测试准备

  1. 准备好测试程序的 Jar 包,假设名字为 mapreduce-examples.jar,本地存放路径为 data\resources
  2. 准备好 Grep 测试表和资源。
    • 创建测试表。
      create table mr_src(key string, value string);
      create table mr_grep_tmp (key string, cnt bigint);
      create table mr_grep_out (key bigint, value string);
    • 添加测试资源。
      add jar data\resources\mapreduce-examples.jar -f;
  3. 使用 tunnel 导入数据。
    tunnel upload data mr_src;
    导入 mr_src 表的数据文件 data 的内容,如下所示:
     hello,odps
     hello,world

测试步骤

在 odpscmd 中执行 Grep,如下所示:
jar -resources mapreduce-examples.jar -classpath data\resources\mapreduce-examples.jar
com.aliyun.odps.mapred.open.example.Grep mr_src mr_grep_tmp mr_grep_out hello;

预期结果

作业成功结束后,输出表 mr_grep_out 中的内容,如下所示:
+------------+------------+
| key        | value      |
+------------+------------+
| 2          | hello      |
+------------+------------+

代码示例

    package com.aliyun.odps.mapred.open.example;
    import java.io.IOException;
    import java.util.Iterator;
    import java.util.regex.Matcher;
    import java.util.regex.Pattern;
    import com.aliyun.odps.data.Record;
    import com.aliyun.odps.data.TableInfo;
    import com.aliyun.odps.mapred.JobClient;
    import com.aliyun.odps.mapred.Mapper;
    import com.aliyun.odps.mapred.MapperBase;
    import com.aliyun.odps.mapred.ReducerBase;
    import com.aliyun.odps.mapred.RunningJob;
    import com.aliyun.odps.mapred.TaskContext;
    import com.aliyun.odps.mapred.conf.JobConf;
    import com.aliyun.odps.mapred.utils.InputUtils;
    import com.aliyun.odps.mapred.utils.OutputUtils;
    import com.aliyun.odps.mapred.utils.SchemaUtils;
    /**
     *
     * Extracts matching regexs from input files and counts them.
     *
     **/
    public class Grep {
      /**
       * RegexMapper
       **/
      public class RegexMapper extends MapperBase {
        private Pattern pattern;
        private int group;
        private Record word;
        private Record one;
        @Override
        public void setup(TaskContext context) throws IOException {
          JobConf job = (JobConf) context.getJobConf();
          pattern = Pattern.compile(job.get("mapred.mapper.regex"));
          group = job.getInt("mapred.mapper.regex.group", 0);
          word = context.createMapOutputKeyRecord();
          one = context.createMapOutputValueRecord();
          one.set(new Object[] { 1L });
        }
        @Override
        public void map(long recordNum, Record record, TaskContext context) throws IOException {
          for (int i = 0; i < record.getColumnCount(); ++i) {
            String text = record.get(i).toString();
            Matcher matcher = pattern.matcher(text);
            while (matcher.find()) {
              word.set(new Object[] { matcher.group(group) });
              context.write(word, one);
            }
          }
        }
      }
      /**
       * LongSumReducer
       **/
      public class LongSumReducer extends ReducerBase {
        private Record result = null;
        @Override
        public void setup(TaskContext context) throws IOException {
          result = context.createOutputRecord();
        }
        @Override
        public void reduce(Record key, Iterator<Record> values, TaskContext context) throws IOException {
          long count = 0;
          while (values.hasNext()) {
            Record val = values.next();
            count += (Long) val.get(0);
          }
          result.set(0, key.get(0));
          result.set(1, count);
          context.write(result);
        }
      }
      /**
       * A {@link Mapper} that swaps keys and values.
       **/
      public class InverseMapper extends MapperBase {
        private Record word;
        private Record count;
        @Override
        public void setup(TaskContext context) throws IOException {
          word = context.createMapOutputValueRecord();
          count = context.createMapOutputKeyRecord();
        }
        /**
         * The inverse function. Input keys and values are swapped.
         **/
        @Override
        public void map(long recordNum, Record record, TaskContext context) throws IOException {
          word.set(new Object[] { record.get(0).toString() });
          count.set(new Object[] { (Long) record.get(1) });
          context.write(count, word);
        }
      }
      /**
       * IdentityReducer
       **/
      public class IdentityReducer extends ReducerBase {
        private Record result = null;
        @Override
        public void setup(TaskContext context) throws IOException {
          result = context.createOutputRecord();
        }
        /** Writes all keys and values directly to output. **/
        @Override
        public void reduce(Record key, Iterator<Record> values, TaskContext context) throws IOException {
          result.set(0, key.get(0));
          while (values.hasNext()) {
            Record val = values.next();
            result.set(1, val.get(0));
            context.write(result);
          }
        }
      }
      public static void main(String[] args) throws Exception {
        if (args.length < 4) {
          System.err.println("Grep <inDir> <tmpDir> <outDir> <regex> [<group>]");
          System.exit(2);
        }
        JobConf grepJob = new JobConf();
        grepJob.setMapperClass(RegexMapper.class);
        grepJob.setReducerClass(LongSumReducer.class);
        grepJob.setMapOutputKeySchema(SchemaUtils.fromString("word:string"));
        grepJob.setMapOutputValueSchema(SchemaUtils.fromString("count:bigint"));
        InputUtils.addTable(TableInfo.builder().tableName(args[0]).build(), grepJob);
        OutputUtils.addTable(TableInfo.builder().tableName(args[1]).build(), grepJob);
        // 设置grepJob的grep的正则表达式
        grepJob.set("mapred.mapper.regex", args[3]);
        if (args.length == 5) {
          grepJob.set("mapred.mapper.regex.group", args[4]);
        }
        @SuppressWarnings("unused")
        RunningJob rjGrep = JobClient.runJob(grepJob);
        // grepJob的输出作为sortJob的输入
        JobConf sortJob = new JobConf();
        sortJob.setMapperClass(InverseMapper.class);
        sortJob.setReducerClass(IdentityReducer.class);
        sortJob.setMapOutputKeySchema(SchemaUtils.fromString("count:bigint"));
        sortJob.setMapOutputValueSchema(SchemaUtils.fromString("word:string"));
        InputUtils.addTable(TableInfo.builder().tableName(args[1]).build(), sortJob);
        OutputUtils.addTable(TableInfo.builder().tableName(args[2]).build(), sortJob);
        sortJob.setNumReduceTasks(1); // write a single file
        sortJob.setOutputKeySortColumns(new String[] { "count" }); 
        @SuppressWarnings("unused")
        RunningJob rjSort = JobClient.runJob(sortJob);
      }
    }