Skip to content
Tutorials Center
My learning

Topics

Lesson 8 of 37 ยท Models and Entities

Primary Keys and Generated Values

Choose between auto-increment, UUID and composite primary keys in TypeORM and Sequelize, see how IDs are generated, and compare them with runnable examples.

Every row needs a stable identity. A primary key is the column (or columns) that uniquely identifies a row, and every relation in later chapters points at it. The main choice is where the value comes from: the database counts, or your application invents it.

Library code is read-only; the runnable stand-ins show key behaviour in plain Node and SQLite.

Auto-increment integers

The simplest option: the database assigns 1, 2, 3, ...

TypeScript
// TypeORM
@PrimaryGeneratedColumn()
id!: number;
TypeScript
// Sequelize
id: { type: DataTypes.INTEGER, autoIncrement: true, primaryKey: true }

Pros: small, fast, readable. Cons: guessable (users can try /orders/1001), awkward to merge from several databases, and the id is only known after the INSERT.

UUIDs

A UUID is a 128-bit random identifier such as 3f2b8c9e-.... You can generate it in the application before saving.

TypeScript
// TypeORM: generated by the database or driver
@PrimaryGeneratedColumn("uuid")
id!: string;
TypeScript
// Sequelize: generated in JavaScript
id: { type: DataTypes.UUID, defaultValue: DataTypes.UUIDV4, primaryKey: true }

Pros: unguessable, can be created offline, safe across databases. Cons: bigger than an integer and random (random inserts fragment B-tree indexes). Time-ordered variants (UUIDv7) keep inserts roughly sequential; check whether your database and driver support them.

Node.js Integer vs UUID

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with node:readline or process.stdin), it gets the next line from this box.

Composite primary keys

A table can be identified by two columns together, typical for link tables.

TypeScript
// TypeORM: mark both columns
@Entity()
export class Enrollment {
  @PrimaryColumn() studentId!: number;
  @PrimaryColumn() courseId!: number;
  @Column({ type: "date" }) enrolledOn!: string;
}
TypeScript
// Sequelize: primaryKey: true on both
studentId: { type: DataTypes.INTEGER, primaryKey: true },
courseId:  { type: DataTypes.INTEGER, primaryKey: true },

Many teams still prefer a single surrogate id plus a unique constraint on the pair, because single-column keys make relations and URLs simpler.

Node.js Composite key rejects duplicates

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with node:readline or process.stdin), it gets the next line from this box.

Natural vs surrogate keys

A natural key is real data (an email, an ISBN). A surrogate key is a meaningless id. Natural keys change (people change emails) and are often wrong in the real world. Default to a surrogate key and put a unique constraint on the natural identifier instead.

Reading the generated id

  • TypeORM: after await repo.save(entity) the same object now has its id.
  • Sequelize: const row = await Model.create(...) returns the instance with its id; with bulkCreate on some databases the ids are only filled when you pass returning: true.
  • Both copy values the database generated (defaults, ids) back onto the object on PostgreSQL via RETURNING; on MySQL, only the auto-increment id is available.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. What is an advantage of a UUID primary key over auto-increment?

    UUIDs can be generated anywhere and do not reveal how many rows exist.
  2. 2. Which TypeORM decorator creates an auto-increment key?

    PrimaryGeneratedColumn lets the database generate the value.
  3. 3. What happens when you insert a row that repeats an existing composite key?

    A primary key is unique; the second insert fails.
  4. 4. Why prefer a surrogate key plus a unique constraint on the email?

    A meaningless id stays stable when real-world data changes.

Summary

  • Auto-increment ids are small and simple but guessable and known only after insert.
  • UUIDs are unguessable and can be generated in code, at the cost of size and index locality.
  • Composite keys fit link tables; a surrogate id plus a unique pair is often easier.
  • Use surrogate keys and unique constraints for natural identifiers.
  • After saving, the generated id is copied back onto your object.

Next: Validation, Hooks and Timestamps