CODEWITHSUNDEEP
sqlamazon-redshift
Py.rookie89
Py.rookie89

Reputation: 129

AWS Redshift SQL using the results of a query to execute another query

I have a table that has descriptions. I am using a query based on author IDs to find the relevant descriptions. I use this code:

 SELECT
    count(*), h.book_desc
    FROM
    native.authbill p, native.chg h
    where p.book_chg_id = h.book_chg_id
    and 
    (p.aut_key in (
    select aut_key
    from native.authcodes p
    where p.auth_code in (74233, 23421) )
    or p.aut_key in (
    select aut_key
    from native.pubisbn_proc pat
    where isbn_code in ('373423','0256543','0257535')))
    group by h.book_desc

Then I have a another query that finds book descriptions based on a genre

SELECT
    count(*), h.book_desc
    FROM
    native.authbill p, native.chg h
    where p.book_chg_id = h.book_chg_id
    and p.genre_code in (
    SELECT distinct chg.genre_code
    FROM native.chgset chg
    where chg.genre_desc in ('Sci-fi', 'Action', 'Rom-com')
                        )

What I want to do is take all the results from the first query and narrow them down further by the second query. I get 150000 results from the first query and 250000 when i run the second query. How can I narrow down the first query results by the second query. So I want to use the results from the first query and then make sure those descriptions also are in the genre description array. This is in AWS Redshift SQL. Any help is appreciated.

Upvotes: 0

Views: 756

Answers (2)

Gordon Linoff
Gordon Linoff

Reputation: 1269463

I think you can just use an and condition in the where clause:

select count(*), h.book_desc
from native.authbill p join
     native.chg h
     on p.book_chg_id = h.book_chg_id
where (p.aut_key in (select aut_key
                     from native.authcodes p
                     where p.auth_code in (74233, 23421)
                    ) or
       p.aut_key in (select aut_key
                     from native.pubisbn_proc pat
                     where isbn_code in ('373423', '0256543', '0257535')
                    )
      ) and
      p.genre_code in (select chg.genre_code
                       from native.chgset chg
                       where chg.genre_desc in ('Sci-fi', 'Action', 'Rom-com')
                      ) 
group by h.book_desc;

Upvotes: 0

Neal Fultz
Neal Fultz

Reputation: 9687

You could use a CTE, and a join to intersect the two sets:

with q1 as (
SELECT
    count(*) as cnt, h.book_desc
    FROM
    native.authbill p, native.chg h
    where p.book_chg_id = h.book_chg_id
    and 
    (p.aut_key in (
    select aut_key
    from native.authcodes p
    where p.auth_code in (74233, 23421) )
    or p.aut_key in (
    select aut_key
    from native.pubisbn_proc pat
    where isbn_code in ('373423','0256543','0257535')))
    group by h.book_desc
),
q2 as (
SELECT
    count(*) as cnt, h.book_desc
    FROM
    native.authbill p, native.chg h
    where p.book_chg_id = h.book_chg_id
    and p.genre_code in (
    SELECT distinct chg.genre_code
    FROM native.chgset chg
    where chg.genre_desc in ('Sci-fi', 'Action', 'Rom-com')
                        )
)
select book_desc, q1.cnt, q2.cnt 
from q1 join q2 using book_desc

Upvotes: 2

Related Questions