hadoopmapreduceavromrunit

MRUnit with Avro NullPointerException in Serialization


I'm trying to test a Hadoop .mapreduce Avro job using MRUnit. I am receiving a NullPointerException as seen below. I've attached a portion of the pom and source code. Any assistance would be appreciated.

Thanks

The error I'm getting is :

java.lang.NullPointerException
at org.apache.hadoop.mrunit.internal.io.Serialization.copy(Serialization.java:73)
at org.apache.hadoop.mrunit.internal.io.Serialization.copy(Serialization.java:91)
at org.apache.hadoop.mrunit.internal.io.Serialization.copyWithConf(Serialization.java:104)
at org.apache.hadoop.mrunit.TestDriver.copy(TestDriver.java:608)
at org.apache.hadoop.mrunit.MapDriverBase.setInputKey(MapDriverBase.java:64)
at org.apache.hadoop.mrunit.MapDriverBase.setInput(MapDriverBase.java:104)
at org.apache.hadoop.mrunit.MapDriverBase.withInput(MapDriverBase.java:218)
at org.lab41.project.mapreduce.ParseMetadataAsTextIntoAvroTest.testMap(ParseMetadataAsTextIntoAvroTest.java:115)
.....

pom snippet:

<dependency>
    <groupId>org.apache.mrunit</groupId>
    <artifactId>mrunit</artifactId>
    <version>0.9.0-incubating</version>
    <classifier>hadoop2</classifier>
    <scope>test</scope>
</dependency>


<avro.version>1.7.4</avro.version>
<hadoop.version>2.0.0-mr1-cdh4.1.3</hadoop.version>

<dependency>
    <groupId>org.apache.avro</groupId>
    <artifactId>avro</artifactId>
    <version>${avro.version}</version>
</dependency>

<dependency>
    <groupId>org.apache.hadoop</groupId>
    <artifactId>hadoop-client</artifactId>
    <version>${hadoop.version}</version>
    <scope>provided</scope>
</dependency>
<dependency>
    <groupId>org.apache.hadoop</groupId>
    <artifactId>hadoop-core</artifactId>
    <version>${hadoop.version}</version>
    <scope>provided</scope>
 </dependency>
 <dependency>
    <groupId>org.apache.avro</groupId>
    <artifactId>avro-mapred</artifactId>
    <version>${avro.version}</version>
    <classifier>hadoop2</classifier>
 </dependency>

Here is an excerpt of the test :

import static org.junit.Assert.*;

import java.io.IOException;
import java.nio.ByteBuffer;
import java.nio.CharBuffer;
import java.nio.charset.Charset;
import java.nio.charset.CharsetEncoder;
import java.text.ParseException;
import java.text.SimpleDateFormat;

import org.apache.avro.mapred.AvroKey;
import org.apache.avro.hadoop.io.AvroSerialization;
import org.apache.avro.mapred.AvroValue;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mrunit.mapreduce.MapDriver;
import org.apache.hadoop.mrunit.types.Pair;
import org.junit.After;
import org.junit.AfterClass;
import org.junit.Before;
import org.junit.BeforeClass;
import org.junit.Test;
import org.lab41.project.domain.DataRecord;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

public class ParseMetadataAsTextIntoAvroTest {

    Logger logger = LoggerFactory
            .getLogger(ParseMetadataAsTextIntoAvroTest.class);
    private MapDriver<LongWritable, Text, AvroKey<Long>, AvroValue<DataRecord>> mapDriver;

    @BeforeClass
    public static void setUpClass() {
    }

    @AfterClass
    public static void tearDownClass() {
    }

    @Before
    public void setUp() throws IOException {
        ParseMetadataAsTextIntoAvroMapper mapper = new ParseMetadataAsTextIntoAvroMapper();

        mapDriver = new MapDriver<LongWritable, Text, AvroKey<Long>, AvroValue<DataRecord>>();
        mapDriver.setMapper(mapper);
        mapDriver.getConfiguration().setStrings("io.serializations", new String[]{  
            AvroSerialization.class.getName()
        });
    }

    @Test
    public void testMap() throws ParseException, IOException {
        Text testInputText = new Text(test0);

        DataRecord record = new DataRecord();
       ….

        AvroKey<Long> expectedPivot = new AvroKey<Long>(1L);
        AvroValue<DataRecord> expectedRecord = new AvroValue<DataRecord>(record);

        mapDriver.withInput(new Pair<LongWritable, Text>(new LongWritable(1), testInputText));
        mapDriver.withOutput(new Pair<AvroKey<Long>, AvroValue<DataRecord>>(expectedPivot, expectedRecord));
        mapDriver.runTest();

    }
}

Solution

  • In order to get this to work you have add the AvroSerializatio to the default serailizations. You also have to configure AvroSerializationn.

     @Before
    public void setUp() throws IOException {
        ParseMetadataAsTextIntoAvroMapper mapper = new ParseMetadataAsTextIntoAvroMapper();
        mapDriver = new MapDriver<LongWritable, Text, AvroKey<Long>, AvroValue<NetworkRecord>>();
        mapDriver.setMapper(mapper);
    
        //Copy over the default io.serializations. If you don't do this then you will 
        //not be able to deserialize the inputs to the mapper
        String[] strings = mapDriver.getConfiguration().getStrings("io.serializations");
        String[] newStrings = new String[strings.length +1];
        System.arraycopy( strings, 0, newStrings, 0, strings.length );
        newStrings[newStrings.length-1] = AvroSerialization.class.getName();
    
        //Now you have to configure AvroSerialization by sepecifying the key
        //writer Schema and the value writer schema.
        mapDriver.getConfiguration().setStrings("io.serializations", newStrings);
        mapDriver.getConfiguration().setStrings("avro.serialization.key.writer.schema", Schema.create(Schema.Type.LONG).toString(true));
        mapDriver.getConfiguration().setStrings("avro.serialization.value.writer.schema", NetworkRecord.SCHEMA$.toString(true));
    }