URL Decoder / Encoder

Characters and their URL codes

A reserved character (# & + / : = ? @) keeps its meaning in an address, so when it is part of a value it must be encoded.

URL Decoder and Encoder Online: Percent-Encoding of Text and Addresses

Type or paste text on the left to get it URL-encoded on the right, or paste a string with %20 and other %XX codes on the right to read it as text. Both boxes follow each other as you type. Everything is converted in your browser: nothing you type is uploaded.

What is URL encoding

A URL can contain only a small set of characters: letters, digits and a few signs. Every other character, such as a space, a letter with an accent, a Cyrillic letter or an emoji, is written as one or more codes of the form %XX, where XX is the value of a byte in hex. This is called percent-encoding. A space is %20, the ampersand is %26, and the letter é is %C3%A9, which is the two bytes of its UTF-8 form. The table under the converter lists the characters you meet most often.

Which mode to use

• A value (like the JavaScript function encodeURIComponent): use it for one piece of an address, such as the value of a query parameter or a path segment. It encodes the characters that have a meaning in an address, such as / ? # & =, so they cannot break the address.
• A whole URL (like encodeURI): use it for a full address that contains spaces or non-English letters. It keeps : / ? # & = in place and keeps the %XX codes that are already in the text, so an address is not encoded twice.
• Form data (like PHP urlencode or the browser when it sends a form): a space becomes + instead of %20.

Plus or %20 for a space

In the query part of an address and in form data a plus sign means a space, but in a path a plus is just a plus. When you decode, the option "read + as a space" is on by default, which is right for query strings. Turn it off for a path or for any text where the plus is a real plus. A real plus in a value is always %2B.

UTF-8 and other character encodings

Today almost every address uses UTF-8: a Cyrillic letter is two bytes and an emoji is four. Older sites may use another encoding, and then the same letters have other codes: "Привет" in Windows-1251 is %CF%F0%E8%E2%E5%F2, in KOI8-R it is %F0%D2%C9%D7%C5%D4. If a decoded text is full of strange signs, choose the right character encoding. The converter reads and writes UTF-8, Windows-1251, KOI8-R and Windows-1252 (Latin-1).

Broken and double-encoded text

A % that is not followed by two hex digits cannot be decoded. The converter keeps it as it is, decodes the rest, and tells you where the problem is. If the result still contains codes like %41, the text was probably encoded twice (a % was encoded as %25); the "Decode again" button decodes it once more.

Limitations

• Only the characters that the chosen character encoding has can be encoded; others are reported.
• The tool changes the characters, not the structure: it does not check whether the address is valid. To take an address apart, use the URL parser; to find addresses in a text, use the URL extractor.
• Line breaks are encoded as %0A and %0D%0A, unless every line is processed on its own.