Calling all gophers!

It’s time to talk about… Calls!

Calls

Given an expression f with a core type F of function type,

f(a1, a2, … an)

calls f with arguments a1, a2, … an.

This basic syntax is common to many languages, so should not feel at all strange if you have any programming experience at all.

So let’s dive right into the nitty-gritty details…

… Except for one special case, arguments must be single-valued expressions assignable to the parameter types of F and are evaluated before the function is called.

So we’ll talk about the special case later. For today, we’ll address the two other points made in this sentence.

1. Arguments must be single-valued expressions. Let’s consider some examples and counter-examples. Assuming a function f of type func(int, int, int):

f(1, 2, 3) // valid: Three single-valued expressions, 1, 2, and 3
f(1+2, 10-9, 3*4) // valid: Each expression evaluates to a single value

func one() int {
	return 1
}
f(one(), one(), one()) // valid: one() evaluates to a single expression

func two() (int, int) {
	return 2, 3
}

f(two(), one()) // invalid: two() evaluates to two expressions

// However, the  likely intent of the above can be rewritten as:
a, b := two()
f(a, b, one())

2. Arguments must be assignable to the parameter types of F. This one should be pretty obvious, especially if you’re familiar with strictly-typed languages.

f("one", float64(1.23), struct{ Name string}{Name: "Bob"}) // Invalid: f expects three ints

Quotes from The Go Programming Language Specification Version of August 2, 2023