Can I optimize this UNION query? [duplicate] - mysql

Question

This question already has answers here:
How do I create an SQL query that groups by value ranges
(7 answers)
Closed 1 year ago.
Hello I'm taking some data from my db and I want to get this output:
How many times I have an amount less then 100€ (inbound, outbound and inner)
How many times I have an amount between 100 and 500€ (inbound, outbound and inner)
How many times I have an amount between 500 and 2000€ (inbound, outbound and inner)
...
How many times I have an amount greater then 200000€ (inbound, outbound and inner)
I can obtain this with this query:
SELECT
'< 100 €' as amount,
SUM(CASE WHEN direction = 'inbound' AND to_address is null THEN 1 ELSE 0 END) AS inbound,
SUM(CASE WHEN direction = 'outbound' AND to_address is null THEN 1 ELSE 0 END) AS outbound,
SUM(CASE WHEN to_address is not null THEN 1 ELSE 0 END) AS inside
FROM user_wallet_movement
WHERE
(amount*base_rate) < 100
AND status = 'execute'
AND mov_date > '2020-01-07'
AND mov_date < '2021-06-30'
UNION
SELECT
'100 ... 500 €' as amount,
SUM(CASE WHEN direction = 'inbound' AND to_address is null THEN 1 ELSE 0 END) AS inbound,
SUM(CASE WHEN direction = 'outbound' AND to_address is null THEN 1 ELSE 0 END) AS outbound,
SUM(CASE WHEN to_address is not null THEN 1 ELSE 0 END) AS inside
FROM user_wallet_movement
WHERE
(amount*base_rate) > 100 AND (amount*base_rate) < 500
AND status = 'execute'
AND mov_date > '2020-01-07'
AND mov_date < '2021-06-30'
UNION
SELECT
'500 ... 2000 €' as amount,
SUM(CASE WHEN direction = 'inbound' AND to_address is null THEN 1 ELSE 0 END) AS inbound,
SUM(CASE WHEN direction = 'outbound' AND to_address is null THEN 1 ELSE 0 END) AS outbound,
SUM(CASE WHEN to_address is not null THEN 1 ELSE 0 END) AS inside
FROM user_wallet_movement
WHERE
(amount*base_rate) > 500 AND (amount*base_rate) < 2000
AND status = 'execute'
AND mov_date > '2020-01-07'
AND mov_date < '2021-06-30'
UNION
SELECT
'2000 ... 10000 €' as amount,
SUM(CASE WHEN direction = 'inbound' AND to_address is null THEN 1 ELSE 0 END) AS inbound,
SUM(CASE WHEN direction = 'outbound' AND to_address is null THEN 1 ELSE 0 END) AS outbound,
SUM(CASE WHEN to_address is not null THEN 1 ELSE 0 END) AS inside
FROM user_wallet_movement
WHERE
(amount*base_rate) > 2000 AND (amount*base_rate) < 10000
AND status = 'execute'
AND mov_date > '2020-01-07'
AND mov_date < '2021-06-30'
UNION
SELECT
'10000 ... 50000 €' as amount,
SUM(CASE WHEN direction = 'inbound' AND to_address is null THEN 1 ELSE 0 END) AS inbound,
SUM(CASE WHEN direction = 'outbound' AND to_address is null THEN 1 ELSE 0 END) AS outbound,
SUM(CASE WHEN to_address is not null THEN 1 ELSE 0 END) AS inside
FROM user_wallet_movement
WHERE
(amount*base_rate) > 10000 AND (amount*base_rate) < 50000
AND status = 'execute'
AND mov_date > '2020-01-07'
AND mov_date < '2021-06-30'
UNION
SELECT
'50000 ... 200000 €' as amount,
SUM(CASE WHEN direction = 'inbound' AND to_address is null THEN 1 ELSE 0 END) AS inbound,
SUM(CASE WHEN direction = 'outbound' AND to_address is null THEN 1 ELSE 0 END) AS outbound,
SUM(CASE WHEN to_address is not null THEN 1 ELSE 0 END) AS inside
FROM user_wallet_movement
WHERE
(amount*base_rate) > 50000 AND (amount*base_rate) < 200000
AND status = 'execute'
AND mov_date > '2020-01-07'
AND mov_date < '2021-06-30'
UNION
SELECT
'> 200000 €' as amount,
SUM(CASE WHEN direction = 'inbound' AND to_address is null THEN 1 ELSE 0 END) AS inbound,
SUM(CASE WHEN direction = 'outbound' AND to_address is null THEN 1 ELSE 0 END) AS outbound,
SUM(CASE WHEN to_address is not null THEN 1 ELSE 0 END) AS inside
FROM user_wallet_movement
WHERE
(amount*base_rate) > 200000
AND status = 'execute'
AND mov_date > '2020-01-07'
AND mov_date < '2021-06-30'
As you can see the conditions are all the same, the only thing that changes is the condition on (amount*base_rate).
Is there a way to optimize this?
Thank you

Answer 1

Thanks to #user202729's comment I got the solution
SELECT
Conds.Description,
SUM(CASE WHEN direction = 'inbound' AND to_address is null THEN 1 ELSE 0 END) AS inbound,
SUM(CASE WHEN direction = 'outbound' AND to_address is null THEN 1 ELSE 0 END) AS outbound,
SUM(CASE WHEN to_address is not null THEN 1 ELSE 0 END) AS inside
FROM user_wallet_movement uwm
JOIN (
SELECT '< 100' AS Description, 0 AS min, 100 AS max
UNION ALL
SELECT '100 ... 500' AS Description, 100 AS min, 500 AS max
UNION ALL
SELECT '500 ... 2000' AS Description, 500 AS min, 2000 AS max
UNION ALL
SELECT '2000 ... 10000' AS Description, 2000 AS min, 10000 AS max
UNION ALL
SELECT '10000 ... 50000' AS Description, 10000 AS min, 50000 AS max
UNION ALL
SELECT '50000 ... 200000' AS Description, 50000 AS min, 200000 AS max
UNION ALL
SELECT '> 200000' AS Description, 200000 AS min, 9999999999 AS max
) as Conds ON (uwm.amount*uwm.base_rate) >= min AND (uwm.amount*uwm.base_rate) < Conds.max
WHERE uwm.status = 'execute'
AND uwm.mov_date > '2020-01-07'
AND uwm.mov_date < '2021-06-30'
GROUP BY Conds.Description

Answer 2

If this is going to be a frequent query and the date range is likely to change, then build and maintain a summary table.
Such a table would have (perhaps) one row per day and status. The columns would be counts for each amount range.
Each night, update the table with the previous day's counts.
Then write the query to reach into the summary table; it will be a lot faster.
More: http://mysql.rjweb.org/doc.php/summarytables