What is partial application and how does it differ from currying?

Both techniques create specialized functions from general ones — but differ in how arguments are collected.

// Partial Application — pre-fill SOME args
function partial(fn, ...presetArgs) {
  return function(...laterArgs) {
    return fn(...presetArgs, ...laterArgs);
  };
}

const add = (a, b, c) => a + b + c;
const add10 = partial(add, 10);         // pre-fill first arg
const add10and20 = partial(add, 10, 20); // pre-fill two args

add10(5, 3);     // 18 — takes remaining 2 args AT ONCE
add10and20(7);   // 37 — takes remaining 1 arg

// Currying — always ONE arg at a time
const curriedAdd = a => b => c => a + b + c;
curriedAdd(1)(2)(3); // 6 — strictly one at a time

// Practical partial application with bind()
function greet(greeting, punct, name) {
  return `${greeting}, ${name}${punct}`;
}
const hello = greet.bind(null, 'Hello', '!'); // partial via bind
hello('Alice'); // 'Hello, Alice!'
hello('Bob');   // 'Hello, Bob!'

Summary:

• Currying: f(a, b, c) → f(a)(b)(c) — each call takes exactly ONE argument
• Partial application: f(a, b, c) → f(a, b)(c) — pre-fill any number of args