使用自定义writable从Hadoop Map Reduce作业输出列表

Question

我试图通过改变hadoop给出的wordcount示例来创建一个简单的map reduce作业。

我试着输出一个列表，而不是单词计数。wordcount示例给出了以下输出

hello 2
world 2

我正试图把它作为一个列表输出，这将构成未来工作的基础

hello 1 1
world 1 1

我想我的思路是对的，但我在写清单时遇到了麻烦。而不是上面的，我得到

Hello   foo.MyArrayWritable@61250ff2
World   foo.MyArrayWritable@483a0ab1

这是MyArrayWritable。我在write(DataOuptut arg0)中放了一个sys，但它从不输出任何东西，所以我认为该方法可能不会被调用，我不知道为什么。

class MyArrayWritable extends ArrayWritable{
public MyArrayWritable(Class<? extends Writable> valueClass, Writable[] values) {
    super(valueClass, values);
}
public MyArrayWritable(Class<? extends Writable> valueClass) {
    super(valueClass);
}
@Override
public IntWritable[] get() {
    return (IntWritable[]) super.get();
}
@Override
public void write(DataOutput arg0) throws IOException {
    for(IntWritable i : get()){
        i.write(arg0);
    }
}
}

EDIT -添加更多源代码

public class WordCount {
public static class Map extends Mapper<LongWritable, Text, Text, IntWritable> {
    private final static IntWritable one = new IntWritable(1);
    private Text word = new Text();
    public void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
        String line = value.toString();
        StringTokenizer tokenizer = new StringTokenizer(line);
        while (tokenizer.hasMoreTokens()) {
            word.set(tokenizer.nextToken());
            context.write(word, one);
        }
    }
} 
public static class Reduce extends Reducer<Text, IntWritable, Text, MyArrayWritable> {
    public void reduce(Text key, Iterable<IntWritable> values, Context context) throws IOException, InterruptedException {
        ArrayList<IntWritable> list = new ArrayList<IntWritable>();    
        for (IntWritable val : values) {
            list.add(val);
        }
        context.write(key, new MyArrayWritable(IntWritable.class, list.toArray(new IntWritable[list.size()])));
    }
}
public static void main(String[] args) throws Exception {
    if(args == null || args.length == 0)
        args = new String[]{"./wordcount/input","./wordcount/output"};
    Path p = new Path(args[1]);
    FileSystem fs = FileSystem.get(new Configuration());
    fs.exists(p);
    fs.delete(p, true);
    Configuration conf = new Configuration();
    Job job = new Job(conf, "wordcount");
    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(IntWritable.class);
    job.setMapperClass(Map.class);
    job.setReducerClass(Reduce.class);
    job.setJarByClass(WordCount.class);
    job.setInputFormatClass(TextInputFormat.class);
    FileInputFormat.addInputPath(job, new Path(args[0]));
    FileOutputFormat.setOutputPath(job, new Path(args[1]));
    job.waitForCompletion(true);
}

}

Answer 1

你的reducer中有一个'bug' -值迭代器在整个循环中重用相同的IntWritable，所以你应该像下面这样包装添加到列表中的值:

public void reduce(Text key, Iterable<IntWritable> values, Context context)
                                      throws IOException, InterruptedException {
    ArrayList<IntWritable> list = new ArrayList<IntWritable>();    
    for (IntWritable val : values) {
        list.add(new IntWritable(val));
    }
    context.write(key, new MyArrayWritable(IntWritable.class, list.toArray(new IntWritable[list.size()])));
}

这实际上不是一个问题，因为你使用的是一个数组列表，你的映射器只输出一个值(一个)，但如果你扩展这段代码，这可能会让你出错。

您还需要在作业中定义您的map和reducer输出类型不同:

// map output types
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(IntWritable.class);
// reducer output types
job.setOutputValueClass(Text.class);
job.setOutputValueClass(MyArrayWritable.class);

您可能想要显式地定义reducer的数量(这可能就是为什么您从未看到您的系统被写入任务日志的原因，特别是如果您的集群管理员已将默认数量定义为0):

job.setNumReduceTasks(1);

你使用默认的文本输出格式，它调用toString()输出键和值对- MyArrayWritable没有一个覆盖的toString方法，所以你应该在你的MyArrayWritable:

@Override
public String toString() {
  return Arrays.toString(get());
}

最后从MyArrayWritable中删除覆盖的write方法-这不是与附加的readFields方法兼容的有效实现。你不需要重写这个方法，但是如果你这样做了(比如你想看到一个系统来验证它是否被调用了)，那么你可以这样做:

@Override
public void write(DataOutput arg0) throws IOException {
  System.out.println("write method called");
  super.write(arg0);
}