TupleHash

TupleHash is a cryptographic hash function designed to produce a fixed-length digest from a sequence of input elements, or tuples. It is part of the Keccak family of hash functions and is described in NIST SP 800-185 as a tuple-oriented hashing primitive. The function operates on a list of data elements rather than a single byte string, making it convenient for hashing structured data such as records or multi-field messages. The underlying primitive is a sponge function, typically instantiated with a fixed digest length chosen by the user.

Input and encoding: TupleHash takes as input an ordered sequence of elements, each element being an arbitrary byte string. Each element is encoded with a length field and a separator to ensure unambiguous delimitation between elements. The resulting encoded sequence is absorbed into the sponge, and a digest of the desired length is squeezed out. A domain separation prefix distinguishes TupleHash from other Keccak-based constructions, helping to prevent cross-function collisions with functions like SHAKE or ParallelHash.

Variants and usage: TupleHash supports different digest lengths by selecting the sponge’s output length, and it can be used with a variety of element sizes because of its length-prefix encoding. It is presented alongside ParallelHash in the SP 800-185 framework, where both functions serve as flexible, input-format-specific hashing options. TupleHash is especially suited for hashing composite data structures, lists of fields, or tuples in a way that preserves the boundaries between elements.

Security and notes: As with other sponge-based hashes, TupleHash relies on the security properties of the Keccak permutation and on correct encoding and domain separation. Proper implementation of the input encoding rules and parameter choices is essential to maintain collision and preimage resistance for the selected digest length.