Home > Web Front-end > JS Tutorial > body text

Summary of knowledge about Vuejs technology stack

零下一度
Release: 2017-06-27 09:41:15
Original
2114 people have browsed it

Preface

The projects of my previous company mainly used jQuery and Angular1, and then I learned React. I didn't expect that when I came to this company, I suddenly started doing Vue, but Vue is still quite easy to get started. The following is some summary of the vue technology stack, all from the official website. It is mainly my own collection of the knowledge points of the vue technology stack. Therefore, this article is very watery. It is recommended to read my previous article Vuejs technology stack from CLI to packaging and online implementation. Full analysis

Vue

Independent build and runtime build

There are two build methods, independent build and runtime build. The difference between them is that the former includes a template compiler and the latter does not.

The responsibility of the template compiler is to compile the template string into a pure JavaScript rendering function. If you want to use the template option in a component, you need a compiler.

Life cycle

Look at the flow chart on the official website for details. Please note the difference between created and mounted. Created means that the vm instance has been created but not mounted, so some DOM operations should be placed in mounted. . There is no difference in whether the asynchronous request is placed in created or mounted. If you know the difference, please comment.

Computed attributes

Expressions within templates should not contain too much logic. For any complex logic, computed attributes should be used

Computed attributes and methods The difference is that computed properties are cached based on their dependencies.

Computed properties and computed properties, usually a better idea is to use computed properties instead of imperative watch callbacks. While computed properties are more appropriate in most cases, sometimes a custom watcher is needed. This is useful when you want to perform asynchronous or expensive operations in response to data changes.

Array update detection

The mutation method of the array (mutation method, which will change the original array called by these methods) will trigger a view update. There are seven following:

push()
pop()
shift()
unshift()
splice()
sort()
reverse()
Copy after login

When using the non-mutation method, you can replace the old array with the new array, or use the Vue.set method.

Object update

You can replace the old object with the new object, or use the Vue.set method

Vue.set(vm.someObject, 'foo', 'bar')

this.someObject = Object.assign({}, this.someObject, { a:1, b:2 })
Copy after login

Event handler

Vue.js is v- on provides event modifiers and key modifiers

Form control binding

You can use the v-model directive to create two-way data binding on form control elements. Common modifiers include .lazy, .number, and .trim.

You can also use the form input component of custom events.

Component

The API of Vue component comes from three parts: props, events and slots:

  • ##Props allows the external environment to pass data to the component

  • Events allow components to trigger side effects from the external environment

  • Slots allow external environments to combine additional content into the component.

1) The data attribute of the component must be a function

2) Parent-child component

In Vue.js, the relationship between parent-child components can be summarized as props down, events up. The parent component passes data down to the child component through props, and the child component sends messages to the parent component through events.

prop is one-way binding: when the properties of the parent component change, it will be transmitted to the child component, but not the other way around. This is to prevent child components from inadvertently modifying the parent component's state - which would make the application's data flow difficult to understand.

In addition, every time the parent component is updated, all props of the child component will be updated to the latest values. This means you should not change props inside child components. If you do this, Vue will give a warning in the console.

Why do we have the urge to modify the data in prop? Usually there are two reasons:

1. After the prop is passed in as the initial value, the subcomponent wants to use it as local data;

2.prop is passed in as the initial value. Processed by subcomponents into other data output.

The correct response to these two reasons is:

1. Define a local variable and initialize it with the value of prop:

props: ['initialCounter'],
data: function () {
  return { counter: this.initialCounter }
}
Copy after login
2. Define a calculation Property, handles the value of prop and returns it.

props: ['size'],
computed: {
  normalizedSize: function () {
    return this.size.trim().toLowerCase()
  }
}
Copy after login

Note that objects and arrays are reference types in JavaScript, pointing to the same memory space. If prop is an object or array, changing it inside the child component will affect the state of the parent component.

3) Non-parent-child components

Sometimes two components also need to communicate (non-parent-child relationship). In simple scenarios, an empty Vue instance can be used as the central event bus. In complex cases, we should consider using specialized state management patterns.

4).sync modifier

In some cases, we may need to perform "two-way binding" on a prop.

.sync was removed in 2.0, and Vue2.3.0+ added it back, but this time it only exists as a compile-time syntactic sugar, and it will be expanded into an automatically updated parent component property. v-on listener. The following code

<comp :foo.sync="bar"></comp>
Copy after login
will be expanded to:

<comp :foo="bar" @update:foo="val => bar = val"></comp>
Copy after login
When the subcomponent needs to update the value of foo, it needs to explicitly trigger an update event:

this.$emit('update:foo', newValue)
Copy after login
5) Using slots for content distribution

Scope slot: Receive prop objects passed from child components. A more representative use case for scope slots is list components, allowing components to customize how each item in the list should be rendered

6) Dynamic components, is attributes and keep-alive directives

7 )Subcomponent index

尽管有props和events,但是有时仍然需要JavaScript中直接访问子组件。为此可以使用ref为子组件指定一个索引ID。

异步更新队列

虽然 Vue.js 通常鼓励开发人员沿着“数据驱动”的方式思考,避免直接接触 DOM,但是有时我们确实要这么做。为了在数据变化之后等待 Vue 完成更新 DOM ,可以在数据变化之后立即使用 Vue.nextTick(callback) 。这样回调函数在 DOM 更新完成后就会调用。

过渡效果

Vue在插入、更新或者移除DOM时,提供多种不同方式的应用过渡效果。包括以下工具:

  • 在CSS过渡和动画中自动应用class

  • 可以配合使用第三方CSS动画库,如Animate.css

  • 在过渡钩子函数中使用JavaScript直接操作DOM

  • 可以配合使用第三方JavaScript动画库,如Velocity.js

1)单元素/组件的过渡

Vue提供了transition的封装组件,在下列情形中,可以给任何元素和组件添加过渡

  • 条件渲染(使用v-if)

  • 条件展示(使用v-show)

  • 动态组件

  • 组件根节点

2)多个元素的过渡

对于原生标签可以使用 v-if/v-else

3)多个组件的过渡

多个组件的过渡我们可以使用动态组件。

4)列表过渡

Render函数和JSX

自定义指令

和Angular的指令类似,主要操作DOM,下面是一个滚动加载的指令,holder暂时没想到什么更好的处理方法:

let scrollCallback = function (callback) {
  let windowH = window.innerHeight
  let getDocumentHeight = function () {
    var body = document.body
    var html = document.documentElement
    return Math.max(
      body.offsetHeight,
      body.scrollHeight,
      html.clientHeight,
      html.offsetHeight,
      html.scrollHeight
    )
  }
  let scrollH = document.documentElement.scrollTop || document.body.scrollTop
  if (windowH + scrollH >= getDocumentHeight() - (this.holder || 20)) {
    callback()
  }
}

let callBackWarpped

export default {
  bind (el, binding, vnode) {
    let holder
    if (vnode.data && vnode.data.attrs && vnode.data.attrs['scroll-placeholder']) {
      holder = parseInt(vnode.data.attrs['scroll-placeholder'])
    } else {
      holder = 20
    }
    callBackWarpped = scrollCallback.bind({el, holder}, binding.value)
    window.addEventListener('scroll', callBackWarpped, false)
  },

  unbind: function () {
    window.removeEventListener('scroll', callBackWarpped, false)
  }
}
Copy after login

混合

混合是一种灵活的分布式复用Vue组件的方式。混合对象可以包含任意组件选项。以组件使用混合对象时,所有混合对象的选项将被混入该组件本身的选项。

插件

1)创建插件

Vue.js的插件应当有一个公开方法install。这个方法的第一个参数是Vue构造器 , 第二个参数是一个可选的选项对象。

2)使用插件

通过全局方法Vue.use()使用插件:

// 调用 `MyPlugin.install(Vue)`
Vue.use(MyPlugin)
Copy after login

也可以传入一个选项对象:

Vue.use(MyPlugin, { someOption: true })
Copy after login

vue-router

两种导航方式

1)router-link声明式导航

<router-link to="/foo">Go to Foo</router-link>
Copy after login

router-link对应的路由匹配成功,将自动设置class属性值.router-link-active。

2)编程式导航

// 字符串
router.push('home')

// 对象
router.push({ path: 'home' })

// 命名的路由
router.push({ name: 'user', params: { userId: 123 }})

// 带查询参数,变成 /register?plan=private
router.push({ path: 'register', query: { plan: 'private' }})
Copy after login

重命名(redirect)和别名(alias)

两种路由模式

vue-router默认hash模式,也可以设置为路由的history模式。

导航钩子

vue-router提供的导航钩子主要用来拦截导航,让它完成跳转或取消。有多种方式可以在路由导航发生时执行钩子:全局的, 单个路由独享的, 或者组件级的。

路由meta

一个路由匹配到的所有路由记录会暴露为$route对象(还有在导航钩子中的route对象)的$route.matched数组。因此,我们需要遍历$route.matched来检查路由记录中的meta字段。

过渡动效

router-view是基本的动态组件,所以我们可以用transition组件给它添加一些过渡效果。

数据获取

有时候,进入某个路由后,需要从服务器获取数据。例如,在渲染用户信息时,你需要从服务器获取用户的数据。我们可以通过两种方式来实现:

导航完成之后获取:先完成导航,然后在接下来的组件生命周期钩子中获取数据。在数据获取期间显示『加载中』之类的指示。

导航完成之前获取:导航完成前,在路由的enter钩子中获取数据,在数据获取成功后执行导航。

滚动行为(scrollBehavior)

注意: 这个功能只在HTML5 history模式下可用。

路由懒加载

当打包构建应用时,Javascript包会变得非常大,影响页面加载。如果我们能把不同路由对应的组件分割成不同的代码块,然后当路由被访问的时候才加载对应组件,这样就更加高效了。

结合Vue的异步组件和Webpack的code splitting功能,轻松实现路由组件的懒加载。

const Foo = resolve => require(['./Foo.vue'], resolve)

const router = new VueRouter({
  routes: [
    { path: '/foo', component: Foo }
  ]
})
Copy after login

router-link

可以设置tag、append、active-class、exact等属性

有时候我们要让 "激活时的CSS类名" 应用在外层元素,而不是a标签本身,那么可以用router-link渲染外层元素,包裹着内层的原生a标签:

<router-link tag="li" to="/foo">
  <a>/foo</a>
</router-link>
Copy after login

在这种情况下,a将作为真实的链接(它会获得正确的 href 的),而"激活时的CSS类名"则设置到外层的li。

Router构造配置

routes、mode、base、linkActiveClass、scrollBehavior

对组件注入

1)注入的属性

通过在Vue根实例的router配置传入router实例,下面两个属性成员会被注入到每个子组件。

$router:router实例

$route:当前激活的路由信息对象。这个属性是只读的,里面的属性是immutable(不可变)的,不过你可以watch(监测变化)它。

2)允许的额外配置:beforeRouteEnter、beforeRouteLeave

vuex

state

1)单一状态树

Vuex使用单一状态树--是的,用一个对象就包含了全部的应用层级状态。至此它便作为一个『唯一数据源(SSOT)』而存在。这也意味着,每个应用将仅仅包含一个store实例。单一状态树让我们能够直接地定位任一特定的状态片段,在调试的过程中也能轻易地取得整个当前应用状态的快照。

2)在Vue组件中获得Vuex状态

最好在根实例中注册store选项,该store实例会注入到根组件下的所有子组件中,且子组件能通过this.$store访问到。而不是在每个需要使用state的组件中需要频繁地导入。

3)mapState辅助函数

当一个组件需要获取多个状态时候,可以使用mapState辅助函数帮助我们生成计算属性,这样可以简化代码书写。mapState函数会返回一个对象,然后可以使用对象展开运算符将它与局部计算属性混合使用。

4)不要滥用vuex

使用Vuex并不意味着你需要将所有的状态放入Vuex。虽然将所有的状态放到Vuex会使状态变化更显式和易调试,但也会使代码变得冗长和不直观。如果有些状态严格属于单个组件,最好还是作为组件的局部状态。你应该根据你的应用开发需要进行权衡和确定。

getters

getters用来从store中的state中派生出一些状态,例如对列表进行过滤并计数:

computed: {
  doneTodosCount () {
    return this.$store.state.todos.filter(todo => todo.done).length
  }
}
Copy after login

getters可以认为是store的计算属性。和state类似,有mapGetters辅助函数。

mutations

更改Vuex的store中的状态的唯一方法是提交mutation。Vuex中的mutations非常类似于事件:每个mutation都有一个字符串的事件类型(type)和一个回调函数(handler)。这个回调函数就是我们实际进行状态更改的地方,并且它会接受state作为第一个参数

1)提交载荷(Payload)

你可以向store.commit传入额外的参数,即mutation的载荷(payload):

// ...
mutations: {
  increment (state, n) {
    state.count += n
  }
}
store.commit('increment', 10)
Copy after login

2)Mutations需遵守Vue的响应规则

3)使用常量替代Mutation事件类型

4)mutation必须是同步函数

5)在组件中提交Mutations

你可以在组件中使用this.$store.commit('xxx')提交mutation,或者使用mapMutations辅助函数将组件中的methods映射为store.commit调用(需要在根节点注入store)。

actions

actions类似于mutation,不同在于:

  • actions提交的是mutation,而不是直接变更状态。

  • actions可以包含任意异步操作。

1)在组件中分发Action

你在组件中使用this.$store.dispatch('xxx')分发action,或者使用mapActions辅助函数将组件的methods映射为store.dispatch调用(需要先在根节点注入store)

2)组合Actions

Action通常是异步的,那么如何知道action什么时候结束呢?更重要的是,我们如何才能组合多个action,以处理更加复杂的异步流程?

首先,你需要明白store.dispatch可以处理被触发的action的回调函数返回的Promise,并且store.dispatch仍旧返回Promise。

使用async/await会更加简单:

// 假设 getData() 和 getOtherData() 返回的是 Promise

actions: {
  async actionA ({ commit }) {
    commit('gotData', await getData())
  },
  async actionB ({ dispatch, commit }) {
    await dispatch('actionA') // 等待 actionA 完成
    commit('gotOtherData', await getOtherData())
  }
}
Copy after login

Modules

由于使用单一状态树,应用的所有状态会集中到一个比较大的对象。当应用变得非常复杂时,store对象就有可能变得相当臃肿。

为了解决以上问题,Vuex允许我们将store分割成模块(module)。每个模块拥有自己的state、mutation、action、getter、甚至是嵌套子模块——从上至下进行同样方式的分割:

const moduleA = {
  state: { ... },
  mutations: { ... },
  actions: { ... },
  getters: { ... }
}

const moduleB = {
  state: { ... },
  mutations: { ... },
  actions: { ... }
}

const store = new Vuex.Store({
  modules: {
    a: moduleA,
    b: moduleB
  }
})

store.state.a // -> moduleA 的状态
store.state.b // -> moduleB 的状态
Copy after login

插件

Vuex的store接受plugins选项,这个选项暴露出每次mutation的钩子。Vuex插件就是一个函数,它接收store作为唯一参数。

严格模式

开启严格模式,仅需在创建store的时候传入strict:true

在严格模式下,无论何时发生了状态变更且不是由mutation函数引起的,将会抛出错误。这能保证所有的状态变更都能被调试工具跟踪到。

不要在发布环境下启用严格模式!严格模式会深度监测状态树来检测不合规的状态变更--请确保在发布环境下关闭严格模式,以避免性能损失。

表单处理

测试

热重载

The above is the detailed content of Summary of knowledge about Vuejs technology stack. For more information, please follow other related articles on the PHP Chinese website!

source:php.cn
Statement of this Website
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Popular Tutorials
More>
Latest Downloads
More>
Web Effects
Website Source Code
Website Materials
Front End Template