setState vs replaceState in React.js

Solution 1:

With setState the current and previous states are merged. With replaceState, it throws out the current state, and replaces it with only what you provide. Usually setState is used unless you really need to remove keys for some reason; but setting them to false/null is usually a more explicit tactic.

While it's possible it could change; replaceState currently uses the object passed as the state, i.e. replaceState(x), and once it's set this.state === x. This is a little lighter than setState, so it could be used as an optimization if thousands of components are setting their states frequently.
_{I asserted this with this test case.}

If your current state is {a: 1}, and you call this.setState({b: 2}); when the state is applied, it will be {a: 1, b: 2}. If you called this.replaceState({b: 2}) your state would be {b: 2}.

Side note: State isn't set instantly, so don't do this.setState({b: 1}); console.log(this.state) when testing.

Solution 2:

Definition by example:

// let's say that this.state is {foo: 42}

this.setState({bar: 117})

// this.state is now {foo: 42, bar: 117}

this.setState({foo: 43})

// this.state is now {foo: 43, bar: 117}

this.replaceState({baz: "hello"})

// this.state. is now {baz: "hello"}

Take note of this from the docs, though:

setState() does not immediately mutate this.state but creates a pending state transition. Accessing this.state after calling this method can potentially return the existing value.

Same goes for replaceState()