消除零值并将值放在一行中的两个不同列中

Question

DB-Fiddle

CREATE TABLE check_inbound 
(
    id SERIAL PRIMARY KEY,
    data_source VARCHAR,
    product VARCHAR,
    inbound_type VARCHAR,
    quantity DECIMAL
);

INSERT INTO check_inbound (data_source, product, inbound_type, quantity)
VALUES 
('data_source_01', 'product_a', 'type_01', '500'),
('data_source_01', 'product_a', 'type_02', '200'),
('data_source_01', 'product_b', 'type_01', '130'),
('data_source_01', 'product_b', 'type_02', '320'),
('data_source_01', 'product_c', 'type_01', '700'),
('data_source_01', 'product_c', 'type_02', '850'),

('data_source_02', 'product_a', 'type_01', '980'),
('data_source_02', 'product_a', 'type_02', '315'),
('data_source_02', 'product_b', 'type_01', '760'),
('data_source_02', 'product_b', 'type_02', '512'),
('data_source_02', 'product_c', 'type_01', '125'),
('data_source_02', 'product_c', 'type_02', '720');

Expected result:

product	inbound_type	quantity_data_source_01	quantity_data_source_02
product_a	type_01	500	980
product_a	type_02	200	315
product_b	type_01	130	760
product_b	type_02	320	512
product_c	type_01	700	125
product_c	type_02	850	720

I want to compare the quantity per product and inbound_type in two different data_sources .

I was able to come close to the result with this query:

SELECT
    i.data_source AS date_source,
    i.product AS product,
    i.inbound_type AS inbound_type,
    (CASE WHEN i.data_source = 'data_source_01' THEN SUM(i.quantity) ELSE 0 END) AS quantity_data_source_01,
    (CASE WHEN i.data_source = 'data_source_02' THEN SUM(i.quantity) ELSE 0 END) AS quantity_data_source_02
FROM 
    check_inbound i
GROUP BY 
    1, 2, 3
ORDER BY 2,1

However, I could not figure out how I can get rid of the rows with the zero values and put the quantities on the same line as in the expected result.

Do you have any idea what I need to change to make it work?

Answer 1

You are doing a pivot query here, and should be aggregating only by product and type. I suggest using MAX with FILTER here:

SELECT
    i.product AS product,
    i.inbound_type AS inbound_type,
    MAX(i.quantity) FILTER (WHERE i.data_source = 'data_source_01') AS quantity_data_source_01,
    MAX(i.quantity) FILTER (WHERE i.data_source = 'data_source_02') AS quantity_data_source_02
FROM check_inbound i
GROUP BY 1, 2
ORDER BY 2, 1;

If you wanted to use CASE expressions, then use this version eg

MAX(CASE WHEN i.data_source = 'data_source_01'
         THEN i.quantity END) AS quantity_data_source_01