Quick access

Saved in this browser. Sign in to keep it on all your devices.

URL Parser

Parse and analyze URL components.

Quick Examples

Version 1.1 Updated What's new

Release notes

  1. v1.1 Latest
    • Added an in-depth guide below the tool, written from what the tool actually does.
    • Corrected guide text that overstated what the parser does.

About the URL Parser

The URL parser splits a web address into its parts, such as https, example.com, /path, and ?key=value, and decodes every query parameter.

How to use it

  1. 1 Paste a URL, or load one of the quick examples.
  2. 2 See each part listed separately.
  3. 3 Read the decoded query parameters in the table below.

Frequently asked questions

What is the difference between the path and the query string?

The path (/blog/post) identifies the page. The query string (?page=2&sort=new) passes extra parameters to it.

What is the # fragment?

The part after # points to a section within the page. Browsers do not send it to the server.

Why do some parameters look like %20?

They are URL-encoded. The parser shows the decoded value next to each parameter.

Why is the port blank when my URL has :443?

Port 443 is the default for https and port 80 is the default for http, so the URL standard drops them when a URL is parsed. The tool shows "(default)" in that case.

What happens if I leave out https://?

The tool adds https:// for you, so example.com/page and localhost:3000 both work. This also applies to addresses such as mailto:[email protected], which have no double slash and get misread as a web address.

Is the URL I paste sent anywhere?

No. Parsing happens in your browser with its built-in URL parser.

One address, taken apart

Load the Complex URL example and the tool splits this single string into its fields:

https://user:[email protected]:8080/path/subpath?foo=bar&baz=qux#section
  • Protocol is https:, the scheme that says how to connect. The colon is part of the value.
  • Hostname is example.com, the name looked up in DNS.
  • Port is 8080. Host combines the two as example.com:8080.
  • Origin is https://example.com:8080. Browsers use this combination of scheme, host and port as a security boundary, and a page on one origin is restricted from reading data on another.
  • Pathname is /path/subpath, the resource on that server.
  • Search is ?foo=bar&baz=qux, and the table underneath splits it into foo = bar and baz = qux.
  • Hash is #section. It stays in the browser and is never included in the request to the server, which is why server logs do not record it.

The user:pass part is a real piece of URL syntax called userinfo. The parser reads it, but this page has no field that displays it, so check the address itself if credentials might be embedded.

Your browser tidies a URL as it reads it

This page uses the URL parser built into your browser, which follows the WHATWG URL Standard rather than the older RFC 3986. The standard rewrites addresses into a normal form, so what comes out can differ from what you typed. Paste this:

HTTP://Example.COM:80/a/./b/../c

and the parsed address is:

http://example.com/a/c

Three things happened. The scheme and hostname were lowercased, because they are case-insensitive. The port disappeared since 80 is the default for http. The /./ and /../ segments were resolved, in the same way that folder paths are. The rest of the path keeps its case, because /Page and /page can be different resources on many servers.

Non-ASCII text is converted too. A hostname such as münchen.de becomes xn--mnchen-3ya.de, its Punycode form, and a path containing ß becomes %C3%9F, the percent-encoded UTF-8 bytes of that letter.

When a plus sign means a space

Spaces cannot appear in a URL, and there are two encodings for them. %20 works anywhere. A plus sign stands for a space only inside a query string, a convention inherited from HTML form submissions. In a path, + is a literal plus. For example, in https://example.com/a+b?c=d+e&f=g%2Bh:

  • The path stays /a+b, with its plus intact.
  • The parameter c decodes to "d e", with a space.
  • The parameter f decodes to "g+h", because %2B is how a real plus is written in a query.

This difference causes real bugs. A phone number such as +8801712345678 placed in a query without encoding arrives at the server with a space where the plus was. When a value looks wrong on arrival, paste the URL here and compare the decoded table with what you intended to send.

Spotting a disguised link

Parsing is a quick safety check for a link you do not trust, since everything before an @ sign in the authority is userinfo and not the destination. Look at this address:

https://[email protected]/login

At a glance it appears to start with google.com, but the Hostname field shows evil.example, and that is the server the browser would contact. Scammers rely on the same habit of reading from the left with long subdomains, such as paypal.com.account-check.example, whose registered domain is account-check.example. Read a hostname from the right, and treat the last two labels as a first approximation of who controls it. Country domains like .co.uk use three.

Auditing campaign links

Marketing links tend to collect parameters. Take https://example.com/search?q=shoes&utm_source=news&utm_medium=email#reviews. The table separates q, which the site itself uses, from utm_source and utm_medium, which only analytics tools read. A misspelled key such as utm_sorce is easy to miss in a long link and obvious when each key has its own row. To build tagged links from scratch, use the UTM Builder.

Limits of this page

  • If a key is repeated, as in ?q=1&q=2, the table may show it only once, even though the server receives both values. Read the Search field to see every occurrence.
  • The Full JSON panel and Copy All give the normalized URL as one quoted string, not a set of separate fields.
  • An address the browser cannot parse clears the results without an error message, so an empty panel means the URL is invalid.