currying

Currying is a transformation in functional programming that converts a function that takes multiple arguments into a sequence of functions, each taking a single argument. In other words, a function f(a, b, c) can be viewed as f(a)(b)(c). The overall result remains the same, but intermediate results are themselves functions waiting for the next argument.

The term currying is named after Haskell Curry, a logician whose name became associated with the technique. The idea has roots in mathematical logic and lambda calculus, with earlier work by other thinkers, but currying as a programming concept is most commonly discussed in relation to functional languages and their handling of function application.

How it works in practice varies by language. In a curried view, a two-argument function is seen as a function that takes the first argument and returns a new function that takes the second. For example, a function add(x) that returns a function y -> x + y can be applied as add(2)(3) to produce 5. In a language like Haskell, functions are curried by default: a function of two arguments has a type a -> b -> c and can be partially applied. In other languages, currying may be achieved explicitly via closures or library helpers, or it may not be built in at all.

Relation to partial application is close but distinct. Partial application fixes some arguments of a function to produce a new function with fewer arguments; in languages with automatic currying, partial applications often align with currying semantics. Currying supports function composition and modular reuse by enabling higher-order construction of specialized functions from more general ones.