Parse any URL into protocol, host, port, path, query and fragment using the WHATWG URL standard. Query parameters are decoded into a readable table.
Split a URL into every component the browser sees. Paste a URL and each part is extracted using the same WHATWG parser your browser uses, so the result matches exactly how the URL will actually be interpreted. Query parameters are decoded into a table.
Every absolute URL follows the same skeleton:
scheme://user:password@host:port/path?query#fragmentMost URLs use only a handful of those slots, but the grammar always allows all of them. Reading a URL is largely a matter of knowing where one component stops and the next begins, which is decided by the delimiters ://, @, :, /, ? and #.
Protocol (scheme) identifies how to fetch the resource — https:, mailto:, ftp:. Note that the WHATWG parser includes the trailing colon in this value, which surprises people comparing it against the string "https".
Username and password are the deprecated userinfo section. Browsers still parse them but strip them from requests and often warn, because embedding credentials in a URL leaks them into history, logs and referrer headers. Treat their presence in a URL you did not write as a phishing signal.
Hostname is the domain or IP address alone. Host is the hostname plus the port when a non-default port is present, which is why the two fields here often look identical.
Port is empty when the URL uses the scheme's default — 443 for https, 80 for http. The parser normalises this: example.com:443/ reports an empty port because 443 adds no information.
Path is everything from the first / up to the query. It is the only component that is never empty for an http(s) URL; a bare origin gets a path of /.
Query string starts at ? and carries the parameters. Fragment starts at # and is the only part never sent to the server — it is handled purely client-side, which is why single-page apps once used it for routing and why an access token in a fragment stays out of server logs.
Origin is the security-relevant triple of scheme, host and port. Two URLs share an origin only when all three match, and that is exactly what the same-origin policy and CORS decisions are based on.
The WHATWG URL parser only accepts absolute URLs. example.com/page has no scheme, so it cannot be parsed on its own — the parser has no way to know whether you meant example.com/page or a relative path. Real code resolves relative references against a base URL; this tool deliberately requires the absolute form so that what you see is unambiguous.
The parser does not hand back the raw string. It lowercases the scheme and host, resolves . and .. segments in the path, drops default ports, and converts internationalised domain names to their Punycode form, so münchen.de reports a hostname of xn--mnchen-3ya.de. This is a feature: it shows you the URL as the network stack will actually treat it, not as it was typed.
A query string may legitimately repeat a key, as in ?tag=a&tag=b. There is no specification saying what that means — some frameworks take the first value, some the last, some collect an array. The table here lists every occurrence in order so you can see exactly what was sent rather than a collapsed view.