I have a DB structured like this:
CATEGORIES > SUBCATEGORIES > PRODUCTS
In just want show the categories which have an number of products related to it, but I don't know if my method is the best. I thinking in putting some SELECT statement inside the first WHERE, but this sounds "unpratical". I searched into Goog开发者_StackOverflowle/MySQl docs and nothing helped me.
Example of what I have done with CATEGORIES > SUBCATEGORIES:
SELECT c.*
FROM categories c
WHERE
(
SELECT count(*)
FROM subcategories sc
WHERE sc.id_categories = c.id
) > 2
With this query I can sucessfully see which categories have more than 2 subcategories related to them, I just could do the same, adding the PRODUCTS table to this query. But I almost sure this is going to slow the query. There is some more fast way of doing this type of query? Thanks!
Something like this, I don't know your FK relations, so I'm just guessing here.
SELECT *
FROM Categories
WHERE EXISTS(
SELECT NULL
FROM products
JOIN SubCategories ON products.fkSubCatID = SubCategories.PkSubCatID
WHERE SubCategories.fkCatID = Categories.pkCatID
HAVING Count(*) > 2)
Sub selects are designed for this purpose.
SELECT * FROM categories
WHERE
( SELECT COUNT(id)
FROM subcategories
WHERE
category=categories.id AND
( SELECT COUNT(id)
FROM products
WHERE
subcategory=subcategories.id
)
)
Note that this query might not be optimal. And might not work as is not tested.
EDIT:
Following probably will work faster:
SELECT * FROM categories
WHERE
EXISTS(
SELECT COUNT(id)
FROM subcategories
WHERE
category=categories.id AND
EXISTS(
SELECT *
FROM products
WHERE
subcategory=subcategories.id
)
)
精彩评论