Dealing with changes in the shape of your application's data over time can be a challenging task. The conventional methods often result in a complicated mess, making it hard to ensure that your data is always up-to-date. This is where future-proof comes into play.
Here's how you can define migration steps using Future-proof:
// In the beginning
const initialState = { x: 0, y: 0 };
const { version, migrate } = from({ x: 0, y: 0 }).init(initialState);
As your data evolves, you can add migration steps:
// Later on
const initialState = { x: 0, y: 0, z: 0 };
const { version, migrate } = from({ x: 0, y: 0 })
.to((state) => ({ ...state, z: 0 }))
.init(initialState);
Each to function takes a callback that receives the current state object and returns a new state object with the desired changes.
Applying migrations is as simple as calling the migrate function with your data object and its version:
const data = migrate(
{
x: 200,
y: 200,
},
0
);
Future-proof has been designed with Zustand persisted stores in mind, making it a seamless integration:
import { create } from "zustand";
import { persist } from "zustand/middleware";
import { from } from "future-proof";
type State = { x: number; y: number; z: number; θ: number };
const initialState: State = {
x: 100,
y: 100,
z: 100,
θ: 0,
};
const { version, migrate } = from({
x: 100,
y: 100,
})
.to((data) => ({ ...data, z: 100 }))
.to((data) => ({ ...data, θ: 0 }))
.init(initialState);
const useStore = create<State>()(
persist((set) => initialState, {
name: "my-persisted-store",
version,
migrate,
})
);
By using Future-proof, you can confidently change the shape of your data as your app evolves, keeping your data migration logic clean and easy to read.