Question

我正在构建一个可以将URL存储在数据库中的NodeJS应用程序。我想使用URL作为主键，以避免重复存储。为此，我需要尽可能采用最简单的网址，并删除多余的斜杠，参数和前缀。

如何将下面列出的所有URL转换为与列出的第一个URL相同的字符串？有办法安全地解决我可能未在下面列出的其他变化吗？

website.com/coolpage/938921

https://website.com/coolpage/938921/

https://www.website.com/coolpage/938921/

http://website.com/coolpage/938921/

https://website.com/coolpage/938921/

https://website.com/coolpage/938921/?awesome=1

https://website.com/coolpage/938921?awesome=1

https:///website.com//coolpage//938921//

Answer 1

使用标准的Node.js url模块。

解决方案：

require('url');

function getBaseUrl(url){
    const u = new URL(url);
    const result =`${u.host}${u.pathname}`
        .split('//').join('/')
        .replace('www.', '');

    // cut off the trailing '/' character from the result
    if (result.length && result[result.length - 1] === '/')
        return result.substring(0, result.length - 1)

    return result;
}

测试：

const urls = [
    "https://website.com/coolpage/938921/",
    "https://www.website.com/coolpage/938921/",
    "http://website.com/coolpage/938921/",
    "https://website.com/coolpage/938921/",
    "https://website.com/coolpage/938921/?awesome=1",
    "https://website.com/coolpage/938921?awesome=1",
    "https:///website.com//coolpage//938921//"
    ];

for (let i = 0; i < urls.length; i++) {
    const u = getBaseUrl(urls[i]);
    console.log(`${i}: ${u}`);
}

控制台输出：

0: website.com/coolpage/938921
1: website.com/coolpage/938921
2: website.com/coolpage/938921
3: website.com/coolpage/938921
4: website.com/coolpage/938921
5: website.com/coolpage/938921
6: website.com/coolpage/938921

Live example on runkit.com

Answer 2

在这里，您可以执行所需的功能：

function convertURL(url) {
	var urlParts = url.split('/')
	var finalURL = ''
	urlParts.forEach((p, i) => {
		if(finalURL.length == 0){
			if(p.includes('.com')){
				finalURL += p
			}
		}

		else if (p.length > 0 && i < urlParts.length - 1){
			finalURL += '/' + p
		}
	})

	return finalURL
}

var url = convertURL('https://website.com/coolpage/938921/?awesome=1')
console.log(url)

Answer 3

您可以将String.prototype.replace与RegExp \/+配合使用，以匹配用单个/和/替换的一个或多个正斜杠字符String.prototype.match() RegExp /[a-z0-9]+\.[a-z0-9]+(?=\/+)\/[a-z0-9]+(?=\/+)\/[a-z0-9]+/ig以匹配URL的主机名和路径名。

let urls = ["https://website.com/coolpage/938921/", "https://www.website.com/coolpage/938921/", "http://website.com/coolpage/938921/", "https://website.com/coolpage/938921/", "https://website.com/coolpage/938921/?awesome=1", "https://website.com/coolpage/938921?awesome=1", "https:///website.com//coolpage//938921//"];

let _URL = "website.com/coolpage/938921";

let replaceForwardSlashes = /\/+/g;

let matchHostAndPathNames = /[a-z0-9]+\.[a-z0-9]+(?=\/+)\/[a-z0-9]+(?=\/+)\/[a-z0-9]+/ig;

let matchedURLS = urls.map(url => url.replace(replaceForwardSlashes,'/').match(matchHostAndPathNames));

console.log(matchedURLS, new Set(...matchedURLS).size === 1, matchedURLS.every(u => u == _URL));

JS：将URL转换为其最简单的形式

3 个答案: