Calling SQL foreign key data with PHP - php

Question

I need to get a better grasp on the process of manipulating and utilizing the SQL tables I need to make so I can continue figuring out exactly how I should make them and structure them to work.
If I have a table for shirts and another table for sizes and I use a foreign key on the shirts table to link to the sizes table to represent multiple options for that column. Do I only need to call on the shirts table in the PHP coding? If so how do I tell the PHP to gather whatever options are available for each row on the sizes table?
If in the table it has
vneck sizes,
scoop neck sizes
and I have it set where the vnecks only have s,m,l,1x and the scoop necks have xs,s,m,l,1x,2x,3x. How can I code the PHP to recognize the difference I have logically set in each row for that column?

Answer 1

It sounds like you actually need to have at least three tables, one for shirts, one for sizes , and one to relate shirts to sizes. There are any number of way you can use PHP to query the data, but most likely you would want to simply query using a JOIN to get data from all tables at the same time.
So perhaps something like this:
shirts table:
shirt_id (auto-incrementing primary key)
...other shirt-related fields
sizes table:
size_id (auto-incrementing primary key)
size_value (i.e. S, M, L)
...other size-related fields
shirt_sizes table:
shirt_id (foreign key to shirts table)
size_id (foreign key to sizes table)
(you have compound primary key across these two fields)
An you would query it like
SELECT * (or whatever fields you need)
FROM shirts
INNER JOIN shirt_sizes ON shirts.shirt_id = shirt_sizes.shirt_id
INNER JOIN size ON shirt_sizes.size_id = sizes.size_id

Answer 2

With the following table structure:
CREATE TABLE `shirt` (
`id` INTEGER NOT NULL,
`name` VARCHAR(32),
PRIMARY KEY( `id` )
);
CREATE TABLE `size` (
`id` INTEGER NOT NULL,
`name` VARCHAR(4),
PRIMARY KEY( `id` )
);
CREATE TABLE `shirt_size` (
`shirtId` INTEGER NOT NULL,
`sizeId` INTEGER NOT NULL,
PRIMARY KEY( `shirtId`, `sizeId` ),
FOREIGN KEY( `shirtId` ) REFERENCES `shirt`( `id` ),
FOREIGN KEY( `sizeId` ) REFERENCES `size`( `id` )
);
And this data:
INSERT INTO
`shirt` ( `id`, `name` )
VALUES
( 1, "vneck" ),
( 2, "scoop neck" );
INSERT INTO
`size` ( `id`, `name` )
VALUES
( 1, "xs" ), ( 2, "s" ), ( 3, "m" ),
( 4, "l" ), ( 5, "1x" ), ( 6, "2x" ), ( 7, "3x" );
INSERT INTO
`shirt_size` ( `shirtId`, `sizeId` )
VALUES
( 1, 2 ), ( 1, 3 ), ( 1, 4 ), ( 1, 5 ),
( 2, 1 ), ( 2, 2 ), ( 2, 3 ), ( 2, 4 ), ( 2, 5 ), ( 2, 6 ), ( 2, 7 );
In MySQL you could do:
SELECT
`shirt`.`id`,
`shirt`.`name`,
GROUP_CONCAT( `size`.`name` ) as `sizes`
FROM
`shirt`
JOIN
`shirt_size`
ON `shirt_size`.`shirtId` = `shirt`.`id`
JOIN
`size`
ON `size`.`id` = `shirt_size`.`sizeId`
GROUP BY `shirt`.`id`;
Which would result in something like:
+----+------------+-------------------+
| id | name | sizes |
+----+------------+-------------------+
| 1 | vneck | s,m,l,1x |
+----+------------+-------------------+
| 2 | snoop neck | xs,s,m,l,1x,2x,3x |
+----+------------+-------------------+
Not sure if other RDBMS's have aggregate functions similar to MySQL's GROUP_CONCAT(). If not, then use something like:
SELECT
`shirt`.`id`,
`shirt`.`name` as `shirtName`,
`size`.`name` as `sizeName`
FROM
`shirt`
JOIN
`shirt_size`
ON `shirt_size`.`shirtId` = `shirt`.`id`
JOIN
`size`
ON `size`.`id` = `shirt_size`.`sizeId`;
Which will give you multiple rows for every size, with each shirt.