How does Vue3 implement global exception handling? The following article will introduce to you the method of global exception handling in Vue3. I hope it will be helpful to you!
When developing component libraries or plug-ins, global exception handling is often required to achieve:
So how to implement the above function? This article first briefly implements an exception handling method, then introduces it in detail based on the implementation in the Vue3 source code, and finally summarizes several core elements for implementing exception handling. [Related recommendations: vuejs video tutorial]
The Vue3 version of this article is 3.0.11
For the front end, there are many common exceptions, such as:
The most commonly used examples are:
Through the window.onerror
document, it can be seen that when an error occurs when JS is running (including syntax errors), window.onerror()
:
window.onerror = function(message, source, lineno, colno, error) { console.log('捕获到异常:',{message, source, lineno, colno, error}); }
Function parameters:
event
in HTML onerror=""
handler. If this function returns true, the execution of the default event handling function is prevented.
In addition, we often use try...catch
statement to handle exceptions:
try { // do something } catch (error) { console.error(error); }
For more processing methods, you can read the previously recommended articles.
You can think about it, do you often have to deal with these errors during the business development process?
So does a complex library like Vue3 use try...catch
to handle exceptions everywhere?
Let’s take a look next.
When developing plug-ins or libraries, we can encapsulate a global exception handling method through try...catch
, which will require The executed method is passed in as a parameter, and the caller only needs to care about the call result without knowing the internal logic of the global exception handling method.
The general usage is as follows:
const errorHandling = (fn, args) => { let result; try{ result = args ? fn(...args) : fn(); } catch (error){ console.error(error) } return result; }
Test it:
const f1 = () => { console.log('[f1 running]') throw new Error('[f1 error!]') } errorHandling(f1); /* 输出: [f1 running] Error: [f1 error!] at f1 (/Users/wangpingan/leo/www/node/www/a.js:14:11) at errorHandling (/Users/wangpingan/leo/www/node/www/a.js:4:39) at Object.<anonymous> (/Users/wangpingan/leo/www/node/www/a.js:17:1) at Module._compile (node:internal/modules/cjs/loader:1095:14) at Object.Module._extensions..js (node:internal/modules/cjs/loader:1147:10) at Module.load (node:internal/modules/cjs/loader:975:32) at Function.Module._load (node:internal/modules/cjs/loader:822:12) at Function.executeUserEntryPoint [as runMain] (node:internal/modules/run_main:81:12) at node:internal/main/run_main_module:17:47 */
As you can see, when you need to handle exceptions for a method, you only need to pass the method in as a parameter. However, the above example is quite different from the logic of actual business development. In actual business, we often encounter nested calls of methods, so let’s try:
const f1 = () => { console.log('[f1]') f2(); } const f2 = () => { console.log('[f2]') f3(); } const f3 = () => { console.log('[f3]') throw new Error('[f3 error!]') } errorHandling(f1) /* 输出: [f1 running] [f2 running] [f3 running] Error: [f3 error!] at f3 (/Users/wangpingan/leo/www/node/www/a.js:24:11) at f2 (/Users/wangpingan/leo/www/node/www/a.js:19:5) at f1 (/Users/wangpingan/leo/www/node/www/a.js:14:5) at errorHandling (/Users/wangpingan/leo/www/node/www/a.js:4:39) at Object.<anonymous> (/Users/wangpingan/leo/www/node/www/a.js:27:1) at Module._compile (node:internal/modules/cjs/loader:1095:14) at Object.Module._extensions..js (node:internal/modules/cjs/loader:1147:10) at Module.load (node:internal/modules/cjs/loader:975:32) at Function.Module._load (node:internal/modules/cjs/loader:822:12) at Function.executeUserEntryPoint [as runMain] (node:internal/modules/run_main:81:12) */
This is no problem. Then the next step is to implement the corresponding exception handling in the catch
branch of the errorHandling
method.
Next, let’s see how it is handled in the Vue3 source code?
After understanding the above example, let’s take a look at how to implement exception handling in the Vue3 source code. It is also very simple to implement.
Two processes, callWithErrorHandling
and callWithAsyncErrorHandling
, are defined in the errorHandling.ts
file Global exception method.
As the name suggests, these two methods handle respectively:
callWithErrorHandling
: handle exceptions of synchronous methods; callWithAsyncErrorHandling
: handle exceptions of asynchronous methods abnormal. The usage method is as follows:
callWithAsyncErrorHandling( handler, instance, ErrorCodes.COMPONENT_EVENT_HANDLER, args )
The code implementation is roughly as follows:
// packages/runtime-core/src/errorHandling.ts // 处理同步方法的异常 export function callWithErrorHandling( fn: Function, instance: ComponentInternalInstance | null, type: ErrorTypes, args?: unknown[] ) { let res try { res = args ? fn(...args) : fn(); // 调用原方法 } catch (err) { handleError(err, instance, type) } return res } // 处理异步方法的异常 export function callWithAsyncErrorHandling( fn: Function | Function[], instance: ComponentInternalInstance | null, type: ErrorTypes, args?: unknown[] ): any[] { // 省略其他代码 const res = callWithErrorHandling(fn, instance, type, args) if (res && isPromise(res)) { res.catch(err => { handleError(err, instance, type) }) } // 省略其他代码 }
callWithErrorHandling
The logic of the method processing is relatively simple, through a simpletry...catch
Create a layer of encapsulation.
The callWithAsyncErrorHandling
method is more clever, by passing the method that needs to be executed into the callWithErrorHandling
method, and processing the result through the .catch
method.
In the above code, if an error is reported, the exception will be handled through handleError()
. The implementation is roughly as follows:
// packages/runtime-core/src/errorHandling.ts // 异常处理方法 export function handleError( err: unknown, instance: ComponentInternalInstance | null, type: ErrorTypes, throwInDev = true ) { // 省略其他代码 logError(err, type, contextVNode, throwInDev) } function logError( err: unknown, type: ErrorTypes, contextVNode: VNode | null, throwInDev = true ) { // 省略其他代码 console.error(err) }
After retaining the core processing logic, you can see that the processing here is also quite simple, and the error content is output directly through console.error(err)
.
在使用 Vue3 时,也支持指定自定义异常处理函数,来处理组件渲染函数和侦听器执行期间抛出的未捕获错误。这个处理函数被调用时,可获取错误信息和相应的应用实例。
文档参考:《errorHandler》
使用方法如下,在项目 main.js
文件中配置:
// src/main.js app.config.errorHandler = (err, vm, info) => { // 处理错误 // `info` 是 Vue 特定的错误信息,比如错误所在的生命周期钩子 }
那么 errorHandler()
是何时执行的呢?我们继续看看源码中 handleError()
的内容,可以发现:
// packages/runtime-core/src/errorHandling.ts export function handleError( err: unknown, instance: ComponentInternalInstance | null, type: ErrorTypes, throwInDev = true ) { const contextVNode = instance ? instance.vnode : null if (instance) { // 省略其他代码 // 读取 errorHandler 配置项 const appErrorHandler = instance.appContext.config.errorHandler if (appErrorHandler) { callWithErrorHandling( appErrorHandler, null, ErrorCodes.APP_ERROR_HANDLER, [err, exposedInstance, errorInfo] ) return } } logError(err, type, contextVNode, throwInDev) }
通过 instance.appContext.config.errorHandler
取到全局配置的自定义错误处理函数,存在时则执行,当然,这边也是通过前面定义的 callWithErrorHandling
来调用。
在使用 Vue3 的时候,也可以通过 errorCaptured
生命周期钩子来捕获来自后代组件的错误。
文档参考:《errorCaptured》
入参如下:
(err: Error, instance: Component, info: string) => ?boolean
此钩子会收到三个参数:错误对象、发生错误的组件实例以及一个包含错误来源信息的字符串。
此钩子可以返回 false
以阻止该错误继续向上传播。有兴趣的同学可以通过文档,查看具体的错误传播规则。
使用方法如下,父组件监听 onErrorCaptured
生命周期(示例代码使用 Vue3 setup 语法):
<template> <Message></Message> </template> <script setup> // App.vue import { onErrorCaptured } from 'vue'; import Message from './components/Message.vue' onErrorCaptured(function(err, instance, info){ console.log('[errorCaptured]', err, instance, info) }) </script>
子组件如下:
<template> <button @click="sendMessage">发送消息</button> </template> <script setup> // Message.vue const sendMessage = () => { throw new Error('[test onErrorCaptured]') } </script>
当点击「发送消息」按钮,控制台便输出错误:
[errorCaptured] Error: [test onErrorCaptured] at Proxy.sendMessage (Message.vue:36:15) at _createElementVNode.onClick._cache.<computed>._cache.<computed> (Message.vue:3:39) at callWithErrorHandling (runtime-core.esm-bundler.js:6706:22) at callWithAsyncErrorHandling (runtime-core.esm-bundler.js:6715:21) at HTMLButtonElement.invoker (runtime-dom.esm-bundler.js:350:13) Proxy {sendMessage: ƒ, …} native event handler
可以看到 onErrorCaptured
生命周期钩子正常执行,并输出子组件 Message.vue
内的异常。
那么这个又是如何实现呢?还是看 errorHandling.ts
中的 handleError()
方法:
// packages/runtime-core/src/errorHandling.ts export function handleError( err: unknown, instance: ComponentInternalInstance | null, type: ErrorTypes, throwInDev = true ) { const contextVNode = instance ? instance.vnode : null if (instance) { let cur = instance.parent // the exposed instance is the render proxy to keep it consistent with 2.x const exposedInstance = instance.proxy // in production the hook receives only the error code const errorInfo = __DEV__ ? ErrorTypeStrings[type] : type while (cur) { const errorCapturedHooks = cur.ec // ①取出组件配置的 errorCaptured 生命周期方法 if (errorCapturedHooks) { // ②循环执行 errorCaptured 中的每个 Hook for (let i = 0; i < errorCapturedHooks.length; i++) { if ( errorCapturedHooks[i](err, exposedInstance, errorInfo) === false ) { return } } } cur = cur.parent } // 省略其他代码 } logError(err, type, contextVNode, throwInDev) }
这边会先获取 instance.parent
作为当前处理的组件实例进行递归,每次将取出组件配置的 errorCaptured
生命周期方法的数组并循环调用其每一个钩子,然后再取出当前组件的父组件作为参数,最后继续递归调用下去。
Vue3 还为异常定义了错误码和错误信息,在不同的错误情况有不同的错误码和错误信息,让我们能很方便定位到发生异常的地方。 错误码和错误信息如下:
// packages/runtime-core/src/errorHandling.ts export const enum ErrorCodes { SETUP_FUNCTION, RENDER_FUNCTION, WATCH_GETTER, WATCH_CALLBACK, // ... 省略其他 } export const ErrorTypeStrings: Record<number | string, string> = { // 省略其他 [LifecycleHooks.RENDER_TRACKED]: 'renderTracked hook', [LifecycleHooks.RENDER_TRIGGERED]: 'renderTriggered hook', [ErrorCodes.SETUP_FUNCTION]: 'setup function', [ErrorCodes.RENDER_FUNCTION]: 'render function', // 省略其他 [ErrorCodes.SCHEDULER]: 'scheduler flush. This is likely a Vue internals bug. ' + 'Please open an issue at https://new-issue.vuejs.org/?repo=vuejs/vue-next' }
当不同错误情况,根据错误码 ErrorCodes
来获取 ErrorTypeStrings
错误信息进行提示:
// packages/runtime-core/src/errorHandling.ts function logError( err: unknown, type: ErrorTypes, contextVNode: VNode | null, throwInDev = true ) { if (__DEV__) { const info = ErrorTypeStrings[type] warn(`Unhandled error${info ? ` during execution of ${info}` : ``}`) // 省略其他 } else { console.error(err) } }
关于 Vue3 实现 Tree Shaking 的介绍,可以看我之前写的高效实现框架和 JS 库瘦身。
其中,logError
方法中就使用到了:
// packages/runtime-core/src/errorHandling.ts function logError( err: unknown, type: ErrorTypes, contextVNode: VNode | null, throwInDev = true ) { if (__DEV__) { // 省略其他 } else { console.error(err) } }
当编译成 production 环境后,__DEV__
分支的代码不会被打包进去,从而优化包的体积。
到上面一部分,我们就差不多搞清楚 Vue3 中全局异常处理的核心逻辑了。我们在开发自己的错误处理方法时,也可以考虑这几个核心点:
支持同步和异步的异常处理;
设置业务错误码、业务错误信息;
支持自定义错误处理方法;
支持开发环境错误提示;
支持 Tree Shaking。
原文地址:https://juejin.cn/post/7071982812668100616
The above is the detailed content of An article explaining in detail how Vue3 handles global exceptions. For more information, please follow other related articles on the PHP Chinese website!