Postresql, max returns all rows with same ids

Question

System

PostgreSQL 11.5 (Ubuntu 11.5-1.pgdg19.04+1) on x86_64-pc-linux-gnu, compiled by gcc (Ubuntu 8.3.0-6ubuntu1) 8.3.0, 64-bit

Issue

Need return one max value from table for every product_id (1 product_id has 10 store_ids) based on available_count.

Sql

select
    product_id, store_id, max(available_count)
from
    j_product_store_availability
group by
    product_id,
    store_id,
    available_count
order by
    product_id,
    available_count desc,
    store_id
--limit 1

Result

product_id store_id available_count

130475  4   7
130475  1   4
130475  8   4
130475  2   3
130475  5   3
130475  7   3
130475  10  3
130475  3   2
130475  6   2
130475  9   0
130479  1   0
130479  2   0
130479  3   0
...

It returns rows for every store_id. Need to return just one max value based on available_count.

Expected result

130475  4   7
130479  1   0
(next product_id)
...

Thanks.

Solution (Laurenz solution)

SELECT DISTINCT ON (product_id)
product_id,
store_id,
max(available_count)
from
j_product_store_availability
--where store_id = '7'
group by
product_id,
store_id,
available_count
ORDER by
product_id,
available_count desc;

Answer 1

If using a non-standard PostgreSQL construct is no problem, the easiest way would be

SELECT DISTINCT ON (product_id)
       product_id, store_id, available_count
FROM j_product_store_availability
ORDER BY product_id, available_count DESC, store_id;