URL Match Pattern

a JavaScript package handles match patterns described in browser extension

demo page

See articles on MDN and Chrome Developers.

Usage

/// check if a string is a valid URL match pattern
URLMatchPattern.test('*://github.com/'); // true

/// check if a string matches a URL match pattern
URLMatchPattern.test('*://*.google.com/*', 'https://google.com/search'); // true

/// get a RegExp object which represents a URL match pattern
const regexp = URLMatchPattern.toRegExp('*://*.name/');
regexp.test('https://domainA.name/'); // true
regexp.test('https://domainB.name/'); // true

/// or create an object with `test` method
const pattern = new URLMatchPattern('*://*.name/');
pattern.test('https://domainA.name/'); // true
pattern.test('https://domainB.name/'); // true

Note

Method test (both static and instance) allows url string:

• with port
• with hash
• without path

However, that's not the case after converting to RegExp.

URLMatchPattern.test('*://*/', 'http://username:password@abc:666/#hash=hash?a'); // true
URLMatchPattern.test('*://*/', 'http://abc'); // true

const pattern = new URLMatchPattern('*://*/');
pattern.test('http://username:password@abc:666/#hash=hash?a'); // true
pattern.test('http://abc'); // true

/// converting to RegExp object would lose that compatibility for current version
const regexp = URLMatchPattern.toRegExp('*://*/');
regexp.test('http://abc/'); // true
regexp.test('http://abc:80/'); // false
regexp.test('http://abc/#foo'); // false
regexp.test('http://abc'); // false

Compatibility

Either browser supports http, https, and ftp.

scheme | this project | Firefox 55+ | Chrome 91+ | Edge, Opera | Safari ----------------|--------------|-------------|------------|-------------|------ ws, wss | Yes | Yes | No | No | No urn | No | No | Yes | No | No data | No | partial | No | No | No file | Yes | Yes | Yes | Yes | No

Defects of URL Match Pattern

• not well-defined
• no escaping way for * in path
• * in path also matches / and ?, which usually have special meaning.