Question

I am having trouble with a query that I want to retrieve counts for multiple columns from the same table in the subqueries as the main query.

Here is my query:

SELECT 
    CM.entityId,
    (SELECT COUNT(*) 
     FROM _credit_control_main 
     WHERE invoiceAge < 14 AND entityId = CM.entityId) AS under14,
    (SELECT COUNT(*) 
     FROM _credit_control_main 
     WHERE invoiceAge >= 14 AND invoiceAge < 30 AND entityId = CM.entityId) AS under14to30,
    (SELECT COUNT(*) 
     FROM _credit_control_main 
     WHERE invoiceAge >= 30 AND invoiceAge < 60 AND entityId = CM.entityId) AS under30to60,
    (SELECT COUNT(*) 
     FROM _credit_control_main 
     WHERE invoiceAge >= 60 AND invoiceAge < 90 AND entityId = CM.entityId) AS under60to90,
    (SELECT COUNT(*) 
     FROM _credit_control_main 
     WHERE invoiceAge > 90 AND entityId = CM.entityId) AS over90,
    COUNT(*) AS iCount
FROM
    _credit_control_main AS CM
WHERE 
    ((CM.invoiceNet + CM.invoiceVat) - (creditNet+creditVat)) - (CM.paymentTotal - (CM.creditNet + CM.creditVat)) > 0.00
GROUP BY 
    entityId 
ORDER BY 
    `CM`.`entityId` ASC

If I create a basic query based on one of the subqueries I get the correct count, but as it is I get vastly inflated counts.

What am I doing wrong?

Answer 1

I don’t know why your query is wrong but I know how you should write it instead. All these subcount are killers, try understanding this trick and you’ll gain a lot in readability and performance:

SELECT
    CM.entityId,
    SUM(invoiceAge < 14) as under14,
    SUM(invoiceAge >= 14 AND invoiceAge < 30) as under14to30,
    SUM(invoiceAge >= 30 AND invoiceAge < 60) as under30to60,
    SUM(invoiceAge >= 60 AND invoiceAge < 90) as under60to90,
    SUM(invoiceAge > 90) as over90,
    COUNT(*) AS iCount
FROM _credit_control_main AS CM
WHERE ((CM.invoiceNet+CM.invoiceVat)-(creditNet+creditVat))-(CM.paymentTotal-(CM.creditNet+CM.creditVat)) > 0.00
GROUP BY entityId
ORDER BY `CM`.`entityId` ASC;

And by the way, the invoiceAge = 90 fits in no category here.