MySQL get list of dates and join with another tab - mysql

Question

I'm trying to generate seven rows for each date in the last seven days, and join with a query from transactions table. The aim is to have a table with each date, and the cumulative total of the quantity column in transactions from the first entry up to the date:
| date | stockOnDate |
|---------------|---------------|
| 2021-10-15 | 10 |
| 2021-10-16 | 3 |
| 2021-10-17 | 0 |
| 2021-10-18 | 9 |
| 2021-10-19 | 15 |
| 2021-10-20 | 15 |
| 2021-10-21 | 15 |
I can get the list of dates, and can join, but can't filter the nested queries:
SELECT v.*, t.*
FROM ( SELECT DATE(ADDDATE(DATE_SUB(NOW(),INTERVAL 7 DAY), t3*1000 + t2*100 + t1*10 + t0)) AS `date`
FROM (SELECT 0 t0 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t0,
(SELECT 0 t1 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t1,
(SELECT 0 t2 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t2,
(SELECT 0 t3 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t3 ) AS v
LEFT JOIN (SELECT SUM(quantity) AS stockOnDate, DATE(timestamp) as tDate
FROM `transactions`
WHERE tDate <= v.`date`) AS t ON t.tDate = v.`date`
WHERE v.`date` >= DATE_SUB(NOW(),INTERVAL 7 DAY) AND v.`date` <= DATE(NOW())
But I'm receviing the following error:
#1054 - Unknown column 'tDate' in 'where clause'
If I replace WHERE tDate <= v.date with WHERE DATE(timestamp) <= v.date I get the same error for v.date - I can't seem to access the value of the parent tables.
I'm not great with MySQL but can't seem to find a solution, where am I going wrong?
Solution by ProGuru
Thanks to ProGuru's answer below, the below query works as expected (using <= instead of = in the JOIN was key)
SELECT b.date, SUM(a.quantity) AS stockOnDate
FROM (
SELECT DATE(ADDDATE(DATE_SUB(NOW(),INTERVAL 6 DAY), t3*1000 + t2*100 + t1*10 + t0)) AS `date`
FROM (SELECT 0 t0 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t0,
(SELECT 0 t1 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t1,
(SELECT 0 t2 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t2,
(SELECT 0 t3 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t3
) b
LEFT JOIN transactions a ON DATE(a.timestamp) <= b.date
WHERE b.date BETWEEN DATE(NOW()) - INTERVAL 6 DAY AND DATE(NOW())
AND a.organisationId = 1
GROUP BY b.date
ORDER BY b.date ASC
I can also change the GROUP BY to GROUP BY a.itemID, b.date to get the stock level on the given date for each itemId.

Answer 1

To get the cumulative sum of quantity, the date join needs to use <=
Revised SQL
SELECT b.date, SUM(COALESCE(a.quantity, 0))
FROM (
SELECT DATE(ADDDATE(DATE_SUB(NOW(),INTERVAL 7 DAY), t3*1000 + t2*100 + t1*10 + t0)) AS `date`
FROM (SELECT 0 t0 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t0,
(SELECT 0 t1 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t1,
(SELECT 0 t2 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t2,
(SELECT 0 t3 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t3
) b
LEFT JOIN transactions a ON DATE(a.timestamp) <= b.date
WHERE b.date BETWEEN DATE(NOW()) - INTERVAL 6 DAY AND DATE(NOW())
GROUP BY b.date
ORDER BY b.date ASC
https://dbfiddle.uk/?rdbms=mysql_8.0&fiddle=34f78e3d7c9225727ac2e728588759e2

Answer 2

To use alias in where clause, you must encapsulate your query in another select.
Example :
SELECT * FROM
(SELECT SUM(quantity) AS stockOnDate, DATE(timestamp) as tDate FROM `transactions`) seb
WHERE seb.tDate <= v.`date`
Update try this:
SELECT v.*, t.*
FROM ( SELECT DATE(ADDDATE(DATE_SUB(NOW(),INTERVAL 7 DAY), t3*1000 + t2*100 + t1*10 + t0)) AS `date`
FROM (SELECT 0 t0 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t0,
(SELECT 0 t1 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t1,
(SELECT 0 t2 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t2,
(SELECT 0 t3 UNION SELECT 1 UNION SELECT 2 UNION SELECT 3 UNION SELECT 4 UNION SELECT 5 UNION SELECT 6 UNION SELECT 7 UNION SELECT 8 UNION SELECT 9) t3 ) AS v
LEFT JOIN (SELECT SUM(quantity) AS stockOnDate, DATE(timestamp) as tDate FROM transactions) AS t ON t.tDate = v.`date`
WHERE t.tDate <= v.`date` and v.`date` >= DATE_SUB(NOW(),INTERVAL 7 DAY) AND v.`date` <= DATE(NOW())
here tDate can be used

Answer 3

Firstly you need to understand that in this subquery
(SELECT SUM(quantity) AS stockOnDate, DATE(timestamp) as tDate
FROM `transactions`
WHERE tDate <= v.`date`) AS t
there is nothing defined as date or even the alias v. You're basically telling the query to look for v.date when it doesn't exists in the subquery.
I think this is what you're looking for:
SELECT dt, IFNULL(SUM(quantity),0) AS stockOnDate
FROM (SELECT STR_TO_DATE(CONCAT(LEFT(NOW(),7),LPAD(seq,2,0)), '%Y-%m%d') dt
FROM seq_1_to_31) v
LEFT JOIN `transactions` t ON dt=DATE(timestamp)
WHERE dt >= (SELECT MIN(DATE(timestamp)) FROM `transactions`)
AND dt <= (SELECT MAX(DATE(timestamp)) FROM `transactions`)
GROUP BY dt;
Here I'm using MariaDB's built-in sequence engine to generate days; concatenated with current year-month combination then uses STR_TO_DATE to make it recognizable as proper date format. The query generating those is in subquery v.
Second option is using a recursive common table expression to generate the date range based on existing date data in the table. Here is the query:
WITH RECURSIVE cte AS (
SELECT MIN(DATE(timestamp)) mndt, MAX(DATE(timestamp)) mxdt FROM `transactions`
UNION ALL
SELECT mndt+INTERVAL 1 DAY, mxdt FROM cte WHERE mndt+INTERVAL 1 DAY <= mxdt)
SELECT mndt, IFNULL(SUM(quantity),0) AS stockOnDate
FROM cte
LEFT JOIN `transactions` t ON mndt=DATE(timestamp)
GROUP BY mndt;
Both the query above eliminates the need to include a long subquery that generates the date range. This is the benefit of newer MySQL & MariaDB versions.
Demo fiddle