url

url re-implements Node.js’s core url module as a standalone npm package so that Browserify (and other bundlers that shim Node built-ins) can give browser code the exact same parse, format, and resolve API Node scripts use natively, without needing an actual Node runtime.

It depends on punycode and qs to replicate the same edge-case parsing behavior as Node’s built-in implementation, and is commonly pulled in transitively as a browser polyfill for the url built-in rather than depended on directly by application code.

What You Get

• A drop-in replacement for Node’s core url module, callable via require('url') in bundled browser code
• Legacy parse, format, and resolve functions matching Node’s original (pre-WHATWG) URL API
• Dependencies on punycode and qs to match Node’s IDNA host-name handling and query-string parsing
• A test suite (test/) validating parity with Node core’s own url module behavior

Common Use Cases

• Bundling legacy Node.js code with Browserify or webpack that calls require('url') and needs it to work in browsers
• Acting as the automatic polyfill target when a bundler’s Node-core shimming config resolves url imports
• Maintaining backward compatibility for libraries still using the pre-WHATWG url.parse/url.format API surface

Under The Hood

Architecture - the entire implementation lives in a single url.js file exporting parse, format, resolve, and the legacy Url constructor, mirroring the flat structure of Node’s own core url module rather than splitting into multiple internal modules. Tech Stack - plain JavaScript (CommonJS) with two runtime dependencies, punycode (for IDNA/host-name encoding) and qs (for query-string serialization), both chosen to replicate Node core’s exact parsing semantics. Code Quality - the package ships a test/ directory run via mocha/nyc for coverage and lints with ESLint (@ljharb/eslint-config) as a pretest step, plus a legacy zuul config for cross-browser test runs, but the project has had no commits since mid-2024 and shows zero recent releases, so it’s effectively frozen in maintenance mode. API Design - the API is intentionally a byte-for-byte match of Node’s older url.parse/url.format/url.resolve functions rather than the newer WHATWG URL class, so there’s no learning curve for anyone already familiar with Node’s original module, but it also means the package doesn’t expose the more modern, spec-compliant URL/URLSearchParams interface.