primary-key

I was wondering if anybody knew a good way to create a unique random integer id for a primary key for a table. I'm using MySQL. The value has to be integer. byte_slave If your're open to suggestions and you can implement it, use UUIDs. MySQL's UUID() function will return a 36 chars value which can be used for ID . If you want to use integer, still, I think you need to create a function getRandID() that you will use in the INSERT statement. This function needs to use random + check of existing ids to return one that is not used before. Check RAND() function for MySQL. In response to: "Because I

I have 3 tables: dentists, groups, and groupdentlink. Many dentists link to many groups through the groupdentlink table. So I'm trying to make a query where it will insert rows into groupdentlink (linking all dentists in the state with all the groups in the state) but only if those rows don't already exist. In a nutshell I want to add new rows without overwriting existing ones or duplicating them. So the intent of the query is something like: INSERT INTO groupdentlink (f_dent_id, f_group_id, f_schedule_id) VALUES ('$_POST[id]', '$groupid', '$scheduleid') WHERE NOT EXISTS ('$_POST[id]', '

Why should I choose UUID over an auto increment number for my entity's primary key? What are the pros and cons? Andrey and Mjg both had good points, but I would add a related performance issue that is significant. With the decoupling of database and key generation also allows applications that have complex relationships between objects to create them all with the keys in place, so that bulk inserts are possible. In the case of auto-increment, all of the objects that own relationships (ie the tables with foreign keys) have to wait for the other side of the relationship (ie the table the foreign

Traditionally I have always used an ID column in SQL (mostly mysql and postgresql). However I am wondering if it is really necessary if the rest of the columns in each row make in unique. In my latest project I have the "ID" column set as my primary key, however I never call it or use it in any way, as the data in the row makes it unique and is much more useful for me. So, if every row in a SQL table is unique, does it need a primary key ID table, and are there ant performance changes with or without one? Thanks! EDIT/Additional info: The specific example that made me ask this question is a

My version of rails is 4.0.0, my version of mysql is Ver 14.14 Distrib 5.7.9, for Win64 (x86_64). I am operating of an older version of rails as I was getting some clashes with the mysql as per my previous question Here . (check Kalelc 's approved answer for my recourse) upon running rake db:migrate I get the following error == CreateUsers: migrating ==================================================== -- create_table(:users) rake aborted! StandardError: An error has occurred, all later migrations canceled: Mysql2::Error: All parts of a PRIMARY KEY must be NOT NULL; if you need NULL in a key,

I have a table user with userID as the primary key. I have another table called Friends . In the Friends table, I have two Users as friends represented by the columns UserID and FrndID where both UserID and FrndID should be a userID in table user . I want to enforce data integrity. Could I use something like this? ADD CONSTRAINT `ufd_users_fk` FOREIGN KEY (`userId`, `friendId`) REFERENCES `users` (`userId`, `userId`) ON DELETE CASCADE ON UPDATE CASCADE; I want to know is REFERENCES users ( userId , userId ) referencing a column multiple times correctly? The reason I am not creating 2 separate