Sql - Multiple Layers Of Correlated Subqueries

August 21, 2024 Post a Comment

I have table A, B and C I want to return all entries in table A that do not exist in table B and of that list do not exist in table C. select * from table_A as a where not exists (

Solution 1:

Try this:

select * from (
  select a.*, b.id as b_id, c.id as c_id 
  from table_A as a
  left outer join table_B as b on a.id = b.id
  left outer join table_C as c on c.id = a.id
) T
where b_id isnulland c_id isnull

Another implementation is this:

select a1.*
from table_A as a1
inner join (
  select a.idfrom table_A
  except  
  select b.idfrom table_B
  except
  select c.idfrom table_c
) as a2 on a1.id = a2.id

Note the restrictions on the form of the sub-query as described here. The second implementation, by most succinctly and clearly describing the desired operation to SQL Server, is likely to be the most efficient.

Solution 2:

You have two WHERE clauses in (the external part of) your second query. That is not valid SQL. If you remove it, it should work as expected:

select * from table_A as a
wherenotexists (select1from table_B as b
                  where a.id = b.id)
AND
      notexists (select1from table_C as c            -- WHERE removed
                  where a.id = c.id) ;

Tested in SQL-Fiddle (thnx @Alexander)

Solution 3:

how about using LEFT JOIN

SELECT  a.*FROM    TableA a
        LEFTJOIN TableB b
            ON a.ID = b.ID
        LEFTJOIN TableC c
            ON a.ID = c.ID
WHERE   b.ID ISNULLAND
        c.ID ISNULL

SQLFiddle Demo

Solution 4:

One more option with NOT EXISTS operator

SELECT*FROM dbo.test71 a 
WHERENOTEXISTS(
                 SELECT1FROM (SELECT b.ID
                       FROM dbo.test72 b
                       UNIONALLSELECT c.ID
                       FROM dbo.test73 c) x
                 WHERE a.ID = x.ID
                 )

Demo on SQLFiddle

Option from @ypercube.Thank for the present;)

SELECT*FROM dbo.test71 a 
WHERENOTEXISTS(
                 SELECT1FROM dbo.test72 b
                 WHERE a.ID = b.ID  
                 UNIONALLSELECT1FROM dbo.test73 c
                 WHERE a.ID = c.ID
                 );

Demo on SQLFiddle

Solution 5:

I do not like "not exists" but if for some reason it seems to be more logical to you; then you can use a alias for your first query. Subsequently, you can re apply another "not exists" clause. Something like:

SELECT * FROM 
  ( select * from tableA as a
  wherenot exists (select1from tableB as b
  where a.id = b.id) )
AS A_NOT_IN_B
WHERENOT EXISTS (
  SELECT1FROM tableC as c
  WHERE c.id = A_NOT_IN_B.id
)

Lab Sql