long toLong(byte[] bytes, int offset, final int length) {
if (length != SIZEOF_LONG || offset + length > bytes.length) {
throw new RuntimeException(
"toLong exception. length not equals to SIZE of Long or buffer overflow");
}
long l = 0;
for (int i = offset; i < offset + length; i++) {
l <<= 8;
l ^= bytes[i] & 0xff;
}
return l;
public static int toInt(byte[] bytes, int offset) {
return Ints.fromBytes(bytes[offset],
bytes[offset + 1],
bytes[offset + 2],
bytes[offset + 3]);
}
/**
@ -137,7 +68,7 @@ public class BytesUtil {
* @return Float made from passed byte array.
*/
public static float toFloat(byte [] bytes, int offset) {
return Float.intBitsToFloat(toInt(bytes, offset, SIZEOF_INT));
return Float.intBitsToFloat(toInt(bytes, offset));
}
/**
@ -154,7 +85,7 @@ public class BytesUtil {
* @return Return double made from passed bytes.
*/
public static double toDouble(final byte [] bytes, final int offset) {
return Double.longBitsToDouble(toLong(bytes, offset, SIZEOF_LONG));
return Double.longBitsToDouble(toLong(bytes, offset));
}
/**
@ -210,46 +141,13 @@ public class BytesUtil {
return new byte[] { b ? (byte) -1 : (byte) 0 };
}
/**
* Convert an int value to a byte array. Big-endian. Same as what DataOutputStream.writeInt
* does.
*
* @param val value
* @return the byte array
*/
public static byte[] toBytes(int val) {
byte [] b = new byte[4];
for(int i = 3; i > 0; i--) {
b[i] = (byte) val;
val >>>= 8;
}
b[0] = (byte) val;
return b;
}
/**
* Convert a long value to a byte array using big-endian.
*
* @param val value to convert
* @return the byte array
*/
public static byte[] toBytes(long val) {
byte [] b = new byte[8];
for (int i = 7; i > 0; i--) {
b[i] = (byte) val;
val >>>= 8;
}
b[0] = (byte) val;
return b;
}
/**
* @param f float value
* @return the float represented as byte []
*/
public static byte [] toBytes(final float f) {
// Encode it as int
return toBytes(Float.floatToRawIntBits(f));
return Ints.toByteArray(Float.floatToRawIntBits(f));
}
/**
@ -261,7 +159,7 @@ public class BytesUtil {
*/
public static byte [] toBytes(final double d) {
// Encode it as a long
return toBytes(Double.doubleToRawLongBits(d));
return Longs.toByteArray(Double.doubleToRawLongBits(d));
}
}

5
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-nativetask/src/main/java/org/apache/hadoop/mapred/nativetask/util/ConfigUtil.java
View File

@ -21,14 +21,15 @@ import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import com.google.common.base.Charsets;
import org.apache.hadoop.conf.Configuration;
public class ConfigUtil {
public static byte[][] toBytes(Configuration conf) {
List<byte[]> nativeConfigs = new ArrayList<byte[]>();
for (Map.Entry<String, String> e : conf) {
nativeConfigs.add(BytesUtil.toBytes(e.getKey()));
nativeConfigs.add(BytesUtil.toBytes(e.getValue()));
nativeConfigs.add(e.getKey().getBytes(Charsets.UTF_8));
nativeConfigs.add(e.getValue().getBytes(Charsets.UTF_8));
}
return nativeConfigs.toArray(new byte[nativeConfigs.size()][]);
}

6
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-nativetask/src/main/java/org/apache/hadoop/mapred/nativetask/util/ReadWriteBuffer.java
View File

@ -18,6 +18,8 @@
package org.apache.hadoop.mapred.nativetask.util;
import com.google.common.base.Charsets;
public class ReadWriteBuffer {
private byte[] _buff;
private int _writePoint;
@ -127,13 +129,13 @@ public class ReadWriteBuffer {
}
public void writeString(String str) {
final byte[] bytes = BytesUtil.toBytes(str);
final byte[] bytes = str.getBytes(Charsets.UTF_8);
writeBytes(bytes, 0, bytes.length);
}
public String readString() {
final byte[] bytes = readBytes();
return BytesUtil.fromBytes(bytes);
return new String(bytes, Charsets.UTF_8);
}
private void checkWriteSpaceAndResizeIfNecessary(int toBeWritten) {

19
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-nativetask/src/test/java/org/apache/hadoop/mapred/nativetask/utils/TestBytesUtil.java
View File

@ -17,6 +17,8 @@
*/
package org.apache.hadoop.mapred.nativetask.utils;
import com.google.common.primitives.Ints;
import com.google.common.primitives.Longs;
import junit.framework.Assert;
import junit.framework.TestCase;
@ -24,27 +26,18 @@ import org.apache.hadoop.mapred.nativetask.util.BytesUtil;
@SuppressWarnings({ "deprecation" })
public class TestBytesUtil extends TestCase {
public void testBytesStringConversion() {
final String str = "I am good!";
final byte[] bytes = BytesUtil.toBytes(str);
Assert.assertEquals(str, BytesUtil.fromBytes(bytes));
}
public void testBytesIntConversion() {
final int a = 1000;
final byte[] intBytes = BytesUtil.toBytes(a);
final byte[] intBytes = Ints.toByteArray(a);
Assert.assertEquals(a, BytesUtil.toInt(intBytes));
Assert.assertEquals(a, BytesUtil.toInt(intBytes, 0));
}
public void testBytesLongConversion() {
final long l = 1000000L;
final byte[] longBytes = BytesUtil.toBytes(l);
final byte[] longBytes = Longs.toByteArray(l);
Assert.assertEquals(l, BytesUtil.toLong(longBytes));
Assert.assertEquals(l, BytesUtil.toLong(longBytes, 0));
}
public void testBytesFloatConversion() {

4
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-nativetask/src/test/java/system/function/org/apache/hadoop/mapred/nativetask/kvtest/KVJob.java
View File

@ -20,12 +20,12 @@ package org.apache.hadoop.mapred.nativetask.kvtest;
import java.io.IOException;
import java.util.zip.CRC32;
import com.google.common.primitives.Longs;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.mapred.nativetask.testutil.BytesFactory;
import org.apache.hadoop.mapred.nativetask.testutil.TestConstants;
import org.apache.hadoop.mapred.nativetask.util.BytesUtil;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.Reducer;
@ -63,7 +63,7 @@ public class KVJob {
resultlong += crc32.getValue();
}
final VTYPE V = null;
context.write(key, (VTYPE) BytesFactory.newObject(BytesUtil.toBytes(resultlong), V.getClass().getName()));
context.write(key, (VTYPE) BytesFactory.newObject(Longs.toByteArray(resultlong), V.getClass().getName()));
}
}

23
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-nativetask/src/test/java/system/function/org/apache/hadoop/mapred/nativetask/testutil/BytesFactory.java
View File

@ -19,6 +19,8 @@ package org.apache.hadoop.mapred.nativetask.testutil;
import java.util.Random;
import com.google.common.primitives.Ints;
import com.google.common.primitives.Longs;
import org.apache.hadoop.io.BooleanWritable;
import org.apache.hadoop.io.ByteWritable;
import org.apache.hadoop.io.BytesWritable;
@ -39,17 +41,17 @@ public class BytesFactory {
public static Object newObject(byte[] seed, String className) {
r.setSeed(seed.hashCode());
if (className.equals(IntWritable.class.getName())) {
return new IntWritable(BytesUtil.toInt(seed));
return new IntWritable(Ints.fromByteArray(seed));
} else if (className.equals(FloatWritable.class.getName())) {
return new FloatWritable(r.nextFloat());
} else if (className.equals(DoubleWritable.class.getName())) {
return new DoubleWritable(r.nextDouble());
} else if (className.equals(LongWritable.class.getName())) {
return new LongWritable(BytesUtil.toLong(seed));
return new LongWritable(Longs.fromByteArray(seed));
} else if (className.equals(VIntWritable.class.getName())) {
return new VIntWritable(BytesUtil.toInt(seed));
return new VIntWritable(Ints.fromByteArray(seed));
} else if (className.equals(VLongWritable.class.getName())) {
return new VLongWritable(BytesUtil.toLong(seed));
return new VLongWritable(Longs.fromByteArray(seed));
} else if (className.equals(BooleanWritable.class.getName())) {
return new BooleanWritable(seed[0] % 2 == 1 ? true : false);
} else if (className.equals(Text.class.getName())) {
@ -75,24 +77,25 @@ public class BytesFactory {
public static <VTYPE> byte[] toBytes(VTYPE obj) {
final String className = obj.getClass().getName();
if (className.equals(IntWritable.class.getName())) {
return BytesUtil.toBytes(((IntWritable) obj).get());
return Ints.toByteArray(((IntWritable) obj).get());
} else if (className.equals(FloatWritable.class.getName())) {
return BytesUtil.toBytes(((FloatWritable) obj).get());
} else if (className.equals(DoubleWritable.class.getName())) {
return BytesUtil.toBytes(((DoubleWritable) obj).get());
} else if (className.equals(LongWritable.class.getName())) {
return BytesUtil.toBytes(((LongWritable) obj).get());
return Longs.toByteArray(((LongWritable) obj).get());
} else if (className.equals(VIntWritable.class.getName())) {
return BytesUtil.toBytes(((VIntWritable) obj).get());
return Ints.toByteArray(((VIntWritable) obj).get());
} else if (className.equals(VLongWritable.class.getName())) {
return BytesUtil.toBytes(((VLongWritable) obj).get());
return Longs.toByteArray(((VLongWritable) obj).get());
} else if (className.equals(BooleanWritable.class.getName())) {
return BytesUtil.toBytes(((BooleanWritable) obj).get());
} else if (className.equals(Text.class.getName())) {
return BytesUtil.toBytes(((Text) obj).toString());
return ((Text)obj).copyBytes();
} else if (className.equals(ByteWritable.class.getName())) {
return BytesUtil.toBytes(((ByteWritable) obj).get());
return Ints.toByteArray((int) ((ByteWritable) obj).get());
} else if (className.equals(BytesWritable.class.getName())) {
// TODO: copyBytes instead?
return ((BytesWritable) obj).getBytes();
} else {
return new byte[0];