Convert a Simple SQL Table to a Prisma Model

From CREATE TABLE to Prisma Model

The most fundamental conversion takes a SQL CREATE TABLE statement and produces a Prisma model block. Each column becomes a field with a Prisma scalar type, and constraints like PRIMARY KEY are mapped to Prisma attributes.

Example SQL

CREATE TABLE users (
  id INTEGER PRIMARY KEY,
  name VARCHAR(255) NOT NULL,
  email VARCHAR(320) NOT NULL,
  age INTEGER,
  is_active BOOLEAN NOT NULL DEFAULT true
);

Generated Prisma Schema

model User {
  id       Int     @id
  name     String
  email    String
  age      Int?
  isActive Boolean @default(true)

  @@map("users")
}

Key Mapping Rules

SQL element	Prisma equivalent
Table name	model name (PascalCase) + @@map("original_name")
Column name	Field name (camelCase) + @map("original_name") if different
NOT NULL	Field is required (default in Prisma)
Nullable column	Append ? to the type
PRIMARY KEY	@id attribute
DEFAULT	@default(...) attribute

Table and Column Naming

Prisma conventions use PascalCase for model names and camelCase for field names. When the SQL table uses snake_case, the converter generates a @@map("table_name") directive on the model and @map("column_name") directives on fields whose names differ after camelCase conversion.

Nullable vs Required Fields

In SQL, columns are nullable by default unless marked NOT NULL. Prisma inverts this: fields are required by default. The converter marks nullable SQL columns with ? in Prisma and leaves NOT NULL columns as plain required fields.

Default Values

SQL DEFAULT clauses are translated to Prisma @default() attributes. Common defaults like true, false, numeric literals, and NOW() map directly. More complex SQL expressions may be preserved as @default(dbgenerated("...")).

This basic table-to-model conversion is the foundation for all other patterns. Once you understand this mapping, you can handle relations, indexes, enums, and advanced features.