Regex for URLs

This pattern checks three things: a scheme of http or https followed by ://, a host made of word characters/dots/hyphens, and an optional path that can include a query string. It's deliberately narrow — good for spotting "does this look like an http(s) link" in a chunk of text, not for fully validating one.

Several perfectly valid URL features fall outside what it checks, each rejected rather than silently mishandled: a port (https://example.com:8080), embedded credentials (https://user:pass@example.com), and a fragment (https://example.com/page#section) are all real, common parts of a URL that this pattern doesn't account for. It's also case-sensitive on the scheme — HTTPS://example.com won't match without adding the i flag — and it only recognizes http/https, not ftp, mailto, or any other scheme.

If you actually need to parse or validate a URL — not just eyeball whether a string looks like one — use the real thing instead of a regex: URL Parser runs the browser's own URL parser and breaks a URL into every component correctly, including the port, credentials, and fragment this pattern skips, and URL Inspector adds a security-focused audit on top of that (embedded credentials, punycode hosts, open-redirect-shaped parameters). A regex is for finding candidate URLs in free text; a real parser is for trusting what you find.

For validating just the host portion on its own — without the scheme or path — see Regex for Domains.