base32

Base32 is a family of binary-to-text encoding schemes that represent binary data in ASCII strings using a 32-character alphabet. The most widely used variant is base32 as defined in RFC 4648. It encodes data by dividing the input bits into 5-bit groups; each group maps to one character from the alphabet A–Z and digits 2–7. Eight output characters correspond to 40 bits, i.e., five input bytes. When the final block contains fewer than 40 bits, padding with the equals sign '=' is added to form a complete set of eight characters. Some implementations omit padding, but standard decoding commonly relies on it.

Alphabet and variants: The canonical base32 alphabet is A–Z plus 2–7. RFC 4648 also defines base32hex, which uses a different alphabet (0–9 and A–V). Other variants, such as Crockford’s base32, emphasize human transcription and error-detection considerations. Some URL-safe or DNS-oriented schemes modify padding or alphabet to suit specific contexts.

Usage and characteristics: Base32 is used where ASCII text is required and case-insensitive handling or manual transcription is important. It provides better error tolerance for manual entry than base64 due to its distinct, non-lowercase alphabet and reduced similarity among characters. It is not a form of encryption; it is a reversible encoding. Compared with base64, base32 expands data more (longer strings) but offers advantages in readability and compatibility with systems that restrict character sets or case. Widely supported across programming languages and tools.