hive join使用总结

hive join方式总结

join类型

[INNER] JOIN 内联
LEFT JOIN 左连
RIGHT JOIN 右连
FULL OUTER JOIN 全外连
LEFT SEMI JOIN 左半连
CROSS JOIN 笛卡尔

数据准备

• cl_tmp.tmp_tm_join_testA

id	name
1	张三
2	李四
3	王五

• cl_tmp.tmp_tm_join_testB

id	age
2	26
3	28
4	29

create table cl_tmp.tmp_tm_join_testA(id int,name string);
insert INTO cl_tmp.tmp_tm_join_testA values(1,'张三'),(2,'李四'),(3,'王五');


create table cl_tmp.tmp_tm_join_testB(id int,age int);
insert INTO cl_tmp.tmp_tm_join_testB values(2,26),(3,28),(4,29);

数据关联

• INNER JOIN

# 取 a b 表交集
select * from cl_tmp.tmp_tm_join_testA a INNER JOIN  cl_tmp.tmp_tm_join_testB b ON a.id = b.id;

• LEFT JOIN

# 补充a表中b表列值
select * from cl_tmp.tmp_tm_join_testA a LEFT JOIN  cl_tmp.tmp_tm_join_testB b ON a.id = b.id;

# 获取a表中存在，b表中不存在
select * from cl_tmp.tmp_tm_join_testA a LEFT JOIN  cl_tmp.tmp_tm_join_testB b ON a.id = b.id where b.id is null;

• RIGHT JOIN

# 获取b表数据
select * from cl_tmp.tmp_tm_join_testA a RIGHT JOIN  cl_tmp.tmp_tm_join_testB b ON a.id = b.id;

# 获取b表中存在，a表中不存在的数据： b - a
select * from cl_tmp.tmp_tm_join_testA a RIGHT JOIN  cl_tmp.tmp_tm_join_testB b ON a.id = b.id where a.id is null;

• FULL [OUTER] JOIN

# a表∪b表
select * from cl_tmp.tmp_tm_join_testA a FULL OUTER JOIN  cl_tmp.tmp_tm_join_testB b ON a.id = b.id;

# (a - b) ∪ (b - a) 
select * from cl_tmp.tmp_tm_join_testA a FULL OUTER JOIN  cl_tmp.tmp_tm_join_testB b ON a.id = b.id 
where a.id is null or b.id is null ;

• LEFT SEMI JOIN

# 同INNER JOIN  
select * from cl_tmp.tmp_tm_join_testA a LEFT SEMI JOIN cl_tmp.tmp_tm_join_testB b ON a.id = b.id;

• CROSS JOIN

# 笛卡尔集
select * from cl_tmp.tmp_tm_join_testA a CROSS JOIN cl_tmp.tmp_tm_join_testB b ;

参考blog

Hive中的各种join关系和使用 https://blog.csdn.net/leying521/article/details/93197951