SPA应用性能优化（懒加载）
前提：
如今开发方式都是采用前后台分离的方式，前台采用的方式则是单页面应用开发简称SPA，这种开发模式最大的一个特点就是将有所代码打包成了一个文件，
这会导致了一个问题就是如果这个应用过大，打出来的这个文件也会越大，性能也就随之降低了，那么如何优化这个问题呢，今天就记录下懒加载策略吧。
那么什么是懒加载呢~
懒加载概念：
懒加载也就是按需加载，只要有需要才会加载的一种策略，常用于性能优化上（个人理解）通过懒加载可以避免加载不必要的资源。
工程级别：
在微服务架构里，前台也需要同步，这就需要每个模块都需要build出属于自己的js跟css，当跳转到目标模块时，我们不需要加载其他模块的文件，于是我们就
可以利用懒加载策略进行优化
利用第三方lazyload.js对目标模块进行文件读写
第三方lazyload文件
/*jslint browser: true, eqeqeq: true, bitwise: true, newcap: true, immed: true, regexp: false */ /**
LazyLoad makes it easy and painless to lazily load one or more external
JavaScript or CSS files on demand either during or after the rendering of a web
page. Supported browsers include Firefox 2+, IE6+, Safari 3+ (including Mobile
Safari), Google Chrome, and Opera 9+. Other browsers may or may not work and
are not officially supported. Visit https://github.com/rgrove/lazyload/ for more info. Copyright (c) 2011 Ryan Grove <ryan@wonko.com>
All rights reserved. Permission is hereby granted, free of charge, to any person obtaining a copy of
this software and associated documentation files (the 'Software'), to deal in
the Software without restriction, including without limitation the rights to
use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of
the Software, and to permit persons to whom the Software is furnished to do so,
subject to the following conditions: The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. @module lazyload
@class LazyLoad
@static
*/ LazyLoad = (function (doc) {
// -- Private Variables ------------------------------------------------------ // User agent and feature test information.
var env, // Reference to the <head> element (populated lazily).
head, // Requests currently in progress, if any.
pending = {}, // Number of times we've polled to check whether a pending stylesheet has
// finished loading. If this gets too high, we're probably stalled.
pollCount = 0, // Queued requests.
queue = { css: [], js: [] }, // Reference to the browser's list of stylesheets.
styleSheets = doc.styleSheets; // -- Private Methods -------------------------------------------------------- /**
Creates and returns an HTML element with the specified name and attributes. @method createNode
@param {String} name element name
@param {Object} attrs name/value mapping of element attributes
@return {HTMLElement}
@private
*/
function createNode(name, attrs) {
var node = doc.createElement(name), attr; for (attr in attrs) {
if (attrs.hasOwnProperty(attr)) {
node.setAttribute(attr, attrs[attr]);
}
} return node;
} /**
Called when the current pending resource of the specified type has finished
loading. Executes the associated callback (if any) and loads the next
resource in the queue. @method finish
@param {String} type resource type ('css' or 'js')
@private
*/
function finish(type) {
var p = pending[type],
callback,
urls; if (p) {
callback = p.callback;
urls = p.urls; urls.shift();
pollCount = 0; // If this is the last of the pending URLs, execute the callback and
// start the next request in the queue (if any).
if (!urls.length) {
callback && callback.call(p.context, p.obj);
pending[type] = null;
queue[type].length && load(type);
}
}
} /**
Populates the <code>env</code> variable with user agent and feature test
information. @method getEnv
@private
*/
function getEnv() {
var ua = navigator.userAgent; env = {
// True if this browser supports disabling async mode on dynamically
// created script nodes. See
// http://wiki.whatwg.org/wiki/Dynamic_Script_Execution_Order
async: doc.createElement('script').async === true
}; (env.webkit = /AppleWebKit\//.test(ua))
|| (env.ie = /MSIE|Trident/.test(ua))
|| (env.opera = /Opera/.test(ua))
|| (env.gecko = /Gecko\//.test(ua))
|| (env.unknown = true);
} /**
Loads the specified resources, or the next resource of the specified type
in the queue if no resources are specified. If a resource of the specified
type is already being loaded, the new request will be queued until the
first request has been finished. When an array of resource URLs is specified, those URLs will be loaded in
parallel if it is possible to do so while preserving execution order. All
browsers support parallel loading of CSS, but only Firefox and Opera
support parallel loading of scripts. In other browsers, scripts will be
queued and loaded one at a time to ensure correct execution order. @method load
@param {String} type resource type ('css' or 'js')
@param {String|Array} urls (optional) URL or array of URLs to load
@param {Function} callback (optional) callback function to execute when the
resource is loaded
@param {Object} obj (optional) object to pass to the callback function
@param {Object} context (optional) if provided, the callback function will
be executed in this object's context
@private
*/
function load(type, urls, callback, obj, context) {
var _finish = function () { finish(type); },
isCSS = type === 'css',
nodes = [],
i, len, node, p, pendingUrls, url; env || getEnv(); if (urls) {
// If urls is a string, wrap it in an array. Otherwise assume it's an
// array and create a copy of it so modifications won't be made to the
// original.
urls = typeof urls === 'string' ? [urls] : urls.concat(); // Create a request object for each URL. If multiple URLs are specified,
// the callback will only be executed after all URLs have been loaded.
//
// Sadly, Firefox and Opera are the only browsers capable of loading
// scripts in parallel while preserving execution order. In all other
// browsers, scripts must be loaded sequentially.
//
// All browsers respect CSS specificity based on the order of the link
// elements in the DOM, regardless of the order in which the stylesheets
// are actually downloaded.
if (isCSS || env.async || env.gecko || env.opera) {
// Load in parallel.
queue[type].push({
urls: urls,
callback: callback,
obj: obj,
context: context
});
} else {
// Load sequentially.
for (i = 0, len = urls.length; i < len; ++i) {
queue[type].push({
urls: [urls[i]],
callback: i === len - 1 ? callback : null, // callback is only added to the last URL
obj: obj,
context: context
});
}
}
} // If a previous load request of this type is currently in progress, we'll
// wait our turn. Otherwise, grab the next item in the queue.
if (pending[type] || !(p = pending[type] = queue[type].shift())) {
return;
} head || (head = doc.head || doc.getElementsByTagName('head')[0]);
pendingUrls = p.urls.concat(); for (i = 0, len = pendingUrls.length; i < len; ++i) {
url = pendingUrls[i]; if (isCSS) {
node = env.gecko ? createNode('style') : createNode('link', {
href: url,
rel: 'stylesheet'
});
} else {
node = createNode('script', { src: url });
node.async = false;
} node.className = 'lazyload';
node.setAttribute('charset', 'utf-8'); if (env.ie && !isCSS && 'onreadystatechange' in node && !('draggable' in node)) {
node.onreadystatechange = function () {
if (/loaded|complete/.test(node.readyState)) {
node.onreadystatechange = null;
_finish();
}
};
} else if (isCSS && (env.gecko || env.webkit)) {
// Gecko and WebKit don't support the onload event on link nodes.
if (env.webkit) {
// In WebKit, we can poll for changes to document.styleSheets to
// figure out when stylesheets have loaded.
p.urls[i] = node.href; // resolve relative URLs (or polling won't work)
pollWebKit();
} else {
// In Gecko, we can import the requested URL into a <style> node and
// poll for the existence of node.sheet.cssRules. Props to Zach
// Leatherman for calling my attention to this technique.
node.innerHTML = '@import "' + url + '";';
pollGecko(node);
}
} else {
node.onload = node.onerror = _finish;
} nodes.push(node);
} for (i = 0, len = nodes.length; i < len; ++i) {
head.appendChild(nodes[i]);
}
} /**
Begins polling to determine when the specified stylesheet has finished loading
in Gecko. Polling stops when all pending stylesheets have loaded or after 10
seconds (to prevent stalls). Thanks to Zach Leatherman for calling my attention to the @import-based
cross-domain technique used here, and to Oleg Slobodskoi for an earlier
same-domain implementation. See Zach's blog for more details:
http://www.zachleat.com/web/2010/07/29/load-css-dynamically/ @method pollGecko
@param {HTMLElement} node Style node to poll.
@private
*/
function pollGecko(node) {
var hasRules; try {
// We don't really need to store this value or ever refer to it again, but
// if we don't store it, Closure Compiler assumes the code is useless and
// removes it.
hasRules = !!node.sheet.cssRules;
} catch (ex) {
// An exception means the stylesheet is still loading.
pollCount += 1; if (pollCount < 200) {
setTimeout(function () { pollGecko(node); }, 50);
} else {
// We've been polling for 10 seconds and nothing's happened. Stop
// polling and finish the pending requests to avoid blocking further
// requests.
hasRules && finish('css');
} return;
} // If we get here, the stylesheet has loaded.
finish('css');
} /**
Begins polling to determine when pending stylesheets have finished loading
in WebKit. Polling stops when all pending stylesheets have loaded or after 10
seconds (to prevent stalls). @method pollWebKit
@private
*/
function pollWebKit() {
var css = pending.css, i; if (css) {
i = styleSheets.length; // Look for a stylesheet matching the pending URL.
while (--i >= 0) {
if (styleSheets[i].href === css.urls[0]) {
finish('css');
break;
}
} pollCount += 1; if (css) {
if (pollCount < 200) {
setTimeout(pollWebKit, 50);
} else {
// We've been polling for 10 seconds and nothing's happened, which may
// indicate that the stylesheet has been removed from the document
// before it had a chance to load. Stop polling and finish the pending
// request to prevent blocking further requests.
finish('css');
}
}
}
} return { /**
Requests the specified CSS URL or URLs and executes the specified
callback (if any) when they have finished loading. If an array of URLs is
specified, the stylesheets will be loaded in parallel and the callback
will be executed after all stylesheets have finished loading. @method css
@param {String|Array} urls CSS URL or array of CSS URLs to load
@param {Function} callback (optional) callback function to execute when
the specified stylesheets are loaded
@param {Object} obj (optional) object to pass to the callback function
@param {Object} context (optional) if provided, the callback function
will be executed in this object's context
@static
*/
css: function (urls, callback, obj, context) {
load('css', urls, callback, obj, context);
}, /**
Requests the specified JavaScript URL or URLs and executes the specified
callback (if any) when they have finished loading. If an array of URLs is
specified and the browser supports it, the scripts will be loaded in
parallel and the callback will be executed after all scripts have
finished loading. Currently, only Firefox and Opera support parallel loading of scripts while
preserving execution order. In other browsers, scripts will be
queued and loaded one at a time to ensure correct execution order. @method js
@param {String|Array} urls JS URL or array of JS URLs to load
@param {Function} callback (optional) callback function to execute when
the specified scripts are loaded
@param {Object} obj (optional) object to pass to the callback function
@param {Object} context (optional) if provided, the callback function
will be executed in this object's context
@static
*/
js: function (urls, callback, obj, context) {
load('js', urls, callback, obj, context);
} };
})(this.document);
封装路由
class LazyLoader extends React.Component {
constructor(props) {
super(props);
this.state = {
module: null
}
}
componentWillMount() {
this.load(this.props.module, this.props.path);
}
componentWillReceiveProps(newProps) {
this.load(newProps.module, newProps.path);
}
load(module, path) {
let cssPath = path + '',
jsPath = path + '';
LazyLoad.css(cssPath);
LazyLoad.js(jsPath, () => {
this.state.module = Modules[module];
})
}
render() {
if (!this.state.module) return null;
else this.props.children(this.state.module);
}
}
const HTMLModule = ({ name, path }) => {
let param = location.param;
location.param = null;
let ModuleComponent = Modules[name];
return (
<React.Fragment>
{
ModuleComponent ?
<ModuleComponent param={param} /> :
<LazyLoader path={path} module={name}>
{(Item) => <Item param={param} />}
</LazyLoader>
}
</React.Fragment>
);
};
class ModulesRouter extends React.Component {
constructor(props) {
super(props);
}
render() {
return (
<Switch>
<Route path={'/Module/Target'} render={() => <HTMLModule name='Target' path={'./Module/Target'} />} />
<Redirect from='*' to='/403' />
</Switch>
);
}
}
export default ModulesRouter;
路由与组件级别：
这里我们利用React16.6.X以后的版本新特性React.lazy跟React.Suspense进行路由以及组件部分的优化
路由设计
root路由
const RootRouter = () => {
return <HashRouter>
<React.Suspense fallback={<div>loading...</div>}>
<Switch>
<Route exact path="/" component={Home} />
<Route path="/oAuthPromisition" component={oAuthPromisition} />
<Route path="/WeChatPromisitionCheck" component={WeChatPromisitionCheck} />
<Route path="/403" component={Forbidden} />
<Route path="/sub" component={ModuleRouter} />
</Switch>
</React.Suspense>
</HashRouter>
}
子路由设计
import {
Route,
Switch
} from 'react-router-dom';
function lazyWithPreload(factory) {
const Component = React.lazy(factory);
Component.preload = factory;
return Component;
}
const LeaveMsg = lazyWithPreload(() => { return import(/* webpackChunkName: "leavemessage" */ '../JSX/Components/LeaveMessage/LeaveMessage') });
const Blog = lazyWithPreload(() => { return import(/* webpackChunkName: "blog" */ '../JSX/Components/Blog/Blog') });
const Article = lazyWithPreload(() => { return import(/* webpackChunkName: "article" */ '../JSX/Components/Blog/Article') });
const Production = lazyWithPreload(() => { return import(/* webpackChunkName: "production" */ '../JSX/Components/Production/index') });
const About = lazyWithPreload(() => { return import(/* webpackChunkName: "about" */ '../JSX/Components/About/About') });
const Write = lazyWithPreload(() => { return import(/* webpackChunkName: "write" */ '../JSX/Components/Blog/Write') });
const Overview = lazyWithPreload(() => { return import(/* webpackChunkName: "overview" */ '../JSX/Components/Overview/Overview') });
function moduleHtml(component) {
component.preload()
return component;
}
const SubRouter = () => {
return <Switch>
<Route path="/sub/leavemessage" component={moduleHtml(LeaveMsg)} />
<Route path="/sub/blog" component={moduleHtml(Blog)} />
<Route path="/sub/article" component={moduleHtml(Article)} />
<Route path="/sub/production" component={moduleHtml(Production)} />
<Route path="/sub/about" component={moduleHtml(About)} />
<Route path="/sub/write" component={moduleHtml(Write)} />
<Route path="/sub/overview" component={moduleHtml(Overview)} />
</Switch>
}
export default SubRouter;
路由设计总结：
利用import动态加载组件，以达到按需加载的目的，但是在动态加载的时候webpack会发生代码切割，所有我们同时也要进行webpack配置
首先配置分割规则
output: {
path: path.resolve(__dirname, "./public/"),
filename: '[name].bundle.js',
chunkFilename: '[name].module.js', //非入口文件命名规则
},
配置chunkFilename即非入口文件命名规则，就是这些由webpack分割的文件命名规范
由于webpack分割之后的路径查找方式是通过build输出路径查找的，因此如果我们单纯的将静态文件(index.html)当做根文件来加载，是找不到分割之后的代码路径的，
因此需要配置HtmlWebpackPlugin
new HtmlWebpackPlugin({
template: 'Index.html',
filename: 'index.html',
chunks: [
'index'
]
}),
然后将资源文件，以及第三方文件拷贝到build输出文件夹下
new CopyWebpackPlugin([
{
from: __dirname + '/Lib/editor',
to: __dirname + '/public/editor',
// toType: 'directory', //file 或者 dir 可选，默认是文件
// ignore: ['.*']
},{
from: __dirname + '/picSrc',
to: __dirname + '/public/picSrc',
}
]),
这样webpack配置修改完成了
组件级别跟路由级别基本类似，但是我们要考虑到，如果这么拆分代码，会产生很多文件，根据实际情况使用。
SPA应用性能优化（懒加载）的更多相关文章
- [转]JavaScript 的性能优化：加载和执行
原文链接:http://www.ibm.com/developerworks/cn/web/1308_caiys_jsload/index.html?ca=drs- JavaScript 的性能优化: ...
- 【转】js JavaScript 的性能优化：加载和执行
JavaScript 的性能优化:加载和执行 转自:https://www.ibm.com/developerworks/cn/web/1308_caiys_jsload/ 随着 Web2.0 技术的 ...
- vue性能优化1--懒加载
懒加载也叫延迟加载,即在需要的时候进行加载.随用随载.为什么需要懒加载?像vue这种单页面应用,如果没有应用懒加载,运用webpack打包后的文件将会异常的大,造成进入首页时,需要加载的内容过多,时间 ...
- JavaScript 的性能优化：加载和执行
随着 Web2.0 技术的不断推广,越来越多的应用使用 javascript 技术在客户端进行处理,从而使 JavaScript 在浏览器中的性能成为开发者所面临的最重要的可用性问题.而这个问题又因 ...
- JavaScript的性能优化：加载和执行
随着 Web2.0 技术的不断推广,越来越多的应用使用 javascript 技术在客户端进行处理,从而使 JavaScript 在浏览器中的性能成为开发者所面临的最重要的可用性问题.而这个问题又因 ...
- Android进阶：ListView性能优化异步加载图片 使滑动效果流畅
ListView 是一种可以显示一系列项目并能进行滚动显示的 View,每一行的Item可能包含复杂的结构,可能会从网络上获取icon等的一些图标信息,就现在的网络速度要想保持ListView运行的很 ...
- 【摘要】JavaScript 的性能优化：加载和执行
1.浏览器遇到js代码会暂停页面的下载和渲染,谁晓得js代码会不会把html给强奸(改变)了: 2.延迟脚本加载:defer 属性 <html> <head> <titl ...
- 前端性能优化_css加载会造成哪些阻塞现象？
css的加载是不会阻塞DOM的解析,但是会阻塞DOM的渲染,会阻塞link后面js语句的执行.这是由于浏览器为了防止html页面的重复渲染而降低性能,所以浏览器只会在加载的时候去解析dom树,然后等在 ...
- 移动端性能优化动态加载JS、CSS
JS CODE (function() { /** * update: * 1.0 */ var version = "insure 1.1.0"; var Zepto = Zep ...
- web前端如何性能优化提高加载速度
前端优化有以下几种途径: 一.减少HTTP请求数量和次数: 二.使用CDN: 三.添加Expires头: 四.压缩组件: 五.将样式表放在头部: 六.将脚本放在底部: 七.避免CSS表达式: 八.使用 ...
随机推荐
- Big Data（四）关于Hadoop的HA&CAP理论详解
问题 思路: 主从集群:结构相对简单,主与从协作 主:单点,数据一致好掌握 问题: 单点故障,集群整体不可用 压力过大,内存受限 解决方案 单点故障: 高可用方案:HA(High Available) ...
- Linux日常之命令sort
素材借鉴:https://www.cnblogs.com/51linux/archive/2012/05/23/2515299.html 命令sort 是Linux中常用的排序命令,属于管道命令. 常 ...
- c++判断当前登录账户是否在域环境内
#include <Windows.h> #include <DSRole.h> #pragma comment(lib, "netapi32.lib") ...
- VMware主机使用无线上网
VMware主机使用无线上网,默认的NAT连接在ubuntu下上不了网,需要把网络适配器改成桥接模式.
- Vue props双向绑定
props是不能改变的 项目里遇到一个问题,就是props的组件内是不能改变,就像react中的props值也是不能改变的, 所以我在用的时候竟然忽略了这个点.真的要反省反省, 下面就是这个报错的问题 ...
- Struts 2 实现Action的几种方式_java - JAVA
文章来源:嗨学网 敏而好学论坛www.piaodoo.com 欢迎大家相互学习 Action用于处理用户的请求,因此也被称为业务控制器.每个Action类就是一个工作单元,Struts 2框架负责将用 ...
- 【SpringBoot】spring-session-data-redis 解决集群环境下session共享
为什么会产生Session共享问题 集群情况下,session保存在各自的服务器的tomcat中,当分发地址至不同服务时,导致sesson取不到,就会产生session共享问题. 解决方案 负载均 ...
- 【leetcode】689. Maximum Sum of 3 Non-Overlapping Subarrays
题目如下: In a given array nums of positive integers, find three non-overlapping subarrays with maximum ...
- postman+newman+jenkins 接口自动化
一:Jenkins安装: jenkins是一款跨平台的持续集成和持续交付.基于Java开发的开源软件,提供任务构建.持续集成监控的功能,可以使开发测试人员更方便的构建软件项目, 我安装是安装程序模式, ...
- windows如何禁用惹人烦的开机启动广告
本地组策略编辑器 建立新的路径规则 重启电脑 本地组策略编辑器 你现在还在为那些烦人的互联网开机广告而发愁嘛,比如一下几种广告:这样的 还是这样的: 又或者是这样的: 修改了dns也并没有什么卵用,所 ...