Skip to content
Warlock.js v4

Cascade

Standalone — usable in any Node project, no @warlock.js/core required.

@warlock.js/cascade is the data layer. Define models with decorators, declare relations between them, write migrations to evolve the schema, run queries through a fluent builder, do atomic updates safely under concurrency, search by vector embedding, paginate, and subscribe to model events.

Define a model once — the schema is your validator, your TypeScript type, and your table shape in a single declaration — and the class queries itself:

import { v, type Infer } from "@warlock.js/seal";
import { Model, RegisterModel } from "@warlock.js/cascade";


export const userSchema = v.object({
  name: v.string(),
  email: v.string().email(),
  status: v.enum(["active", "inactive"]).default("active"),
});


@RegisterModel()
export class User extends Model<Infer<typeof userSchema>> {
  public static table = "users";
  public static schema = userSchema;
}


// Write — validated against the schema, returns a hydrated instance
const user = await User.create({ name: "Ada Lovelace", email: "ada@example.com" });
user.id;             // the new ID Cascade assigned
user.get("status");  // "active" — the schema default applied


// Read it back — the model is the query entry point, no repository to import
const found = await User.find(user.id);
found?.get("email"); // "ada@example.com"

One declaration — no parallel type User = { ... }, no db.collection(...), no repository layer. The schema keeps your type and your table in lockstep, and the class is the query entry point.

Read by intent

Section titled “Read by intent”