hive数仓、数据中台数据核对的方法

采用hash函数

with a as(
select 1 as c1
			,null as c2
			,3 as id
	union all
	select 1 as c1
			,0 as c2
			,3 as id
			)
			,
with b as (
select 1 as c1
			,null as c2
	union all
	select 1 as c1
			,0 as c2
			)
select a.c1
		,b.c1
		a.c2
		b.c2
from a
left join b
 on a.id =b.id
 where hash(a.c1+1   --加1的目的是因为hash(0)=hash(null)，这两个意义不一样
 					,b.c2+1
 					)
 					<>
 			hash(b.c1+1
 					,b.c2+1)