在介紹這個實例之前,請各位參考:http://bjyjtdj.iteye.com/blog/1453410。
reduce side join是一種最簡單的join方式,其主要思想如下:
在map階段,map函數同時讀取兩個文件File1和File2,爲了區分兩種來源的key/value數據對,對每條數據打一個標籤(tag),比如:tag=0表示來自文件File1,tag=2表示來自文件File2。即:map階段的主要任務是對不同文件中的數據打標籤。在reduce階段,reduce函數獲取key相同的來自File1和File2文件的value list, 然後對於同一個key,對File1和File2中的數據進行join(笛卡爾乘積)。即:reduce階段進行實際的連接操作。在這個例子中我們假設有兩個數據文件如下:
user.csv文件:
"ID","NAME","SEX"
"1","user1","0"
"2","user2","0"
"3","user3","0"
"4","user4","1"
"5","user5","0"
"6","user6","0"
"7","user7","1"
"8","user8","0"
"9","user9","0"
order.csv文件:
"USER_ID","NAME"
"1","order1"
"2","order2"
"3","order3"
"4","order4"
"7","order7"
"8","order8"
"9","order9"
目前網上的例子大多是基於0.20以前版本的API寫的,所以咱們採用新的API來寫,具體代碼如下:
public class MyJoin
{
public static class MapClass extends
Mapper<LongWritable, Text, Text, Text>
{
//最好在map方法外定義變量,以減少map計算時創建對象的個數
private Text key = new Text();
private Text value = new Text();
private String[] keyValue = null;
@Override
protected void map(LongWritable key, Text value, Context context)
throws IOException, InterruptedException
{
//採用的數據輸入格式是TextInputFormat,
//文件被分爲一系列以換行或者製表符結束的行,
//key是每一行的位置(偏移量,LongWritable類型),
//value是每一行的內容,Text類型,所有我們要把key從value中解析出來
keyValue = value.toString().split(",", 2);
this.key.set(keyValue[0]);
this.value.set(keyValue[1]);
context.write(this.key, this.value);
}
}
public static class Reduce extends Reducer<Text, Text, Text, Text>
{
//最好在reduce方法外定義變量,以減少reduce計算時創建對象的個數
private Text value = new Text();
@Override
protected void reduce(Text key, Iterable<Text> values, Context context)
throws IOException, InterruptedException
{
StringBuilder valueStr = new StringBuilder();
//values中的每一個值是不同數據文件中的具有相同key的值
//即是map中輸出的多個文件相同key的value值集合
for(Text val : values)
{
valueStr.append(val);
valueStr.append(",");
}
this.value.set(valueStr.deleteCharAt(valueStr.length()-1).toString());
context.write(key, this.value);
}
}
public static void main(String[] args) throws Exception
{
Configuration conf = new Configuration();
Job job = new Job(conf, "MyJoin");
job.setJarByClass(MyJoin.class);
job.setMapperClass(MapClass.class);
job.setReducerClass(Reduce.class);
//job.setCombinerClass(Reduce.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(Text.class);
//分別採用TextInputFormat和TextOutputFormat作爲數據的輸入和輸出格式
//如果不設置,這也是Hadoop默認的操作方式
job.setInputFormatClass(TextInputFormat.class);
job.setOutputFormatClass(TextOutputFormat.class);
FileInputFormat.addInputPath(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, new Path(args[1]));
System.exit(job.waitForCompletion(true) ? 0 : 1);
}
}