Selecting distinct values from two tables
开发者_开发问答I have two rather large databases (+1 million rows each). Both tables have the same structure.
How can I check if each value in a column is unique across both tables?
Is there a
SELECT COUNT(DISTINCT
col) FROM tbl
type of query that will consider BOTH tables?
Thanks!
You can UNION
two full sets in a subquery and then select DISTINCT col
from that.
Something like:
SELECT DISTINCT col FROM (SELECT * FROM tbl1 UNION ALL SELECT * FROM tbl2)
You can use
UNION ALL
statement. It doesn't remove duplicate rows so you can see if there are any duplicates.
Here is my initial thought in pseudocode.
select tableOne.distinctcol
from
(select distinct col as distinctcol from tb1) as tableOne
(select distinct col as distinctcol from tb2) as tableTwo
where tableOne.distinctcol = tableTwo.distinctcol
Basic get a distinct list of values from each table, join them on that column.
精彩评论