Module 2, Function Types. Topic 1.1, First-class Values. So, we've been talking about functions in this class already. You can't really program in Go without functions. So you've always gotta have at least the main function. And we've had that in all the examples that we've used. But what we're going to talk about now is the ways in which Go allows you to treat a function as a first-class value. So there's this term first-class value, treating the functions as first-class values and it's associated with functional programming. Functional programming is a way of programming, a programming paradigm let's say, that is different than what people usually use, let's say. So if you've ever programmed in scheme or something like that, that's a functional programming language, ML, there are a bunch of these languages. And so it's a different way of thinking, and we're not going to really do a lot of functional programming. But there are some of features of functional programming that can be very useful. So implement some of these features, and we'll talk about how to use those. So treating a function as a first-class value, generally means being able to treat a function like any other type, like an integer, a string, a float. All the things that you can do to one of those types, you can also do to a function. So as an example, variables can be declared to be a function type. So you can always declare a variable to be an integer type, a floating point type, things like that. You can also declare it to be a function type. And you can then set a variable equal to a function. You can create them dynamically, so in your code, you can create an integer on a fly, right? In the middle of your code, you can create new variable, X or whatever, and X equal three, whatever. And that creates an integer on a fly at run time while you're running the code. So that's creating a variable dynamically, creating some type dynamically. You can do that with functions, too. Now up until this point, we haven't done that. We create them statically, just at the top level you say func main or something like that. But we want to be able to create them dynamically, too. Sometimes it's very useful. So inside a function, you can create a new function, right inside. Functions can be passed as arguments to other functions and returned as values from other functions. So just like an integer, a flow, to something like that. You can pass it as an argument. You can pass a function as an argument to a function, or a function can return a new function as its return value. Also it can be stored in a data structure. So you can have a data structure, maybe a slice. You can have a slice of integers, something like that. You can also have a slice of floats. So in all these ways, Go lang is going to allow us to treat a function just like a regular type. So let's talk about how to implement these things in Go. Start off with variables as function. So one thing you'd like to be able to do sometimes is declare a variable as a function. So the variable basically acts as an alias, another name for that function in this case, we do this the way we're doing right here. So if you look at the example code, see in red, we've got that variable that we're declaring right at the top called funcVar. And that is going to be my new variable that I'm declaring to be a function type. So if you look at the top line, var funcVar, and then to the right of it, I give its type, and that's how you do it. So that's how you declare variables of certain types you say var funcVar, let's say I want it to be an integer, I'd say var funcVar int. So now I want it to be a function, so I say var funcVar func, and I give the argument type, integer, right? And I give its return type, so that right there, func, int in parentheses, and then int to the right of funcVar, you'd call that a signature, a function signature. And so you have to put that to the right of the variable name in the declaration. So that first line just declares a new variable called funcVar, which will be able to be assigned to a function. Now then right after that, I declare a function, inc function, incFn and it just does increment. Right, it takes a value x and it increments it, returns x+1. So it's just a function. Now in the main, first thing I do is I say funcVar = incFn. So right there, I am assigning funcVar to be that function incFn. Now notice in that assignment there, I don't put the parenthesis to the right of incFn. So normally, when you write a function, when you call a function, you'd put parenthesis. Open paren, close paren to the right of it, so if I wanted to call incFn, I'd say incFn open paren, close paren, put some parameters inside the parentheses, but that's how you call it. You put those parentheses. We are not calling it here though, right? So all we're doing is we're saying funcVar is basically appointed to the same function, to incFn, so we just do an assignment just as shown. And then once I've done that, I can use funcVar just the same as I would use that function name. So you can look at the last line. I'm doing a print, I'd say print funcVar 1. That's the same as saying print incFn 1. Pass one to the function, it executes it and increments it, it should print out a two. So in this way, the variable can be assigned directly to a function. So this is treating a function as a first-class object. Functions can also be passed as arguments to other functions. So you can see that here, I got this function called applyIt. And all it's supposed to do is apply a function to an integer. So it takes two arguments, my function applyIt. The first argument is a function. I call it afunct, and its type is func. Func takes an integer and returns an integer. The second argument is a value which I'm calling an integer. And the whole point of this function, this whole function applyIt should basically take that function that is its argument, apply that to the val argument, and it should return an integer. And if you look at what the function does in between the curly brackets, all you have is return afunct(val). So it takes whatever that function is that you passed, passes the value to that function, and returns its return value. So in this way, we're passing a function as an argument to another function which is useless sometimes. We'll talk about that. So here's sort of a slightly bigger example where passing a function as an argument. Again at the top, we've got that applyIt that I told you about, right? So we're just going to show how you might use such a function. So applyIt is a function. It takes a function as an argument, and it applies the argument function to the other argument, the val argument. Now then right after I define that, I define two other functions, an inc function and a dec function, increment, decrement functions. And they do what you would expect, return x + 1, return x- 1. So we got those two functions, then we looked at the main and there are two print statements. I call applyIt both times, applyIt with the incFn as the argument and applyIt with the decFn. So when I call applyIt with the incFn, the other argument is 2, right? So an incFn should increment that, so it should print out a 3. And then the next print statement calls applyIt with the decFn, decrement function. It passes it a 2, so we should return a 1 and so 1 should get printed out. So if you run this code, you should get a 3 and 1. Anonymous functions, so anonymous functions are basically functions without a name. So you don't actually need to name a function if you don't want to. Now, it is usually very useful to name functions. If you want to call them, you need a name for them, right? But if you're going to pass a function as an argument to another function, often you don't need to give it an explicit name. So you saw on the last slide just now, I had this inc function and dec function. And I gave them names and I passed it to my applyIt function. But I don't have to actually give them names in order to pass them as arguments. I can just make the function right there in the call. They call this a lambda calculator, because this is a lambda and it comes from lambda calculus, but we won't use those terms. That's a math term that people, where this actually came from. Because these ideas of passing these functions around came from lambda calculus a long time ago. But if we look at the function, we've still got the same applyIt. But if you look at the main, I call applyIt, and the first argument is a function definition, right? So without a name, there's no name for it, but it is otherwise a function definition. So I have it highlighted in red, func takes x integer as an argument, returns an int, and in curly brackets, return x + 1. So that's the increment function. But I never gave it the name inc function, I don't have to. I just define the function the function right there, define it without a name. So it is now called an anonymous function. And I can just pass it directly to applyIt, and I didn't really need the name, right? ApplyIt will just apply that function. Names are more useful if you're going to call this function directly in your code. You want it named associated with it, but in a case like this, you don't need a name. You can just make the function and just pass it as an argument. Thank you.
