Complex JOIN SQL query in Laravel 5 - mysql

Question

I have three tables that I need to join: products, orders_items, orders
The products table contains, obviously, the products. When a user adds a product to their cart, it creates an order and a line item in the order_items table. When the user pays, it puts the total paid into the orders table. Therefore, purchased products removed from inventory only include products that are attached to an order_item attached to an order that has been marked paid.
I need to return all the products, but also somehow calculate how many are left in inventory. The products table contains a field called "initial_quantity" which is how many of the product we started with. If that field is NULL, there's no limit on the product.
Here is my current query, which works just fine for returning all the products of a certain type:
$swims = Product::where('product_type_id', 1)
->whereDate('visible_date', '<=', Carbon::now())
->where('active', true)
->where(function ($query) {
$query->where('end_date', '>=', Carbon::now())
->orWhere('end_date', null);
})
->get();
Here is the SQL for that query:
select * from `products` where `product_type_id` = 1 and date(`visible_date`) <= '2019-08-08 17:10:12' and `active` = 1 and (`end_date` >= '2019-08-08 17:10:12' or `end_date` is null)
The above query returns three results:
+----+-------------+------------+------------+--------------+------------------+-----------------+--------+
| id | name | base_price | end_date | visible_date | initial_quantity | product_type_id | active |
+----+-------------+------------+------------+--------------+------------------+-----------------+--------+
| 1 | membership | 30.97 | NULL | 2019-08-04 | NULL | 1 | 1 |
| 12 | repellendus | 779.24 | 2027-03-16 | 1990-12-19 | NULL | 1 | 1 |
| 16 | ducimus | 708.33 | NULL | 1999-03-24 | NULL | 1 | 1 |
+----+-------------+------------+------------+--------------+------------------+-----------------+--------+
I can run another query to see how much of each product has been sold:
select product_id, SUM(quantity) sold from `order_items` inner join `orders` on `order_items`.`order_id` = `orders`.`id` and `orders`.`paid` is not null group by `product_id`
This query returns:
+------------+------+
| product_id | sold |
+------------+------+
| 1 | 5 |
| 7 | 3 |
| 11 | 1 |
| 12 | 1 |
+------------+------+
What I'm looking for is a way to combine those two (or similar) queries so I get all the products I'm looking for, plus the quantity of those products that have been sold, to give me a result similar to this:
+----+-------------+------------+------------+--------------+------------------+-----------------+--------+------+
| id | name | base_price | end_date | visible_date | initial_quantity | product_type_id | active | sold |
+----+-------------+------------+------------+--------------+------------------+-----------------+--------+------+
| 1 | membership | 30.97 | NULL | 2019-08-04 | NULL | 1 | 1 | 5 |
| 12 | repellendus | 779.24 | 2027-03-16 | 1990-12-19 | NULL | 1 | 1 | 1 |
| 16 | ducimus | 708.33 | NULL | 1999-03-24 | NULL | 1 | 1 | 0 |
+----+-------------+------------+------------+--------------+------------------+-----------------+--------+------+
I would like a query that returns all products I'm searching for and also how many are remaining. I've tried join, leftJoin, rightJoin, subquery joins and unions. Laravel doesn't seem to have outer joins or I would have tried those too.
I have tried this query and many other iterations and I always get a syntax error:
select * from (`products` where `product_type_id` = 1 and date(`visible_date`) <= '2019-08-08 17:10:12' and `active` = 1 and (`end_date` >= '2019-08-08 17:10:12' or `end_date` is null)) x INNER JOIN (select product_id, SUM(quantity) sold from `order_items` inner join `orders` on `order_items`.`order_id` = `orders`.`id` and `orders`.`paid` is not null group by `product_id`) y ON x.id = y.product_id
Can anybody come up with the query I'm looking for? If not in Laravel then in raw SQL would be fine and I can use it as-is or try to Laravelize it.
Thank you in advance.
Here is my MRE:
CREATE TABLE `products` (
`id` bigint(20) UNSIGNED NOT NULL,
`name` varchar(191) COLLATE utf8mb4_unicode_ci NOT NULL,
`base_price` double(5,2) NOT NULL,
`end_date` date DEFAULT NULL,
`visible_date` date DEFAULT NULL,
`initial_quantity` smallint(5) UNSIGNED DEFAULT NULL,
`product_type_id` int(10) UNSIGNED NOT NULL,
`active` tinyint(3) UNSIGNED NOT NULL DEFAULT '1'
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci;
INSERT INTO `products` (`id`, `name`, `base_price`, `end_date`, `visible_date`, `initial_quantity`, `product_type_id`, `active`) VALUES
(1, 'membership', 30.97, NULL, '2019-08-04', NULL, 1, 1),
(2, 'delectus', 347.44, NULL, '1975-06-08', 45, 1, 0),
(3, 'aut', 283.36, '1981-05-30', '1973-08-18', NULL, 3, 0),
(4, 'adipisci', 984.00, '1986-01-04', '1989-03-29', NULL, 3, 0),
(5, 'voluptas', 310.55, '2012-05-04', '1992-11-16', 45, 2, 1),
(6, 'quia', 657.81, '1976-11-23', '1978-10-23', 57, 2, 0),
(7, 'delectus', 601.91, NULL, '1987-03-09', 53, 2, 1),
(8, 'consequatur', 723.25, '1994-09-24', '1985-01-23', 33, 1, 0),
(9, 'perferendis', 427.33, NULL, '1995-12-08', 51, 3, 1),
(10, 'id', 674.39, '1974-12-13', '2007-05-12', 41, 2, 0),
(11, 'maxime', 133.49, NULL, '1986-07-15', NULL, 2, 1),
(12, 'repellendus', 779.24, '2027-03-16', '1990-12-19', NULL, 1, 1),
(13, 'sit', 956.76, NULL, '1993-07-02', 39, 2, 0),
(14, 'molestiae', 425.16, NULL, '1981-03-25', NULL, 3, 1),
(15, 'omnis', 418.78, '1976-11-03', '2007-10-07', NULL, 3, 1),
(16, 'ducimus', 708.33, NULL, '1999-03-24', NULL, 1, 1),
(17, 'numquam', 328.26, '2011-09-19', '2001-10-07', NULL, 3, 1),
(18, 'est', 962.26, NULL, '1970-12-24', NULL, 1, 0),
(19, 'quis', 520.18, NULL, '1978-03-25', NULL, 3, 0),
(20, 'facere', 891.89, NULL, '1979-10-31', 36, 3, 0),
(21, 'voluptatem', 518.47, '1982-06-26', '1975-03-14', NULL, 1, 1);
COMMIT;
CREATE TABLE `order_items` (
`order_id` bigint(20) UNSIGNED NOT NULL,
`product_id` bigint(20) UNSIGNED NOT NULL,
`quantity` tinyint(3) UNSIGNED NOT NULL DEFAULT '1',
`price` double(5,2) NOT NULL,
`tax` double(5,2) NOT NULL,
`buy_now` tinyint(3) UNSIGNED NOT NULL DEFAULT '1',
`created_at` timestamp NULL DEFAULT NULL,
`updated_at` timestamp NULL DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci;
INSERT INTO `order_items` (`order_id`, `product_id`, `quantity`, `price`, `tax`, `buy_now`, `created_at`, `updated_at`) VALUES
(1, 1, 1, 30.97, 4.03, 1, '2019-08-04 11:35:26', '2019-08-04 11:35:26'),
(1, 12, 1, 779.24, 654.56, 1, '2019-08-04 11:35:26', '2019-08-04 11:35:26'),
(2, 1, 1, 30.97, 4.03, 1, '2019-08-04 11:56:16', '2019-08-04 11:56:16'),
(3, 1, 1, 30.97, 4.03, 1, '2019-08-07 17:25:43', '2019-08-07 17:25:43'),
(3, 7, 1, 601.91, 210.67, 1, '2019-08-07 17:25:43', '2019-08-07 17:25:43'),
(4, 1, 1, 30.97, 4.03, 1, '2019-08-08 15:44:45', '2019-08-08 15:44:45'),
(4, 7, 1, 601.91, 210.67, 1, '2019-08-08 15:44:45', '2019-08-08 15:44:45'),
(5, 1, 1, 30.97, 4.03, 1, '2019-08-08 15:46:07', '2019-08-08 15:46:07'),
(5, 11, 1, 133.49, 88.10, 1, '2019-08-08 15:46:07', '2019-08-08 15:46:07'),
(6, 1, 1, 30.97, 4.03, 1, '2019-08-08 15:46:55', '2019-08-08 15:46:55'),
(6, 16, 1, 708.33, 687.08, 1, '2019-08-08 15:46:55', '2019-08-08 15:46:55'),
(7, 1, 1, 30.97, 4.03, 1, '2019-08-08 15:47:37', '2019-08-08 15:47:37'),
(7, 7, 1, 601.91, 210.67, 1, '2019-08-08 15:47:37', '2019-08-08 15:47:37'),
(8, 1, 1, 30.97, 4.03, 1, '2019-08-08 15:48:13', '2019-08-08 15:48:13'),
(8, 11, 1, 133.49, 88.10, 1, '2019-08-08 15:48:13', '2019-08-08 15:48:13');
CREATE TABLE `orders` (
`id` bigint(20) UNSIGNED NOT NULL,
`user_id` bigint(20) UNSIGNED NOT NULL,
`subtotal` double(7,2) DEFAULT NULL,
`tax` double(7,2) DEFAULT NULL,
`paid` double(7,2) DEFAULT NULL,
`transaction` varchar(191) COLLATE utf8mb4_unicode_ci DEFAULT NULL,
`created_at` timestamp NULL DEFAULT NULL,
`updated_at` timestamp NULL DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci;
INSERT INTO `orders` (`id`, `user_id`, `subtotal`, `tax`, `paid`, `transaction`, `created_at`, `updated_at`) VALUES
(1, 1, 810.21, 658.59, 12343.00, NULL, '2019-08-04 11:35:26', '2019-08-04 11:35:26'),
(2, 1, 30.97, 4.03, NULL, NULL, '2019-08-04 11:56:16', '2019-08-04 11:56:16'),
(3, 1, 632.88, 214.70, 847.58, NULL, '2019-08-05 17:25:43', '2019-08-07 17:25:43'),
(4, 5, 632.88, 214.70, 847.58, NULL, '2019-08-08 15:44:45', '2019-08-08 15:44:45'),
(5, 5, 164.46, 92.13, 256.59, NULL, '2019-08-08 15:46:07', '2019-08-08 15:46:07'),
(6, 5, 739.30, 691.11, NULL, NULL, '2019-08-08 15:46:55', '2019-08-08 15:46:55'),
(7, 10, 632.88, 214.70, 847.58, NULL, '2019-08-08 15:47:37', '2019-08-08 15:47:37'),
(8, 10, 164.46, 92.13, NULL, NULL, '2019-08-08 15:48:13', '2019-08-08 15:48:13');

Answer 1

SELECT a.*
, COALESCE(b.sold,0) sold
FROM
( SELECT *
FROM products p
WHERE p.product_type_id = 1
AND DATE(p.visible_date) <= '2019-08-08 17:10:12'
AND p.active = 1
AND (p.end_date >= '2019-08-08 17:10:12' or p.end_date IS NULL)
) a
LEFT
JOIN
( SELECT product_id
, SUM(quantity) sold
FROM order_items oi
JOIN orders o
ON o.id = oi.order_id
AND o.paid is not null
GROUP
BY product_id
) b
ON b.product_id = a.id;
...or, better...
SELECT p.*
, COALESCE(y.sold,0) sold
FROM products p
LEFT
JOIN
( SELECT product_id
, SUM(quantity) sold
FROM order_items oi
JOIN orders o
ON o.id = oi.order_id
AND o.paid is not null
GROUP
BY product_id
) y
ON y.product_id = p.id
WHERE p.product_type_id = 1
AND DATE(p.visible_date) <= '2019-08-08 17:10:12'
AND p.active = 1
AND (p.end_date >= '2019-08-08 17:10:12' OR p.end_date IS NULL)