Java执行hadoop的基本操作实例代码

时间:2022-08-25 21:09:45

Java执行hadoop的基本操作实例代码

向HDFS上传本地文件

 
?
1
 
2
3
4
5
6
7
8
9
public static void uploadInputFile(String localFile) throws IOException{
    Configuration conf = new Configuration();
    String hdfsPath = "hdfs://localhost:9000/";
    String hdfsInput = "hdfs://localhost:9000/user/hadoop/input";
    FileSystem fs = FileSystem.get(URI.create(hdfsPath), conf);
    fs.copyFromLocalFile(new Path(localFile), new Path(hdfsInput));
    fs.close();
    System.out.println("已经上传文件到input文件夹啦");
  }

将output文件下载到本地

 
?
1
 
2
3
4
5
6
7
8
9
10
public static void getOutput(String outputfile) throws IOException{
    String remoteFile = "hdfs://localhost:9000/user/hadoop/output/part-r-00000";
    Path path = new Path(remoteFile);
    Configuration conf = new Configuration();
    String hdfsPath = "hdfs://localhost:9000/";
    FileSystem fs = FileSystem.get(URI.create(hdfsPath),conf);
    fs.copyToLocalFile(path, new Path(outputfile));
    System.out.println("已经将输出文件保留到本地文件");
    fs.close();
  }

删除hdfs中的文件

 
?
1
 
2
3
4
5
6
7
8
9
10
public static void deleteOutput() throws IOException{
   Configuration conf = new Configuration();
   String hdfsOutput = "hdfs://localhost:9000/user/hadoop/output";
   String hdfsPath = "hdfs://localhost:9000/";
   Path path = new Path(hdfsOutput);
   FileSystem fs = FileSystem.get(URI.create(hdfsPath), conf);
   fs.deleteOnExit(path);
   fs.close();
   System.out.println("output文件已经删除");
 }

执行mapReduce程序

创建Mapper类和Reducer类

 
?
1
 
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
public static class TokenizerMapper extends Mapper<Object, Text, Text, IntWritable>{
 
    private final static IntWritable one = new IntWritable(1);
    private Text word = new Text();
 
    public void map(Object key, Text value, Context context) throws IOException, InterruptedException{
      String line = value.toString();
      line = line.replace("\\", "");
      String regex = "性别:</span><span class=\"pt_detail\">(.*?)</span>";
      Pattern pattern = Pattern.compile(regex);
      Matcher matcher = pattern.matcher(line);
      while(matcher.find()){
        String term = matcher.group(1);
        word.set(term);
        context.write(word, one);
      }
    }
  }
 
  public static class IntSumReducer extends Reducer<Text, IntWritable, Text, IntWritable>{
 
    private IntWritable result = new IntWritable();
 
    public void reduce(Text key, Iterable<IntWritable> values, Context context) throws IOException, InterruptedException{
      int sum = 0;
      for(IntWritable val :values){
        sum+= val.get();
      }
      result.set(sum);
      context.write(key, result);
    }
  }

执行mapReduce程序

 
?
1
 
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
public static void runMapReduce(String[] args) throws Exception {
    Configuration conf = new Configuration();
    String[] otherArgs = new GenericOptionsParser(conf, args).getRemainingArgs();
    if(otherArgs.length != 2){
      System.err.println("Usage: wordcount<in> <out>");
      System.exit(2);
    }
    Job job = new Job(conf, "word count");
    job.setJarByClass(WordCount.class);
    job.setMapperClass(TokenizerMapper.class);
    job.setCombinerClass(IntSumReducer.class);
    job.setReducerClass(IntSumReducer.class);
    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(IntWritable.class);
    FileInputFormat.addInputPath(job, new Path(otherArgs[0]));
    FileOutputFormat.setOutputPath(job, new Path(otherArgs[1]));
    System.out.println("mapReduce 执行完毕!");
    System.exit(job.waitForCompletion(true)?0:1);
 
  }

感谢阅读,希望能帮助到大家,谢谢大家对本站的支持!

原文链接:http://blog.csdn.net/qq_30843221/article/details/54429792