MySQL: optimization of table (indexing, foreign key) with no primary keys - mysql

Each member has 0 or more orders. Each order contains at least 1 item.
memberid - varchar, not integer - that's OK (please do not mention that's not very good, I can't change it).
So, thera 3 tables: members, orders and order_items. Orders and order_items are below:
CREATE TABLE `orders` (
`orderid` INT(11) UNSIGNED NOT NULL AUTO_INCREMENT,
`memberid` VARCHAR( 20 ),
`Time` TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP ,
`info` VARCHAR( 3200 ) NULL ,
PRIMARY KEY (orderid) ,
FOREIGN KEY (memberid) REFERENCES members(memberid)
) ENGINE = InnoDB;
CREATE TABLE `order_items` (
`orderid` INT(11) UNSIGNED NOT NULL,
`item_number_in_cart` tinyint(1) NOT NULL , --- 5 items in cart= 5 rows
`price` DECIMAL (6,2) NOT NULL,
FOREIGN KEY (orderid) REFERENCES orders(orderid)
) ENGINE = InnoDB;
So, order_items table looks like:
orderid - item_number_in_cart - price:
...
1000456 - 1 - 24.99
1000456 - 2 - 39.99
1000456 - 3 - 4.99
1000456 - 4 - 17.97
1000457 - 1 - 20.00
1000458 - 1 - 99.99
1000459 - 1 - 2.99
1000459 - 2 - 69.99
1000460 - 1 - 4.99
...
As you see, order_items table has no primary keys (and I think there is no sense to create an auto_increment id for this table, because once we want to extract data, we always extract it as WHERE orderid='1000456' order by item_number_in_card asc - the whole block, id woudn't be helpful in queries).
Once data is inserted into order_items, it's not UPDATEd, just SELECTed.
The questions are:
I think it's a good idea to put index on item_number_in_cart. Could anybody please confirm that?
Is there anything else I have to do with order_items to increase the performance, or that looks pretty good? I could miss something because I'm a newbie.
Thank you in advance.

Primary keys can span multiple columns. You can't use the PRIMARY attribute of columns to do this, but you can define a separate primary key with multiple columns:
CREATE TABLE `order_items` (
`orderid` INT(11) UNSIGNED NOT NULL,
`item_number_in_cart` tinyint(1) NOT NULL , --- 5 items in cart= 5 rows
`price` DECIMAL (6,2) NOT NULL,
PRIMARY KEY (orderid, item_number_in_cart),
FOREIGN KEY (orderid) REFERENCES orders(orderid)
) ENGINE = InnoDB;
Moreover, a primary key is simply a unique key where every column is not null with a certain name; you can create your own unique keys on non-nullable columns to get the same effects.
You'll not likely get much of a performance improvement by indexing item_number_in_cart; as the number of line items for a given order will tend to be small, sorting by item_number_in_cart won't take much time or memory. However, including the column in a primary key will help with data consistency.

Index on item_number_in_cart won't be used. It's tiny int, not selective enough, and won't even considered by the engine once you have 2 records. You can add it as a second column to the existing index on orderid (since you created FK constraint on orderid, mysql automatically adds an index on this field).
You say that data in order_items never updated, but I think it can be deleted; doing so without primary key will be problematic.

Well I'd be having an autoinc anyway, as I'm a big believer in surrogate keys, but as suggested by alex07 an index, or even primary key of orderid,item_number_in_cart should sort things out. Note the order by item_number will be using a two pass sort, (get the data and then sort it in the number order) so an index / key will chop that out straight off so you'd want that index even with a surrogate key.

Related

MYSQL table for related records in another table

What is the best way to have a table to maintain related records of another table.
Example:
mytbl
-----
id sku
1 sk1
2 sk2
3 sk3
4 sk4
5 sk5
6 sk6
7 sk7
Lets say records 1, 4 and 3 are 'related'
So I want to maintain a table that tells me that they are.
relatedTbl
----------
sku related_sku
sk1 sk3
sk1 sk4
sk3 sk4
This solution would work but, is there a better solution?
EDIT: I used skus in the relatedTbl but I know I could (better) to use ids. The question is about the structure of the table more than what foreign key to use.
You have the correct solution. As you indicated, use the ID. If sku is unique, consider using it as a natural PK.
CREATE TABLE IF NOT EXISTS `mytbl` (
`id` INT UNSIGNED NOT NULL AUTO_INCREMENT,
`sku` VARCHAR(45) NULL,
PRIMARY KEY (`id`))
ENGINE = InnoDB;
CREATE TABLE IF NOT EXISTS `relatedTbl` (
`mytbl_id` INT UNSIGNED NOT NULL,
`mytbl_id1` INT UNSIGNED NOT NULL,
PRIMARY KEY (`mytbl_id`, `mytbl_id1`),
INDEX `fk_mytbl_has_mytbl_mytbl1_idx` (`mytbl_id1` ASC),
INDEX `fk_mytbl_has_mytbl_mytbl_idx` (`mytbl_id` ASC),
CONSTRAINT `fk_mytbl_has_mytbl_mytbl`
FOREIGN KEY (`mytbl_id`)
REFERENCES `mytbl` (`id`)
ON DELETE NO ACTION
ON UPDATE NO ACTION,
CONSTRAINT `fk_mytbl_has_mytbl_mytbl1`
FOREIGN KEY (`mytbl_id1`)
REFERENCES `mytbl` (`id`)
ON DELETE NO ACTION
ON UPDATE NO ACTION)
ENGINE = InnoDB;
You may want to consider adding a third field to 'mytbl' in to store a unique key for common records. for instance, field 3 would be named "uniqID", and records 1, 4 and 3 are 'related' the table would then be:
mytbl
id sku uniqID
1 sk1 1
2 sk2
3 sk3 1
4 sk4 1
5 sk5
6 sk6
7 sk7
you can then use a 'WHERE uniqID=1' clause at the end of your select statement to get the common attributes

In SQL, is it OK for two tables to refer to each other?

In this system, we store products, images of products (there can be many image for a product), and a default image for a product. The database:
CREATE TABLE `products` (
`ID` int(10) unsigned NOT NULL AUTO_INCREMENT,
`NAME` varchar(255) NOT NULL,
`DESCRIPTION` text NOT NULL,
`ENABLED` tinyint(1) NOT NULL DEFAULT '1',
`DATEADDED` datetime NOT NULL,
`DEFAULT_PICTURE_ID` int(10) unsigned DEFAULT NULL,
PRIMARY KEY (`ID`),
KEY `Index_2` (`DATEADDED`),
KEY `FK_products_1` (`DEFAULT_PICTURE_ID`),
CONSTRAINT `FK_products_1` FOREIGN KEY (`DEFAULT_PICTURE_ID`) REFERENCES `products_pictures` (`ID`) ON DELETE SET NULL ON UPDATE SET NULL
) ENGINE=InnoDB AUTO_INCREMENT=30 DEFAULT CHARSET=utf8;
CREATE TABLE `products_pictures` (
`ID` int(10) unsigned NOT NULL AUTO_INCREMENT,
`IMG_PATH` varchar(255) NOT NULL,
`PRODUCT_ID` int(10) unsigned NOT NULL,
PRIMARY KEY (`ID`),
KEY `FK_products_pictures_1` (`PRODUCT_ID`),
CONSTRAINT `FK_products_pictures_1` FOREIGN KEY (`PRODUCT_ID`) REFERENCES `products` (`ID`) ON DELETE CASCADE
) ENGINE=InnoDB AUTO_INCREMENT=20 DEFAULT CHARSET=utf8 ROW_FORMAT=DYNAMIC;
as you can see, products_pictures.PRODUCT_ID -> products.ID and products.DEFAULT_PICTURE_ID -> products_pictures.ID, so a cycle reference. Is it OK?
No, it's not OK. Circular references between tables are messy. See this (decade old) article: SQL By Design: The Circular Reference
Some DBMS can handle these, and with special care, but MySQL will have issues.
Option 1
As your design, to make one of the two FKs nullable. This allows you to solve the chicken-and-egg problem (which table should I first Insert into?).
There is a problem though with your code. It will allow a product to have a default picture where that picture will be referencing another product!
To disallow such an error, your FK constraint should be:
CONSTRAINT FK_products_1
FOREIGN KEY (id, default_picture_id)
REFERENCES products_pictures (product_id, id)
ON DELETE RESTRICT --- the SET NULL options would
ON UPDATE RESTRICT --- lead to other issues
This will require a UNIQUE constraint/index in table products_pictures on (product_id, id) for the above FK to be defined and work properly.
Option 2
Another approach is to remove the Default_Picture_ID column form the product table and add an IsDefault BIT column in the picture table. The problem with this solution is how to allow only one picture per product to have that bit on and all others to have it off. In SQL-Server (and I think in Postgres) this can be done with a partial index:
CREATE UNIQUE INDEX is_DefaultPicture
ON products_pictures (Product_ID)
WHERE IsDefault = 1 ;
But MySQL has no such feature.
Option 3
This approach, allows you to even have both FK columns defined as NOT NULL is to use deferrable constraints. This works in PostgreSQL and I think in Oracle. Check this question and the answer by #Erwin: Complex foreign key constraint in SQLAlchemy (the All key columns NOT NULL Part).
Constraints in MySQL cannot be deferrable.
Option 4
The approach (which I find cleanest) is to remove the Default_Picture_ID column and add another table. No circular path in the FK constraints and all FK columns will be NOT NULL with this solution:
product_default_picture
----------------------
product_id NOT NULL
default_picture_id NOT NULL
PRIMARY KEY (product_id)
FOREIGN KEY (product_id, default_picture_id)
REFERENCES products_pictures (product_id, id)
This will also require a UNIQUE constraint/index in table products_pictures on (product_id, id) as in solution 1.
To summarize, with MySQL you have two options:
option 1 (a nullable FK column) with the correction above to enforce integrity correctly
option 4 (no nullable FK columns)
The only issue you're going to encounter is when you do inserts.
Which one do you insert first?
With this, you will have to do something like:
Insert product with null default picture
Insert picture(s) with the newly created product ID
Update the product to set the default picture to one that you just inserted.
Again, deleting will not be fun.
this is just suggestion but if possible create one join table between this table might be helpfull to tracking
product_productcat_join
------------------------
ID(PK)
ProductID(FK)- product table primary key
PictureID(FK) - category table primary key
In the other table you can just hold that field without the foreign key constraint.
it is useful in some cases where you want to process with the smaller table but connect to the bigger table with the result of the process.
For example if you add a product_location table which holds the country, district, city, address and longitude and latitude information. There might be a case that you want to show the product within a circle on the map.
John what your doing isnt anything bad but using PK-FK actually helps with normalizing your data by removing redundant repeating data. Which has some fantastic advantages from
Improved data integrity owing to the elimination of duplicate storage locations for the same data
Reduced locking contention and improved multiple-user concurrency
Smaller files
that is not a cyclic ref, that is pk-fk

Junction tables with non unique entry

I'm beginning to build a stamp collecting web app. Python/flask backend (i think :)) mySQL as db. I don't know much about db design so please keep that in mind if I do some really stupid mistake in the way I thought it out. I was thinking of splitting the data into 3 tables.
users table (all the users should be added upon registration to this table)
stamps table (all stamps should reside here and only modified by me)
owned table (junction table with user_id and stamp_id as foreign keys)
Question : if I put user_id and stamp_id as primary key , there will only be one unique entry of this type for example user_1 has card_1. But user_1 might have a duplicate of card_1 so i should have 2 rows
user_1 card_1
user_1 card_1
Another problem that arises is that I want to include state of owned stamp. For example user_1 might have a card_1 in mint condition and a card_1 in bad condition. As far as I understand I can only enter one unique pair of user_1 card_1 . What can I do to get the desired result? Also if there's a better way of doing this please let me know.
Aditional question. I was using mysql workbench to try to plot the db so I have a question about the sql it generates. the CONSTRAINT "fk_gibberish", is that normal or ... why is that ?
CREATE TABLE IF NOT EXISTS `stampcollect`.`users` (
`user_id` INT UNSIGNED NOT NULL AUTO_INCREMENT ,
`user_username` VARCHAR(45) NULL ,
`user_password` VARCHAR(45) NULL ,
`user_email` VARCHAR(45) NULL ,
PRIMARY KEY (`user_id`) )
CREATE TABLE IF NOT EXISTS `stampcollect`.`stamps` (
`stamp_id` INT UNSIGNED NOT NULL AUTO_INCREMENT ,
`stamp_name` VARCHAR(45) NULL ,
PRIMARY KEY (`stamp_id`) )
CREATE TABLE IF NOT EXISTS `stampcollect`.`owned` (
`user_id` INT NOT NULL ,
`stamp_id` INT NOT NULL ,
`stamp_status` BIT NULL ,
PRIMARY KEY (`user_id`, `stamp_id`) ,
INDEX `fk_{F5DBEF0D-24E0-4AFF-A5CB-2A6A0D448C96}` (`stamp_id` ASC) ,
CONSTRAINT `fk_{22B4468E-A5FB-4702-A8A9-576AA48A0543}`
FOREIGN KEY (`user_id` )
REFERENCES `stampcollect`.`users` (`user_id` ),
CONSTRAINT `fk_{F5DBEF0D-24E0-4AFF-A5CB-2A6A0D448C96}`
FOREIGN KEY (`stamp_id` )
REFERENCES `stampcollect`.`stamps` (`stamp_id` ));
If users can own the same stamp in multiple states then the state should go in the "owned" table and be part of the key. If he can own multiple copies of the same stamp then it would make sense to have a "quantity" column in that table (not part of the key).
Add an id field with auto-increment on your owned table, and make that the primary key.
Regarding the other question: it's just Workbench generating a unique id for your foreign key. You can rename them, just keep them unique.

MySQL 5-star rating datatype?

Would ENUM('1','2','3','4','5') be a sensible datatype for a product rating which must be between 1 and 5?
Thanks!
Yes, that would be an appropriate data type since it enforces your domain.
If you are going to add (or do any other mathematical operation) them together, however, a numeric data type might be better.
I suggest using
TINYINT UNSIGNED NOT NULL
or, for better ANSI/SQL compatibility, use:
SMALLINT NOT NULL
With an integer type, it is much easier to do calculations. ENUM is not bad, but there is a potential to mess up because it's kind of a dual string/int type (beneath the covers, it's an int, but from the outside, it's a string). And indeed, suppose you do feel the need to go to 3 stars, or 10 stars or so, the migration will be much less painful.
If you are using Mysql 8+ then you use TINYINT with CHECK constraint
-- Product reviews
CREATE TABLE product_review (
product_review_id INT UNSIGNED PRIMARY KEY AUTO_INCREMENT,
user_id INT UNSIGNED NOT NULL,
product_id INT UNSIGNED NOT NULL,
unique(user_id, product_id),
rating TINYINT UNSIGNED NOT NULL CHECK (
rating > 0
AND rating <= 5
),
review VARCHAR(2047) NOT NULL,
created_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP
-- FOREIGN KEY (product_id) REFERENCES product (product_id) ON DELETE CASCADE ON UPDATE CASCADE.
-- FOREIGN KEY (user_id) REFERENCES user (user_id) ON DELETE CASCADE ON UPDATE CASCADE
);
It would not allow value other than 1, 2, 3, 4, 5 and also support any kind of mathematical operation, for example you can get average rating of the product with simple calculation.

Enforce unique rows in MySQL

I have a table in MySQL that has 3 fields and I want to enforce uniqueness among two of the fields. Here is the table DDL:
CREATE TABLE `CLIENT_NAMES` (
`ID` int(11) NOT NULL auto_increment,
`CLIENT_NAME` varchar(500) NOT NULL,
`OWNER_ID` int(11) NOT NULL,
PRIMARY KEY (`ID`),
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
The ID field is a surrogate key (this table is being loaded with ETL).
The CLIENT_NAME is a field that contains names of clients
The OWNER_ID is an id indicates a clients owner.
I thought I could enforce this with a unique index on CLIENT_NAME and OWNER_ID,
ALTER TABLE `DW`.`CLIENT_NAMES`
ADD UNIQUE INDEX enforce_unique_idx(`CLIENT_NAME`, `OWNER_ID`);
but MySQL gives me an error:
Error executing SQL commands to update table.
Specified key was too long; max key length is 765 bytes (error 1071)
Anyone else have any ideas?
MySQL cannot enforce uniqueness on keys that are longer than 765 bytes (and apparently 500 UTF8 characters can surpass this limit).
Does CLIENT_NAME really need to be 500 characters long? Seems a bit excessive.
Add a new (shorter) column that is hash(CLIENT_NAME). Get MySQL to enforce uniqueness on that hash instead.
Have you looked at CONSTRAINT ... UNIQUE?
Something seems a bit odd about this table; I would actually think about refactoring it. What do ID and OWNER_ID refer to, and what is the relationship between them?
Would it make sense to have
CREATE TABLE `CLIENTS` (
`ID` int(11) NOT NULL auto_increment,
`CLIENT_NAME` varchar(500) NOT NULL,
# other client fields - address, phone, whatever
PRIMARY KEY (`ID`),
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
CREATE TABLE `CLIENTS_OWNERS` (
`CLIENT_ID` int(11) NOT NULL,
`OWNER_ID` int(11) NOT NULL,
PRIMARY KEY (`CLIENT_ID`,`OWNER_ID`),
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
I would really avoid adding a unique key like that on a 500 character string. It's much more efficient to enforce uniqueness on two ints, plus an id in a table should really refer to something that needs an id; in your version, the ID field seems to identify just the client/owner relationship, which really doesn't need a separate id, since it's just a mapping.
Here. For the UTF8 charset, MySQL may use up to 3 bytes per character. CLIENT_NAME is 3 x 500 = 1500 bytes. Shorten CLIENT_NAME to 250.
later: +1 to creating a hash of the name and using that as the key.