Question

SQL group by problem 

I have a SQL group by problem. My table has the following form.

Cust_id.  Price_id     Price. 
----------------------------  
1.          556.        5000. 
-----------------------------  
2.          654.         600. 
2.          432.         487. 
2.          546.         500. 
---------------------------  
3.          455.         200. 
3.          877.         143. 
3.          123.         879. 

Now when I run this query:

Select  cust_id,  max(price) as max, min(price) as min. 
From table. 
Group by cust_id. 

I get.

Cust_id.      Max.        Min. 
1.           5000.       5000. 
2.            600.        487. 
3.            879.        143. 

But what I really want is not the max and min price but the price_id associated with the price.
So the results would be.

Cust_id.       Max.        Min.   
1.             556.        556.   
2.             654.        432.   
3.             123.        877.    

I am at a loss for how to do this. I think that the above query would be a sub query of some sort but that is as far as I got.  

Answer 1

Use:

   SELECT x.cust_id,
          y.price_id AS max,
          z.price_id AS min
     FROM (SELECT t.cust_id,  
                  MAX(t.price) as max, 
                  MIN(t.price) as min
             FROM TABLE t
         GROUP BY t.cust_id) x
LEFT JOIN TABLE y ON y.cust_id = x.cust_id
                 AND y.price = x.max
LEFT JOIN TABLE z ON z.cust_id = x.cust_id
                 AND z.price = x.min

The problem is that if a cust_id has two records with the same high (or low) price, you’ll see duplicates and will need to provide logic to deal with ties.