SQL to Prisma Schema

About this SQL to Prisma schema converter

Paste MySQL or MariaDB table definitions and get a Prisma schema: one model per table, both sides of every foreign key written as relation fields, native database types where Prisma needs them, and a plain list of anything that does not convert. It reads the SQL itself (nothing is executed and nothing leaves your browser). This page is the Prisma entry point of the Database Schema Studio, so the imported schema is fully editable, and the same schema can also produce Laravel migrations.

How to use it

1. Paste CREATE TABLE statements or a structure-only dump, or open a .sql file, then Import.
2. Choose the Prisma version and the database the schema targets.
3. Read the diagnostics: they list what was skipped, approximated or renamed.
4. Copy schema.prisma (and prisma.config.ts for Prisma 7) or download them as a ZIP.

A worked example

This SQL:

CREATE TABLE `users` (
  `id` bigint unsigned NOT NULL AUTO_INCREMENT,
  `name` varchar(255) NOT NULL,
  `email` varchar(255) NOT NULL,
  `created_at` timestamp NULL DEFAULT NULL,
  `updated_at` timestamp NULL DEFAULT NULL,
  PRIMARY KEY (`id`),
  UNIQUE KEY `users_email_unique` (`email`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4;

CREATE TABLE `posts` (
  `id` bigint unsigned NOT NULL AUTO_INCREMENT,
  `user_id` bigint unsigned NOT NULL,
  `title` varchar(200) NOT NULL,
  `published` tinyint(1) NOT NULL DEFAULT '0',
  PRIMARY KEY (`id`),
  KEY `posts_user_id_foreign` (`user_id`),
  CONSTRAINT `posts_user_id_foreign` FOREIGN KEY (`user_id`) REFERENCES `users` (`id`) ON DELETE CASCADE
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4;

becomes these two models (Prisma 7, MySQL):

model User {
  id        BigInt    @id @default(autoincrement()) @db.UnsignedBigInt
  name      String    @db.VarChar(255)
  email     String    @unique(map: "users_email_unique") @db.VarChar(255)
  createdAt DateTime? @map("created_at") @db.Timestamp(0)
  updatedAt DateTime? @map("updated_at") @db.Timestamp(0)
  posts     Post[]

  @@map("users")
}

model Post {
  id        BigInt  @id @default(autoincrement()) @db.UnsignedBigInt
  userId    BigInt  @map("user_id") @db.UnsignedBigInt
  title     String  @db.VarChar(200)
  published Boolean @default(false)
  user      User    @relation(fields: [userId], references: [id], onDelete: Cascade, map: "posts_user_id_foreign")

  @@map("posts")
}

Note what changed: table names become singular PascalCase models with @@map; created_at becomes createdAt with @map; the foreign key becomes a user field on Post and a posts list on User (Prisma needs both); tinyint(1) is a Boolean; the constraint names from the SQL are kept through map; and the index that only supported the foreign key is not listed, because MySQL creates it from the key.

Type mapping (MySQL to Prisma)

A native attribute is written only when it differs from Prisma's default for the database (so varchar(191), MySQL's default String, stays a plain String). Other databases use their own natives: for PostgreSQL, uuid is @db.Uuid, jsonb is @db.JsonB and unsigned integers are ignored with a warning.

What Prisma cannot express

• Generated columns and CHECK constraints (dropped, with a warning).
• SetNull on required columns, and SetDefault on MySQL (the action is dropped).
• Indexes on JSON or Unsupported columns, and whole-column indexes on MySQL TEXT and BLOB (a 191-character prefix is written).
• Relations to non-unique columns or between columns of different types.