Redux Toolkit (RTK) 相比经典 Redux 提供了什么？

Question

Accepted Answer

**Redux Toolkit (RTK)** 是当今编写 Redux 的官方推荐方式。它消除了经典 Redux 臭名昭著的样板代码 — 手动 action types、action creators、不可变更新逻辑、store 设置 — 同时保留相同的核心原则。

## RTK 解决的样板代码问题

```js
// ❌ classic Redux — verbose: action types, creators, switch reducers, manual immutability
const INCREMENT = "counter/increment";
const increment = () => ({ type: INCREMENT });
function reducer(state = { value: 0 }, action) {
  switch (action.type) {
    case INCREMENT: return { ...state, value: state.value + 1 };
    default: return state;
  }
}
```

## RTK: createSlice 一应俱全

```js
import { createSlice, configureStore } from "@reduxjs/toolkit";

export const { increment, incrementBy } = counterSlice.actions; // action creators AUTO-generated
const store = configureStore({ reducer: { counter: counterSlice.reducer } });
```

`createSlice` 自动生成 action types 和 creators，并通过 **Immer** — 让你编写"可变"的 reducer 代码，该代码被安全地转换为不可变更新。一个简洁的 slice 替代了所有手动编写的部分。

## RTK 还包括什么

```text
configureStore   → store setup with DevTools + thunk middleware preconfigured
createAsyncThunk → standard async (pending/fulfilled/rejected actions) for data fetching
createEntityAdapter → normalized state + CRUD reducers for collections
RTK Query        → a full data-fetching/caching layer (like React Query, built into Redux)
```

## createAsyncThunk 示例

```js
export const fetchUsers = createAsyncThunk("users/fetch", () => api.getUsers());
// auto-dispatches users/fetch/pending | fulfilled | rejected — handle in extraReducers
```

## RTK 为何成为标准

```text
✓ ~75% less code than classic Redux
✓ Immer → no manual immutable spread logic (fewer bugs)
✓ Good defaults (DevTools, thunk) out of the box
✓ Encourages best practices (slices, normalized data)
✓ RTK Query bundles server-state management
```

## 为什么这很重要

经典 Redux 最大的批评就是其样板代码；RTK 直接解决了这个问题，这就是为什么它现在是**官方推荐**以及在现代代码库中"使用 Redux"的含义。

理解 `createSlice`（Immer 驱动、自动生成的 actions）、`createAsyncThunk`（异步）、`createEntityAdapter`（规范化）和 RTK Query（服务器状态）对当代 Redux 工作至关重要 — 它保留了 Redux 的可预测性，同时消除了驱使许多团队转向替代方案的繁仪。