Hadoop 无法使用Pig写入SequenceFile

我想将一些Pig变量存储到Hadoop SequenceFile中，以便运行外部MapReduce作业

假设我的数据具有（chararray，int）模式：

我编写了这个存储函数：

import java.io.IOException;
import java.util.logging.Level;
import java.util.logging.Logger;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;

import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.OutputFormat;
import org.apache.hadoop.mapreduce.RecordWriter;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;
import org.apache.pig.StoreFunc;
import org.apache.pig.data.Tuple;


public class StoreTest extends StoreFunc {

    private String storeLocation;
    private RecordWriter writer;
    private Job job;

    public StoreTest(){

    }

    @Override
    public OutputFormat getOutputFormat() throws IOException {
        //return new TextOutputFormat();
        return new SequenceFileOutputFormat();
    }

    @Override
    public void setStoreLocation(String location, Job job) throws IOException {
        this.storeLocation = location;
        this.job = job;
        System.out.println("Load location is " + storeLocation);
        FileOutputFormat.setOutputPath(job, new Path(location));        
        System.out.println("Out path " + FileOutputFormat.getOutputPath(job));
    }

    @Override
    public void prepareToWrite(RecordWriter writer) throws IOException {
        this.writer = writer;
    }

    @Override
    public void putNext(Tuple tuple) throws IOException {
        try {
            Text k = new Text(((String)tuple.get(0)));
            IntWritable v = new IntWritable((Integer)tuple.get(1));
            writer.write(k, v);
        } catch (InterruptedException ex) {
            Logger.getLogger(StoreTest.class.getName()).log(Level.SEVERE, null, ex);
        }

    }
}

这只猪的代码是：

register MyUDFs.jar;
x = load '/user/pinoli/input' as (a:chararray,b:int);
store x into '/user/pinoli/output/' using StoreTest(); 

但是，存储失败，我收到以下错误：

ERROR org.apache.pig.tools.pigstats.PigStats - ERROR 0: java.io.IOException: wrong key class: org.apache.hadoop.io.Text is not class org.apache.hadoop.io.LongWritable

---

有没有办法解决这个问题？？？

问题是您没有设置输出键/值类。您可以在

setStoreLocation（）

方法中执行此操作：

@Override
public void setStoreLocation(String location, Job job) throws IOException {
    this.storeLocation = location;
    this.job = job;
    this.job.setOutputKeyClass(Text.class);   // !!!
    this.job.setOutputValueClass(IntWritable.class);  // !!!
    ...

}

我猜您希望将storer与不同的键/值类型一起使用。在这种情况下，您可以将它们的类型传递给构造函数。 例如：

@Override
public void setStoreLocation(String location, Job job) throws IOException {
    this.storeLocation = location;
    this.job = job;
    this.job.setOutputKeyClass(Text.class);   // !!!
    this.job.setOutputValueClass(IntWritable.class);  // !!!
    ...

}

private Class<? extends WritableComparable> keyClass;
private Class<? extends Writable> valueClass;
...

public StoreTest() {...}

@SuppressWarnings({ "unchecked", "rawtypes" })
    public StoreTest(String keyClass, String valueClass) {
        try {
            this.keyClass = (Class<? extends WritableComparable>) Class.forName(keyClass);
            this.valueClass = (Class<? extends Writable>) Class.forName(valueClass);
        }
        catch (Exception e) {
            throw new RuntimeException("Invalid key/value type", e);
        }
    }

...

 @Override
    public void setStoreLocation(String location, Job job) throws IOException {
        this.storeLocation = location;
        this.job = job;
        this.job.setOutputKeyClass(keyClass);
        this.job.setOutputValueClass(valueClass);
        ...
    }

register MyUDFs.jar;
DEFINE myStorer StoreTest('org.apache.hadoop.io.Text', 'org.apache.hadoop.io.IntWritable');
x = load '/user/pinoli/input' as (a:chararray,b:int);
store x into '/user/pinoli/output/' using myStorer();