percentdecoding

Percent-decoding, often referred to as percent-decoding or URL decoding, is the process of converting percent-encoded sequences back into their original characters. It is the inverse operation of percent-encoding (URL encoding), which represents arbitrary data in URLs, query strings, and form submissions to ensure safe transport over text-based protocols.

The decoding works by scanning the input for sequences of the form %xx, where xx consists of two hexadecimal digits. Each such sequence represents a single byte, which is then assembled into a byte sequence and interpreted using a character encoding, most commonly UTF-8. After decoding, the resulting characters are returned to the user or used by the program. In many contexts, the decoding step is followed by a specific character encoding interpretation.

Examples include %20 producing a space, %7B producing a left curly brace, and %C3%A9 producing the character é in UTF-8. In web form data encoded as application/x-www-form-urlencoded, plus signs may be interpreted as spaces after the percent-decoding step, but the plus sign itself is not a percent-encoded character.

Edge cases include incomplete sequences like % and %2, or invalid hexadecimal digits such as %GZ, which can raise errors or be left unchanged depending on the implementation. Decoding multi-byte sequences requires correct UTF-8 interpretation; invalid byte sequences can trigger errors or yield replacement characters.

Common implementations include Python’s urllib.parse.unquote, JavaScript’s decodeURIComponent, and Java’s URLDecoder. Proper use involves decoding to the appropriate text encoding and avoiding double decoding.

See also: URL encoding, percent-encoding, form encoding.