前提條件

已通過快速入門完成測試所需環(huán)境配置。

測試準(zhǔn)備

1. 準(zhǔn)備好測試程序的JAR包，假設(shè)名字為mapreduce-examples.jar，本地存放路徑為MaxCompute客戶端bin目錄下data\resources。

2. 準(zhǔn)備好Unique的測試表和資源。

   1. 創(chuàng)建測試表。

      CREATE TABLE ss_in(key BIGINT, value BIGINT);
      CREATE TABLE ss_out(key BIGINT, value BIGINT);

   2. 添加測試資源。

      -- 首次添加忽略-f覆蓋指令。
      add jar data\resources\mapreduce-examples.jar -f;

3. 使用Tunnel將MaxCompute客戶端bin目錄下data.txt導(dǎo)入ss_in表中。

   tunnel upload data.txt ss_in;

   導(dǎo)入ss_in表的數(shù)據(jù)文件data的內(nèi)容如下。

    1,1
    1,1
    2,2
    2,2

測試步驟

jar -resources mapreduce-examples.jar -classpath data\resources\mapreduce-examples.jar
com.aliyun.odps.mapred.open.example.Unique ss_in ss_out key;

預(yù)期結(jié)果

+------------+------------+
| key        | value      |
+------------+------------+
| 1          |   1        |
| 2          |   2        |
+------------+------------+

代碼示例

Pom依賴信息，請參見注意事項(xiàng)。

package com.aliyun.odps.mapred.open.example;
import java.io.IOException;
import java.util.Iterator;
import com.aliyun.odps.data.Record;
import com.aliyun.odps.data.TableInfo;
import com.aliyun.odps.mapred.JobClient;
import com.aliyun.odps.mapred.MapperBase;
import com.aliyun.odps.mapred.ReducerBase;
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;
/**
     * Unique Remove duplicate words
     *
     **/
public class Unique {
    public static class OutputSchemaMapper extends MapperBase {
        private Record key;
        private Record value;
        @Override
            public void setup(TaskContext context) throws IOException {
            key = context.createMapOutputKeyRecord();
            value = context.createMapOutputValueRecord();
        }
        @Override
            public void map(long recordNum, Record record, TaskContext context)
            throws IOException {
            long left = 0;
            long right = 0;
            if (record.getColumnCount() > 0) {
                left = (Long) record.get(0);
                if (record.getColumnCount() > 1) {
                    right = (Long) record.get(1);
                }
                key.set(new Object[] { (Long) left, (Long) right });
                value.set(new Object[] { (Long) left, (Long) right });
                context.write(key, value);
            }
        }
    }
    public static class OutputSchemaReducer 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 {
            result.set(0, key.get(0));
            while (values.hasNext()) {
                Record value = values.next();
                result.set(1, value.get(1));
            }
            context.write(result);
        }
    }
    public static void main(String[] args) throws Exception {
        if (args.length > 3 || args.length < 2) {
            System.err.println("Usage: unique <in> <out> [key|value|all]");
            System.exit(2);
        }
        String ops = "all";
        if (args.length == 3) {
            ops = args[2];
        }
        /** Reduce的輸入分組是由setOutputGroupingColumns的設(shè)置來決定的，這個(gè)參數(shù)如果不設(shè)置，默認(rèn)就是MapOutputKeySchema。*/
        // Key Unique
        if (ops.equals("key")) {
            JobConf job = new JobConf();
            job.setMapperClass(OutputSchemaMapper.class);
            job.setReducerClass(OutputSchemaReducer.class);
            job.setMapOutputKeySchema(SchemaUtils.fromString("key:bigint,value:bigint"));
            job.setMapOutputValueSchema(SchemaUtils.fromString("key:bigint,value:bigint"));
            job.setPartitionColumns(new String[] { "key" });
            job.setOutputKeySortColumns(new String[] { "key", "value" });
            job.setOutputGroupingColumns(new String[] { "key" });
            job.set("tablename2", args[1]);
            job.setNumReduceTasks(1);
            job.setInt("table.counter", 0);
            InputUtils.addTable(TableInfo.builder().tableName(args[0]).build(), job);
            OutputUtils.addTable(TableInfo.builder().tableName(args[1]).build(), job);
            JobClient.runJob(job);
        }
        // Key&Value Unique
        if (ops.equals("all")) {
            JobConf job = new JobConf();
            job.setMapperClass(OutputSchemaMapper.class);
            job.setReducerClass(OutputSchemaReducer.class);
            job.setMapOutputKeySchema(SchemaUtils.fromString("key:bigint,value:bigint"));
            job.setMapOutputValueSchema(SchemaUtils.fromString("key:bigint,value:bigint"));
            job.setPartitionColumns(new String[] { "key" });
            job.setOutputKeySortColumns(new String[] { "key", "value" });
            job.setOutputGroupingColumns(new String[] { "key", "value" });
            job.set("tablename2", args[1]);
            job.setNumReduceTasks(1);
            job.setInt("table.counter", 0);
            InputUtils.addTable(TableInfo.builder().tableName(args[0]).build(), job);
            OutputUtils.addTable(TableInfo.builder().tableName(args[1]).build(), job);
            JobClient.runJob(job);
        }
        // Value Unique
        if (ops.equals("value")) {
            JobConf job = new JobConf();
            job.setMapperClass(OutputSchemaMapper.class);
            job.setReducerClass(OutputSchemaReducer.class);
            job.setMapOutputKeySchema(SchemaUtils.fromString("key:bigint,value:bigint"));
            job.setMapOutputValueSchema(SchemaUtils.fromString("key:bigint,value:bigint"));
            job.setPartitionColumns(new String[] { "value" });
            job.setOutputKeySortColumns(new String[] { "value" });
            job.setOutputGroupingColumns(new String[] { "value" });
            job.set("tablename2", args[1]);
            job.setNumReduceTasks(1);
            job.setInt("table.counter", 0);
            InputUtils.addTable(TableInfo.builder().tableName(args[0]).build(), job);
            OutputUtils.addTable(TableInfo.builder().tableName(args[1]).build(), job);
            JobClient.runJob(job);
        }
    }
}