How to count distinct values over two categories?

Question

I have a SQL query that shows me the clients and product for each purchase

client    | product
-----------------------
Lucy Lu   | Banana
Lucy Lu   | Banana
Lucy Lu   | Pineapple    
Mad Damon | Banana
Mad Damon | Apple
Mad Damon | Apple
Peter Fox | Banana
Peter Fox | Banana
Peter Fox | Banana
Peter Fox | Apple    
Peter Fox | Apple 

I want to distinct these query but the count of each product for each customer, so I can see for each customer how much of each product he bought:

client    | product   | count
----------------------------
Lucy Lu   | Banana    | 2
Lucy Lu   | Pineapple | 1
Mad Damon | Banana    | 1
Mad Damon | Apple     | 2
Peter Fox | Banana    | 3
Peter Fox | Apple     | 2

I tried it with count(DISTINCT product) , count(DISTINCT client) , count(*) and GROUP BY (client) or GROUP BY (product) , but didn't get a useful solution. When I try it with SELECT DISTINCT [rest of the query], I get what I want but without the count column.

Answer 1

You don't need to use DISTINCT. A simple GROUP BY with COUNT will do the trick.

Oto Shavadze suggested this query:

select client, product, count(*) 
from t 
group by client, product;

Here you make groups based on the distinct values of (client, product) pairs. The COUNT will give you the count of rows in each of these group.

Answer 2

Try this query and you will get the output. Here test_data is the table name. Please replace it with your table_name.

SELECT count(product) as count,client, product from test_data group by product,client