我想运行两个映射器，在不同的目录中产生两个不同的输出。第一个映射器的输出(作为参数发送)应该发送到第二个映射器的输入。我在驱动程序类中有这段代码

import java.io.IOException;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.mapreduce.lib.output.MultipleOutputs;
import org.apache.hadoop.mapreduce.lib.output.TextOutputFormat;


public class Export_Column_Mapping 
{
    private static String[] Detail_output_column_array = new String[27];
    private static String[] Shop_output_column_array = new String[8];
    private static String details_output = null ;
    private static String Shop_output = null;

    public static void main(String[] args) throws Exception 
    {

        String Output_filetype = args[3];
        String Input_column_number = args[4];
        String Output_column_number = args[5];

        Configuration Detailsconf = new Configuration(false);

        Detailsconf.setStrings("output_filetype",Output_filetype);
        Detailsconf.setStrings("Input_column_number",Input_column_number);
        Detailsconf.setStrings("Output_column_number",Output_column_number);

        Job Details = new Job(Detailsconf," Export_Column_Mapping");

        Details.setJarByClass(Export_Column_Mapping.class);
        Details.setJobName("DetailsFile_Job");

        Details.setMapperClass(DetailFile_Mapper.class);
        Details.setNumReduceTasks(0);

        Details.setInputFormatClass(TextInputFormat.class);
        Details.setOutputFormatClass(TextOutputFormat.class);

        FileInputFormat.setInputPaths(Details, new Path(args[0])); 
        FileOutputFormat.setOutputPath(Details, new Path(args[1]));

        if(Details.waitForCompletion(true))
        {

        Configuration Shopconf = new Configuration();

        Job Shop = new Job(Shopconf,"Export_Column_Mapping");
        Shop.setJarByClass(Export_Column_Mapping.class);
        Shop.setJobName("ShopFile_Job");

        Shop.setMapperClass(ShopFile_Mapper.class);
        Shop.setNumReduceTasks(0);

        Shop.setInputFormatClass(TextInputFormat.class);
        Shop.setOutputFormatClass(TextOutputFormat.class);

        FileInputFormat.setInputPaths(Shop, new Path(args[1])); 
        FileOutputFormat.setOutputPath(Shop, new Path(args[2]));

        MultipleOutputs.addNamedOutput(Shop, "text", TextOutputFormat.class,LongWritable.class, Text.class);
        System.exit(Shop.waitForCompletion(true) ? 0 : 1);
        }
    }

    public static class DetailFile_Mapper extends Mapper<LongWritable,Text,Text,Text>
    {   
        public void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException 
        {
            String str_Output_filetype = context.getConfiguration().get("output_filetype"); 

            String str_Input_column_number = context.getConfiguration().get("Input_column_number");
            String[] input_columns_number = str_Input_column_number.split(",");

            String str_Output_column_number= context.getConfiguration().get("Output_column_number");    
            String[] output_columns_number = str_Output_column_number.split(",");

            String str_line = value.toString();
            String[] input_column_array = str_line.split(",");

            try
            {

                for(int i = 0;i<=input_column_array.length+1; i++)
                {
                    int int_outputcolumn = Integer.parseInt(output_columns_number[i]);
                    int int_inputcolumn = Integer.parseInt(input_columns_number[i]);

                    if((int_inputcolumn != 0) && (int_outputcolumn != 0) && output_columns_number.length == input_columns_number.length)
                    {

                        Detail_output_column_array[int_outputcolumn-1] = input_column_array[int_inputcolumn-1];


                        if(details_output != null)
                        {
                            details_output = details_output+"       "+ Detail_output_column_array[int_outputcolumn-1];
                            Shop_output = Shop_output+"     "+ Shop_output_column_array[int_outputcolumn-1];

                        }else
                        {
                            details_output = Detail_output_column_array[int_outputcolumn-1];
                            Shop_output =  Shop_output_column_array[int_outputcolumn-1];

                        }
                    }
                }

            }catch (Exception e)
            {

            }
            context.write(null,new Text(details_output));
        }
    }
    public static class ShopFile_Mapper extends Mapper<LongWritable,Text,Text,Text>
    {   
        public void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException 
        {
            try
            {

                for(int i = 0;i<=Shop_output_column_array.length; i++)
                {
                    Shop_output_column_array[0] = Detail_output_column_array[0];
                    Shop_output_column_array[1] = Detail_output_column_array[1];
                    Shop_output_column_array[2] = Detail_output_column_array[2];
                    Shop_output_column_array[3] = Detail_output_column_array[3];
                    Shop_output_column_array[4] = Detail_output_column_array[14];

                    if(details_output != null)
                    {
                        Shop_output = Shop_output+"     "+ Shop_output_column_array[i];

                    }else
                    {
                        Shop_output =  Shop_output_column_array[i-1];

                    }
                }
            }catch (Exception e){

            }
            context.write(null,new Text(Shop_output));
        }
    }

}

我收到错误..

Error:org.apache.hadoop.mapreduce.lib.input.InvalidInputException: Input path does not exist: file:/home/Barath.B.Natarajan.ap/rules/text.txt

我想一个接一个地运行这些工作，有人可以帮助我吗？...

最佳答案

有一种叫做 jobcontrol 的东西，你可以用它来实现它。

假设有两个工作A和B

ControlledJob A= new ControlledJob(JobConf for A);
ControlledJob B= new ControlledJob(JobConf for B);
B.addDependingJob(A);

JobControl jControl = newJobControl("Name");
jControl.addJob(A);
jControl.addJob(B);
Thread runJControl = new Thread(jControl);
runJControl.start();
while (!jControl.allFinished()) {
code = jControl.getFailedJobList().size() == 0 ? 0 : 1;
Thread.sleep(1000);
}
System.exit(1);

一开始的初始化代码是这样的:

int code =1;

让你的第一个工作是第一个具有零化简器的映射器，第二个工作是第二个具有零化简器的映射器。配置应该是 B 的输入路径和 A 的输出路径应该相同。

关于hadoop - 在 Hadoop MapReduce 中对多个输出目录使用多个映射器，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/33104028/