my-name-is-url

Intelligently recognises many different url formats in a string. For the browser and node. Here, have a play.

About

my-name-is-url was created because I couldn't find a parser with a high enough success rate. The url spec is so vague that many strings could be a url, therefore matching the spec directly results in a lot of false positives. Most parsers get around this by requiring a url to contain a scheme to be matched as a url.

The regular expression used in my-name-is-url tries to match patterns likely to represent a url in a sentence rather than matching the actual url spec. This results in a much wider scope of matchable urls than most other parsers without introducing loads of false positives.

❗️Important note

If you're trying to parse a url into sections (scheme,host) or check a url is valid this module isn't for you. This module is intended to find urls in a string.

Install

npm install --save my-name-is-url

or

jspm install npm:my-name-is-url

Usage

Importing

CommonJS

var Urls = require('my-name-is-url');

ES6

import Urls from 'my-name-is-url';

Regex

If you just wanna do your own thing the regex used internally is helpfully exposed

const UrlRegex = Urls.regex;

Get Urls

The get() method returns an array of urls in a string

const text = 'Check out these sites: foobar.com,//foo.com,http://bar.com.';

Urls(text).get();
// [ 'foobar.com', '//foo.com', 'http://bar.com' ]

Filter Urls

The filter() method runs a filter on each url in a string

const text = 'My GitHub profile: https://github.com/lukechilds';

Urls(text).filter(url => `<a href="${url}">${url}</a>`);
// 'My GitHub profile: <a href="https://github.com/lukechilds">https://github.com/lukechilds</a>'

👍 Pro tip

You can get a parser instance by calling Urls() or new Urls, whichever you prefer.

License

MIT © Luke Childs