该文档涉及的组件

SQL操作:Join (JoinBatchOp)

Java 类名:com.alibaba.alink.operator.batch.sql.JoinBatchOp

Python 类名:JoinBatchOp

功能介绍

对批式数据进行sql的JOIN操作。

参数说明

名称 中文名称 描述 类型 是否必须? 取值范围 默认值
joinPredicate where语句 where语句 String
selectClause select语句 select语句 String
type join类型 join类型: “join”, “leftOuterJoin”, “rightOuterJoin” 或 “fullOuterJoin” String “JOIN”, “LEFTOUTERJOIN”, “RIGHTOUTERJOIN”, “FULLOUTERJOIN” “JOIN”

代码示例

Python 代码

from pyalink.alink import *

import pandas as pd

useLocalEnv(1)

df1 = pd.DataFrame([
    ['Ohio', 2000, 1.5],
    ['Ohio', 2001, 1.7],
    ['Ohio', 2002, 3.6],
    ['Nevada', 2001, 2.4],
    ['Nevada', 2002, 2.9],
    ['Nevada', 2003, 3.2]
])
df2 = pd.DataFrame([
    ['Nevada', 2001, 2.4],
    ['Nevada', 2003, 3.2]
])

batch_data1 = BatchOperator.fromDataframe(df1, schemaStr='f1 string, f2 bigint, f3 double')
batch_data2 = BatchOperator.fromDataframe(df2, schemaStr='f1 string, f2 bigint, f3 double')

op = JoinBatchOp().setJoinPredicate("a.f1=b.f1").setSelectClause("a.f1, a.f2, a.f3")
op.linkFrom(batch_data1, batch_data2).print()

Java 代码

import com.alibaba.alink.operator.batch.BatchOperator;
import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
import com.alibaba.alink.operator.batch.sql.JoinBatchOp;
import org.junit.Test;

public class JoinBatchOpTest {
	@Test
    public void testJoinBatchOp() throws Exception {
    	List <Row> df1 = Arrays.asList(
        	Row.of("Ohio", 2000, 1.5),
    		Row.of("Ohio", 2001, 1.7),
    		Row.of("Ohio", 2002, 3.6),
    		Row.of("Nevada", 2001, 2.4),
    		Row.of("Nevada", 2002, 2.9),
    		Row.of("Nevada", 2003, 3.2)
    	);
    	List <Row> df2 = Arrays.asList(
    		Row.of("Nevada", 2001, 2.4),
    		Row.of("Nevada", 2003, 3.2)
    	);
    	BatchOperator <?> data1 = new MemSourceBatchOp(df1, "f1 string, f2 int, f3 double");
    	BatchOperator <?> data2 = new MemSourceBatchOp(df2, "f1 string, f2 int, f3 double");
    	BatchOperator <?> joinOp = new JoinBatchOp().setJoinPredicate("a.f1=b.f1").setSelectClause(
    		"a.f1, a.f2, a.f3");
    	joinOp.linkFrom(data1, data2).print();
	}
}

运行结果

f1 f2 f3
Nevada 2001 2.4000
Nevada 2002 2.9000
Nevada 2003 3.2000
Nevada 2001 2.4000
Nevada 2002 2.9000
Nevada 2003 3.2000