ClassCastException无法将Writable转换为Text

问题描述

@H_502_0@我创建了一个名为TextArrayWritable的子类来保存文本数组。看起来像这样:

import org.apache.hadoop.io.ArrayWritable;
import org.apache.hadoop.io.Text;
public class TextArrayWritable extends ArrayWritable{
    public TextArrayWritable() {
        super(Text.class);
    }
    public TextArrayWritable(Text[] values) {
        super(Text.class,values);
    }
}
@H_502_0@在我的映射器中,我生成一个键值对,其中键是Text,值是上面的TextArrayWritable。映射器的代码如下所示:

import java.io.IOException;

import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;


public class TagWordMapper extends Mapper<LongWritable,Text,TextArrayWritable>{
    
    @Override
    public void map(LongWritable key,Text value,Context context) throws IOException,InterruptedException{
        if(key.get()==0) {
            return;
        }else {
            String line = value.toString();
            Text title = new Text(line.split("\t")[2]);
            Text likes =  new Text(line.split("\t")[8]);
            Text tags =  new Text(line.split("\t")[6]);
            
            Text[] temp = new Text[2];
            temp[0]=likes;
            temp[1]=tags;
            
            context.write(title,new TextArrayWritable(temp));
        }
    }
}
@H_502_0@问题发生在化简器中,当我循环遍历TextArrayWritable时,即使我将其强制转换为Text [],也发生了异常,因为TextArrayWritable#get()返回了Text数组。减速器代码如下:

import java.io.IOException;
import java.util.regex.*;

import org.apache.hadoop.io.Text;
import org.apache.hadoop.io.Writable;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.mapreduce.Reducer;

public class TagWordCombiner extends Reducer<Text,TextArrayWritable,IntWritable>{

    @Override
    public void reduce(Text title,Iterable<TextArrayWritable> arrays,InterruptedException{
        int count=0;
        Pattern r = Pattern.compile("\\bcute\\b",Pattern.CASE_INSENSITIVE);
        for(TextArrayWritable array:arrays) {
            
            Text[] temp = (Text[]) array.get();
            int likes = Integer.parseInt(temp[0].toString());
            String tags = temp[1].toString();
            
            Matcher matcher = r.matcher(tags);
            boolean found = matcher.find();
            if(found && likes>=3000){
                count+=1;
            }
        }
        context.write(title,new IntWritable(count));
    }
}
@H_502_0@即使我将返回类型转换为Text [],也知道为什么会发生这种情况。任何见解均表示赞赏。谢谢您的阅读。

解决方法

我决定不使用TextArrayWritable,而是选择使用Text以字符串形式发送数据数组。此外,我发现它可以使组合器正常工作。组合器的输入/输出数据类型必须与映射器的输出数据类型匹配。

这是我的代码:

映射器:

package combiner;
import java.io.IOException;

import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;

public class CombinerMapper extends Mapper<LongWritable,Text,Text>{

    @Override
    public void map(LongWritable key,Text value,Context context) throws IOException,InterruptedException{
        if(key.get()==0) {
            return;
        }else {
            String line = value.toString();
            Text title = new Text(line.split("\t")[2]);
            Text arr = new Text(line.split("\t")[8]+"\t"+line.split("\t")[6]);
            context.write(title,arr);
        }
        
    }
}

它的作用是映射一个.csv文件。它跳过第一行,因为第一行是标题并且没有用。在下一行中,它被“ \ t”分隔以按顺序提取“标题”,“喜欢”,“标签”。仅将“喜欢”和“标签”插入到数组中。 Mapper的输出数据类型为Text,Text。

组合器:

package combiner;
import java.util.regex.*;
import java.io.IOException;

import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;

public class CombinerCombiner extends Reducer<Text,Text>{

    @Override
    public void reduce(Text title,Iterable<Text> values,InterruptedException{
        int sum=0;
        for(Text val:values) {
            String line = val.toString();
            int likes = Integer.parseInt(line.split("\t")[0]);
            String tags = line.split("\t")[1];
            
            Pattern r = Pattern.compile("\\bcute\\b",Pattern.CASE_INSENSITIVE);
            Matcher matcher = r.matcher(tags);
            boolean found = matcher.find();
            
            if(found && likes>=3000) {
                sum+=1;
            }
        }
        if(sum==0) {
            return;
        }else {
            context.write(title,new Text(Integer.toString(sum)));
        }
    }
}

组合器如何工作的棘手之处在于其输入/输出必须与Mapper的输出数据类型匹配,否则它将抛出我遇到的错误。在这种情况下,我将组合器用作过滤器,以过滤出没有> = 3000个“赞”且标题中的标签不包含任何形式的“可爱”一词的标题。它输入为数据类型,并输出为数据类型,两者都与Mapper的输出数据类型相同。

最后,减速器:

package combiner;
import java.io.IOException;

import org.apache.hadoop.io.Text;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.mapreduce.Reducer;

public class CombinerReducer extends Reducer<Text,IntWritable>{

    public void reduce(Text title,Iterable<Text>counts,InterruptedException{
        int sum = 0;
        for(Text count:counts) {
            String line = count.toString();
            int temp = Integer.parseInt(line);
            sum+=temp;
        }
        context.write(title,new IntWritable(sum));
    }

它用作常规的Reducer,将数据数组求和并产生最终和。循环遍历Iterable是没有意义的,因为它是在组合器中为我们完成的,但是它可以工作。

这是我的驱动程序类:

package combiner;

import org.apache.hadoop.fs.Path;

import org.apache.hadoop.io.Text;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;

import org.apache.hadoop.mapreduce.Job;

public class Combiner {

    public static void main(String[] args) throws Exception{
        //Check if proper arguments is inputed
        if (args.length != 2) {
              System.out.printf(
                  "Usage: TagWord <input dir> <output dir>\n");
              System.exit(-1);
        }
        
        Job job = new Job();
        job.setJarByClass(Combiner.class);
        job.setJobName("Combiner");
        
        //Declare the input file path
        FileInputFormat.setInputPaths(job,new Path(args[0]));
        
        //Declare the output file path
        FileOutputFormat.setOutputPath(job,new Path(args[1]));
        
        //Initiate the the -cers
        job.setMapperClass(CombinerMapper.class);
        job.setCombinerClass(CombinerCombiner.class);
        job.setReducerClass(CombinerReducer.class);
        
        //map output
        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(Text.class);
        
        //reducer output
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(IntWritable.class);
        
        //Initiate boolean to check whether MapReduce is successful
        boolean success = job.waitForCompletion(true);
        
        //Execute code based on whether the job is successful
        System.exit(success ? 0 : 1);
    }
}

请注意在映射器和化简器类之间的组合器类。另外,请确保包括setMapOutputKeyClass和setMapOutputValueClass。这允许Mapper类使用不同的输出数据类型。它还会强制化简器输出数据类型为您为setOutputKeyClass和setOutputValueClass设置的任何数据,在我们的情况下,我们的化简器必须输出Text和IntWritable。

如有任何疑问,请发表评论。 Hadoop的信息很少。