Redux is an amazing tool if you take the time to get to know it. One of the things about Redux that commonly trips people up is that reducers must be pure functions.
A pure function is a function which:
- Given same arguments, always returns the same result, and
- Has no side-effects (e.g., it won’t mutate its input arguments).
The problem is that sometimes a reducer needs to make a complicated change to some input state, but you can’t just mutate the state argument without causing bugs.
The solution is a handy tool called Immer . In this video, I’ll introduce you to Immer and show you how to use it to reduce the complexity of your reducer code. With one or two small reducers, the difference is pretty subtle, but on a large project, it can significantly simplify your application code.
Here’s an example. Imagine you’re building a social network, and you need to keep track of posts that a user has liked. When they like a post, you add a like object to the user’s likes collection. That might look something like this:
Notice what we’re returning from the like.type
case: Mixing bits of the payload into a nested property of the state object using the JavaScript object spread syntax: ...state
and ...state.likes
on lines 19–25.
After immer, you can simplify that part to a one-liner (line 21):
The produce
function returns a partial application (a function which has been partially applied to its arguments) which then takes the arguments of the reducer function. You pass it a callback function which takes a draft
of the state object instead of the real state object. You’re free to mutate that object as if it were any other mutable object in JavaScript. No more spreading nested properties to avoid mutating the input argument.
After your callback function runs, Immer compares the draft to the original state and then builds a new object with your changes applied, so your function feels like it’s mutating, but still behaves like a pure function. You get the best of both worlds: The simplicity of mutation with the benefits of immutability.
Next Steps
EricElliottJS.com has in-depth lessons on topics like pure functions, immutability, partial applications, and other functional and object oriented programming concepts.
以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持 码农网
猜你喜欢:本站部分资源来源于网络,本站转载出于传递更多信息之目的,版权归原作者或者来源机构所有,如转载稿涉及版权问题,请联系我们。
分布式机器学习:算法、理论与实践
刘铁岩、陈薇、王太峰、高飞 / 机械工业出版社 / 2018-10-20 / 89.00
人工智能和大数据时代,解决最有挑战性问题的主流方案是分布式机器学习!本书旨在全面介绍分布式机器学习的现状,深入分析其中的核心技术问题,并且讨论该领域未来的发展方向。 由微软亚洲研究院机器学习核心团队潜心力作!鄂维南院士、周志华教授倾心撰写推荐序! 本书旨在全面介绍分布式机器学习的现状,深入分析其中的核心技术问题,并且讨论该领域未来的发展方向。 全书共12章。第1章是绪论,向大家展......一起来看看 《分布式机器学习:算法、理论与实践》 这本书的介绍吧!
CSS 压缩/解压工具
在线压缩/解压 CSS 代码
RGB CMYK 转换工具
RGB CMYK 互转工具