Hadoop 多表关联

  1 import java.io.IOException;
  2 import java.util.Iterator;
  3 
  4 import org.apache.hadoop.conf.Configuration;
  5 import org.apache.hadoop.fs.Path;
  6 import org.apache.hadoop.io.Text;
  7 import org.apache.hadoop.mapreduce.Job;
  8 import org.apache.hadoop.mapreduce.Mapper;
  9 import org.apache.hadoop.mapreduce.Reducer;
 10 import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
 11 import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
 12 import org.apache.hadoop.util.GenericOptionsParser;
 13 
 14 
 15 public class MTjoin {
 16     
 17     public static int time = 0;
 18     
 19     public static class Map extends Mapper<Object, Text, Text, Text>{
 20         // 在Map中先区分输入行属于左表还是右表，然后对两列值进行分割，
 21         // 连接列保存在key值，剩余列和左右表标志保存在value中，最后输出
 22         @Override
 23         protected void map(Object key, Text value,Mapper<Object, Text, Text, Text>.Context context)
 24                 throws IOException, InterruptedException {
 25             // super.map(key, value, context);
 26             String line = value.toString();
 27             int i=0;
 28             // 输入文件首行，不处理
 29             if(line.contains("factoryname")==true || line.contains("addressID")==true){
 30                 return ;
 31             }
 32             // 找出数据中的分割点
 33             while(line.charAt(i)>=‘9‘ || line.charAt(i)<=‘0‘){
 34                 i++;
 35             }
 36             if (line.charAt(0)>=‘9‘||line.charAt(0)<=‘0‘) {
 37                 // 左表
 38                 int j = i-1;
 39                 while(line.charAt(j)!=‘ ‘) j--;
 40                 String [] values = {line.substring(0,j),line.substring(i)};
 41                 context.write(new Text(values[1]), new Text("1+"+values[0]));
 42             }else {
 43                 // 右表
 44                 int j = i+1;
 45                 while(line.charAt(j)!=‘ ‘) j++;
 46                 String[] values = {line.substring(0,i+1),line.substring(j)};
 47                 context.write(new Text(values[0]), new Text("2"+values[1]));
 48             }
 49         }
 50     }
 51 
 52     public static class Reduce extends Reducer<Text, Text, Text, Text>{
 53         // Reduce解析Map输出，将value中数据按照左右表分别保存，然后求 // 笛卡尔积，输出
 54         @Override
 55         protected void reduce(Text key, Iterable<Text> values,Reducer<Text, Text, Text, Text>.Context context)
 56                 throws IOException, InterruptedException {
 57             // super.reduce(arg0, arg1, arg2);
 58             if (time==0) {
 59                 //  输出文件第一行
 60                 context.write(new Text("factoryname"), new Text("addressname"));
 61                 time++;
 62             }
 63             int factorynum = 0;
 64             String[] factory = new String[10];
 65             int addressnum = 0;
 66             String[] address = new String[10];
 67             Iterator ite = values.iterator();
 68             while (ite.hasNext()) {
 69                 String record = ite.next().toString();
 70                 int len = record.length();
 71                 int i = 2;
 72                 char type = record.charAt(0);
 73                 String factoryname = new String();
 74                 String addressname = new String();
 75                 if (type==‘1‘) {
 76                     // 左表
 77                     factory[factorynum] = record.substring(2);
 78                     factorynum++;
 79                 }else {
 80                     // 右表
 81                     address[addressnum] = record.substring(2);
 82                     addressnum++;
 83                 }
 84             }
 85             if (factorynum != 0 && addressnum !=0) {
 86                 // 求笛卡尔积
 87                 for(int m=0;m<factorynum;m++){
 88                     for(int n=0;n<addressnum;n++){
 89                         context.write(new Text(factory[m]), new Text(address[n]));
 90                     }
 91                 }
 92             }
 93         }
 94     }
 95     
 96     public static void main(String[] args) throws IOException, ClassNotFoundException, InterruptedException {
 97         
 98         Configuration conf = new Configuration();
 99         String[] otherArgs = new GenericOptionsParser(conf,args).getRemainingArgs();
100         if(otherArgs.length!=2){
101             System.out.println("Usage:wordcount <in> <out>");
102             System.exit(2);
103         }
104         Job job = new Job(conf,"multiple table join");
105         job.setJarByClass(MTjoin.class);
106         job.setMapperClass(Map.class);
107         job.setReducerClass(Reduce.class);
108         job.setOutputKeyClass(Text.class);
109         job.setOutputValueClass(Text.class);
110         FileInputFormat.addInputPath(job,new Path(otherArgs[0]));
111         FileOutputFormat.setOutputPath(job, new Path(otherArgs[1]));
112         System.exit(job.waitForCompletion(true)?0:1);
113     }
114 
115 }