Account

What new tool do you want?

Developer Tools

URL Encode / Decode

Percent-encode text for URLs and query strings, or decode %20-style encoded URLs back into readable text.

How to use

  1. 1Choose Encode or Decode and the scope.
  2. 2Paste your text or URL and copy the result.

Frequently asked questions

What's the difference between encoding a component and a full URL?

Component mode encodes everything except letters, digits and - _ . ! ~ * ' ( ). Full-URL mode leaves characters like : / ? & = intact.

Should spaces be %20 or +?

%20 is correct everywhere. + means a space only in HTML form data (application/x-www-form-urlencoded); enable the option for that.

Why does decoding fail?

A % must be followed by two hex digits. A stray % (like '100%') makes the input invalid — encode it as %25.

A worked example showing why scope matters

Encode a b&c=d/é as a component and every special character gets escaped: a%20b%26c%3Dd%2F%C3%A9 — the & and = are encoded because encodeURIComponent treats them as literal characters to protect, not query-string syntax. Encode the full URL https://x.com/a b?q=1&r=é in "Full URL" scope instead, and you get https://x.com/a%20b?q=1&r=%C3%A9 — the : / ? & = characters that make the URL structurally valid are left alone, while the space and the accented é still get escaped. Using component encoding on a full URL would wrongly escape its own ? and & and break it; the two modes exist because encoding "a value going into a URL" and "a URL as a whole" are genuinely different operations.

Why %20 and + aren't interchangeable

Percent-encoding is a general mechanism (RFC 3986) where %20 always means a literal space, valid anywhere in a URL. The + for space convention is specific to one encoding format — application/x-www-form-urlencoded, which is what browsers use for classic HTML form submissions — and a parser that isn't expecting that format will take a literal + character at face value. The toggle exists specifically for reproducing that form-encoding behavior; leave it off unless you know you're building a string destined for form-encoded data.

Why a stray % breaks decoding outright

Percent-encoding's rule is strict: a % must always be followed by exactly two hexadecimal digits. The string "100%" fails decoding here precisely because that trailing % isn't followed by two hex digits — it's not percent-encoding syntax at all, just a literal percent sign that itself needed to be encoded as %25 to survive the round trip. This is a common real bug source: any text containing a raw % (prices, percentages) needs to be encoded before it goes into a URL, or decoding it later throws exactly this error.

The scope menu hides while I am decoding

Mode opens on encode. Scope opens on query value, and that menu is hidden as soon as I switch to decode. Spaces as + starts off. While encoding, turning it on replaces each encoded space with a plus, so a space becomes a plus rather than %20, while a percent sign I typed is still encoded and is not rewritten into a plus. While decoding, the same switch turns plus signs into spaces before the percent sequences are read, so a plus that was meant as a plus becomes a space. A percent sign that is not followed by two hex digits stops the whole decode. The message says every % must be followed by two hex digits, and I do not get a partial result.

Related tools

More: Developer Tools