MaxCompute：Pipeline樣本

前提條件

已通過快速入門完成測試所需環境配置。

測試準備

1. 準備好測試程式的JAR包，假設名字為mapreduce-examples.jar，本地存放路徑為MaxCompute用戶端bin目錄下data\resources。

2. 準備好Pipeline的測試表和資源。

   1. 建立測試表。

      CREATE TABLE wc_in (key STRING, value STRING);
      CREATE TABLE wc_out(key STRING, cnt 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用戶端bin目錄下data.txt匯入wc_in表中。

   tunnel upload data.txt wc_in;

   匯入wc_in表的資料檔案data的內容。

   hello,odps

測試步驟

jar -resources mapreduce-examples.jar -classpath data\resources\mapreduce-examples.jar
com.aliyun.odps.mapred.open.example.WordCountPipeline wc_in wc_out;

預期結果

+------------+------------+
| key        | cnt        |
+------------+------------+
| hello      | 1          |
| odps       | 1          |
+------------+------------+

程式碼範例

Pom依賴資訊，請參見注意事項。

package com.aliyun.odps.mapred.open.example;
import java.io.IOException;
import java.util.Iterator;
import com.aliyun.odps.Column;
import com.aliyun.odps.OdpsException;
import com.aliyun.odps.OdpsType;
import com.aliyun.odps.data.Record;
import com.aliyun.odps.data.TableInfo;
import com.aliyun.odps.mapred.Job;
import com.aliyun.odps.mapred.MapperBase;
import com.aliyun.odps.mapred.ReducerBase;
import com.aliyun.odps.pipeline.Pipeline;
public class WordCountPipelineTest {
    public static class TokenizerMapper extends MapperBase {
        Record word;
        Record one;
        @Override
            public void setup(TaskContext context) throws IOException {
            word = context.createMapOutputKeyRecord();
            one = context.createMapOutputValueRecord();
            one.setBigint(0, 1L);
        }
        @Override
            public void map(long recordNum, Record record, TaskContext context)
            throws IOException {
            for (int i = 0; i < record.getColumnCount(); i++) {
                String[] words = record.get(i).toString().split("\\s+");
                for (String w : words) {
                    word.setString(0, w);
                    context.write(word, one);
                }
            }
        }
    }
    public static class SumReducer extends ReducerBase {
        private Record value;
        @Override
            public void setup(TaskContext context) throws IOException {
            value = context.createOutputValueRecord();
        }
        @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);
            }
            value.set(0, count);
            context.write(key, value);
        }
    }
    public static class IdentityReducer extends ReducerBase {
        private Record result;
        @Override
            public void setup(TaskContext context) throws IOException {
            result = context.createOutputRecord();
        }
        @Override
            public void reduce(Record key, Iterator<Record> values, TaskContext context)
            throws IOException {
            while (values.hasNext()) {
                result.set(0, key.get(0));
                result.set(1, values.next().get(0));
                context.write(result);
            }
        }
    }
    public static void main(String[] args) throws OdpsException {
        if (args.length != 2) {
            System.err.println("Usage: WordCountPipeline <in_table> <out_table>");
            System.exit(2);
        }
        Job job = new Job();
        /**構造Pipeline的過程中，如果不指定Mapper的OutputKeySortColumns、PartitionColumns、OutputGroupingColumns，架構會預設使用其OutputKey作為此三者的預設配置。
         */
        Pipeline pipeline = Pipeline.builder()
            .addMapper(TokenizerMapper.class)
            .setOutputKeySchema(
            new Column[] { new Column("word", OdpsType.STRING) })
            .setOutputValueSchema(
            new Column[] { new Column("count", OdpsType.BIGINT) })
            .setOutputKeySortColumns(new String[] { "word" })
            .setPartitionColumns(new String[] { "word" })
            .setOutputGroupingColumns(new String[] { "word" })
            .addReducer(SumReducer.class)
            .setOutputKeySchema(
            new Column[] { new Column("word", OdpsType.STRING) })
            .setOutputValueSchema(
            new Column[] { new Column("count", OdpsType.BIGINT)})
            .addReducer(IdentityReducer.class).createPipeline();
        /**將pipeline的設定到jobconf中，如果需要設定combiner，是通過jobconf來設定。*/
        job.setPipeline(pipeline);
        /**設定輸入輸出表。*/
        job.addInput(TableInfo.builder().tableName(args[0]).build());
        job.addOutput(TableInfo.builder().tableName(args[1]).build());
        /**作業提交並等待結束。*/
        job.submit();
        job.waitForCompletion();
        System.exit(job.isSuccessful() == true ? 0 : 1);
    }
}