Getting data from SQL query where another column is unique

Question

Can somebody help me with this query I am trying to write.

SELECT sku FROM product p where sku
not in(SELECT sku FROM price_adjustment p) and sku 
in (SELECT sku FROM searchable_product s where is_visible = 'T') and sku 
in (SELECT sku FROM channel_pricing c 
where offer_price > 10.00 and offer_price % 2 = 0) limit 5;

This query works fine but I also want to only pick skus where the search_prod_id is distinct. I thought this would work but it doesn't:

SELECT sku, search_prod_id FROM product p where sku 
not in(SELECT sku FROM price_adjustment p) and sku 
in (SELECT sku FROM searchable_product s 
where is_visible = 'T') and sku in 
(SELECT sku FROM channel_pricing c 
where offer_price > 10.00 and offer_price % 2 = 0) and sku 
in (SELECT distinct search_prod_id from product p ) limit 5;

When I run that I can see the search_prod_id is the same for each sku.

Can somebody tell me what I'm doing wrong?

Answer 1

You can force the search_prod_id to be different by grouping them

SELECT MAX(sku) sku, search_prod_id 
FROM   product p 
WHERE  sku NOT IN(SELECT sku 
                  FROM   price_adjustment) 
  AND  sku IN (SELECT sku 
               FROM   searchable_product
               WHERE  is_visible = 'T') 
  AND  sku IN (SELECT sku 
               FROM   channel_pricing c 
               WHERE  offer_price > 10.00 
                 AND  offer_price % 2 = 0)
GROUP BY search_prod_id
LIMIT 5;

also in your query you're checking sku against search_prod_id, every match it's just luck (bad luck IMO)

sku in (SELECT distinct search_prod_id from product p ) 

Answer 2

Try this Query it may work,

'SELECT sku FROM product p where search_prod_id in(
SELECT DISTINCT search_prod_id ,sku FROM product p where sku not in(
SELECT sku FROM price_adjustment p) and sku in (
SELECT sku FROM searchable_product s 
where is_visible = 'T') and sku in 
(SELECT sku FROM channel_pricing c 
where offer_price > 10.00 and offer_price % 2 = 0) limit 5;'

Answer 3

Not entirely sure I have the syntax correct, but something similar. I used a GROUPING in the last AND clause to only give search_prod_id s where there is only one!

SELECT sku, search_prod_id 
FROM product p 
WHERE sku not in(SELECT sku FROM price_adjustment  p) 
and sku in (SELECT sku FROM searchable_product s where is_visible = 'T' ) 
and sku in (SELECT sku FROM channel_pricing c where offer_price > 10.00 and offer_price % 2 = 0) 
and sku in (SELECT search_prod_id from product p GROUP BY search_prod_id HAVING count(*) = 1 ) limit 5;

Answer 4

The problem with your query is that when doing

sku in (SELECT distinct search_prod_id from product p ) 

you are adding to the query sku values which are equal to search_prod_id. I think that the query you are looking for is this one:

SELECT distinct sku, search_prod_id  FROM product p where sku not in(SELECT sku FROM price_adjustment p) and sku in (SELECT sku FROM searchable_product s where is_visible = 'T') and sku in (SELECT sku FROM channel_pricing c where offer_price > 10.00 and offer_price % 2 = 0) limit 5;

Answer 5

In the last statement (where .. and sku in (SELECT distinct search_prod_id ... you select only the sku's which are equal to the search_prod_id. I think you will need something like where ... and search_prod_id in (SELECT distinct search_prod_id from product ...

I think this should work:

SELECT sku, search_prod_id FROM product p where sku 
not in(SELECT sku FROM price_adjustment p) and sku 
in (SELECT sku FROM searchable_product s 
where is_visible = 'T') and sku in 
(SELECT sku FROM channel_pricing c 
where offer_price > 10.00 and offer_price % 2 = 0) and search_prod_id 
in (SELECT distinct search_prod_id from product p ) limit 5;