hadoop - 运行 MRUnit 时出现 java.lang.NoSuchMethodError : org. apache.hadoop.mapreduce.TaskAttemptID

Question

我正在尝试使用 Cloudera hadoop 发行版开发 mr-job。我正在使用 api 版本 2。 我确实遇到了 mr-unit 的麻烦。请建议该怎么做。我已经使用了标准的原型(prototype)并且完全迷失了，我不明白问题的根源在哪里。 这是我的依赖项:

<dependency>
            <groupId>com.cloudera.hadoop</groupId>
            <artifactId>hadoop-core</artifactId>
            <version>0.20.2-320</version>
            <scope>provided</scope>
        </dependency>

        <dependency>
            <groupId>com.cloudera.hadoop</groupId>
            <artifactId>hadoop-mrunit</artifactId>
            <version>0.20.2-320</version>
            <scope>test</scope>
        </dependency>

这是我的测试代码:

@Test
    public void testEmptyOutput() throws Exception{
        for(String line : linesFromFlatFile){
            //List<Pair<GetReq, IntWritable>> output = 
                    driver.withInput(UNUSED_LONG_KEY, new Text(line) )
                 // .withOutput(null, null )
                  .run();
            //assertTrue("", output.isEmpty());
        }
    }

这里有一个异常(exception):

> Tests run: 1, Failures: 0, Errors: 1, Skipped: 0, Time elapsed: 0.283
> sec <<< FAILURE!
> testEmptyOutput(MapperTest)
> Time elapsed: 0.258 sec  <<< ERROR! java.lang.NoSuchMethodError:
> org.apache.hadoop.mapreduce.TaskAttemptID.<init>(Ljava/lang/String;IZII)V
>   at
> org.apache.hadoop.mrunit.mapreduce.mock.MockMapContextWrapper$MockMapContext.<init>(MockMapContextWrapper.java:71)
>   at
> org.apache.hadoop.mrunit.mapreduce.mock.MockMapContextWrapper.getMockContext(MockMapContextWrapper.java:144)
>   at
> org.apache.hadoop.mrunit.mapreduce.MapDriver.run(MapDriver.java:197)
>   at
MapperTest.testEmptyOutput(ScoringCounterMapperTest.java:42)

package mypackage;

import java.util.Date;
import java.util.List;

import junit.framework.TestCase;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mrunit.mapreduce.MapDriver;
import org.apache.hadoop.mrunit.types.Pair;
import org.junit.Before;
import org.junit.Test;

import Sample;
import GetReq;



public class MapperTest extends TestCase {
    private static final IntWritable ONE_OCCURANCE = new IntWritable(1);
    private static final LongWritable UNUSED_LONG_KEY = new LongWritable(new Date().getTime());

    private Mapper<LongWritable, Text, GetReq, IntWritable> mapper;
    private MapDriver<LongWritable, Text, GetReq, IntWritable> driver;

    List<String> linesFromFlatFileNoOutput = null;
    List<String> linesFromFlatFileWithOutput = null;
    @Before 
    public void setUp() {
        mapper = newMapper();
        driver = new MapDriver<LongWritable, Text, GetReq, IntWritable>(mapper);
        Mapper.METADATA_CSV ="../../data/metadata.csv"; //ugly hook
        linesFromFlatFileNoOutput = Sample.instance.getLinesFromFlatFileNoOutput();  
        linesFromFlatFileWithOutput = Sample.instance.getLinesFromFlatFileWithOutput(); 
    }

    @Test
    public void testEmptyOutput() throws Exception{
        for(String line : linesFromFlatFileNoOutput){
            //List<Pair<GetReq, IntWritable>> output = 
                    driver.withInput(UNUSED_LONG_KEY, new Text(line) )
                  .withOutput(null, null )
                  .runTest();
            //assertTrue("", output.isEmpty());
        }
    }

    @Test
    public void testResultOutput() throws Exception{
        for(String line : linesFromFlatFileWithOutput){
            driver.withInput(UNUSED_LONG_KEY, new Text(line) )
                //.withOutput(null, null )
                .runTest();
        }
    }
}

嗯...我没有更改 pom.xml 中的任何内容 现在我得到输出和相同的异常。看起来映射器运行一次。或者 rties 运行。我从映射器主体获得调试输出。

UPD:我添加了分类器并更改了依赖项:

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

现在我确实遇到了另一个问题:

Found interface org.apache.hadoop.mapreduce.Counter, but class was expected

在线:

context.getCounter(EnumCounter.MATCHED_RECORDS).increment(1);

我又做错了什么？

Answer 1

我找到了解决方案:需要为 mr-unit 添加分类器标签。它应该看起来像:

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

现在我有另一个问题:找到接口(interface) org.apache.hadoop.mapreduce.Counter，但类是预期的计数器增量。此问题与一些错误有关。