1 // Rust-101, Part 02: Generic types, Traits
2 // ========================================
6 // Let us for a moment reconsider the type `NumberOrNothing`. Isn't it a bit
7 // annoying that we had to hard-code the type `i32` in there? What if tomorrow,
8 // we want a `CharOrNothing`, and later a `FloatOrNothing`? Certainly we don't
9 // want to re-write the type and all its inherent methods.
11 // ## Generic datatypes
13 // The solution to this is called *generics* or *polymorphism* (the latter is Greek,
14 // meaning "many shapes"). You may know something similar from C++ (where it's called
15 // *templates*) or Java, or one of the many functional languages. So here, we define
16 // a generic type `SomethingOrNothing`.
17 pub enum SomethingOrNothing<T> {
21 // Instead of writing out all the variants, we can also just import them all at once.
22 pub use self::SomethingOrNothing::*;
23 // What this does is to define an entire family of types: We can now write
24 // `SomethingOrNothing<i32>` to get back our `NumberOrNothing`.
25 type NumberOrNothing = SomethingOrNothing<i32>;
26 // However, we can also write `SomethingOrNothing<bool>` or even `SomethingOrNothing<SomethingOrNothing<i32>>`.
27 // In fact, such a type is so useful that it is already present in the standard
28 // library: It's called an *option type*, written `Option<T>`.
29 // Go check out its [documentation](http://doc.rust-lang.org/stable/std/option/index.html)!
30 // (And don't worry, there's indeed lots of material mentioned there that we did not cover yet.)
32 // ## Generic `impl`, Static functions
33 // The types are so similar, that we can provide a generic function to construct a `SomethingOrNothing<T>`
34 // from an `Option<T>`, and vice versa.
35 // **Exercise 02.1**: Implement such functions! I provided a skeleton of the solution. Here,
36 // `unimplemented!` is another macro. This one terminates execution saying that something has not yet
39 // Notice the syntax for giving generic implementations to generic types: Think of the first `<T>`
40 // as *declaring* a type variable ("I am doing something for all types `T`"), and the second `<T>` as
41 // *using* that variable ("The thing I do, is implement `SomethingOrNothing<T>`").
43 // Inside an `impl`, `Self` refers to the type we are implementing things for. Here, it is
44 // an alias for `SomethingOrNothing<T>`.
45 // Remember that `self` is the `this` of Rust, and implicitly has type `Self`.
46 impl<T> SomethingOrNothing<T> {
47 fn new(o: Option<T>) -> Self {
51 fn to_option(self) -> Option<T> {
55 // Observe how `new` does *not* have a `self` parameter. This corresponds to a `static` method
56 // in Java or C++. In fact, `new` is the Rust convention for defining constructors: They are
57 // nothing special, just static functions returning `Self`.
59 // You can call static functions, and in particular constructors, as demonstrated in `call_constructor`.
60 fn call_constructor(x: i32) -> SomethingOrNothing<i32> {
61 SomethingOrNothing::new(Some(x))
65 // Now that we have a generic `SomethingOrNothing`, wouldn't it be nice to also gave a generic
66 // `vec_min`? Of course, we can't take the minimum of a vector of *any* type. It has to be a type
67 // supporting a `min` operation. Rust calls such properties that we may demand of types *traits*.
69 // So, as a first step towards a generic `vec_min`, we define a `Minimum` trait.
70 // For now, just ignore the `Copy`, we will come back to this point later.
71 // A `trait` is a lot like interfaces in Java: You define a bunch of functions
72 // you want to have implemented, and their argument and return types.<br/>
73 // The function `min` takes to arguments of the same type, but I made the
74 // first argument the special `self` argument. I could, alternatively, have
75 // made `min` a static function as follows: `fn min(a: Self, b: Self) -> Self`.
76 // However, in Rust one typically prefers methods over static function wherever possible.
77 pub trait Minimum : Copy {
78 fn min(self, b: Self) -> Self;
81 // Next, we write `vec_min` as a generic function over a type `T` that we demand to satisfy the `Minimum` trait.
82 // This requirement is called a *trait bound*.
83 // The only difference to the version from the previous part is that we call `e.min(n)` instead
84 // of `std::cmp::min(n, e)`. Rust automatically figures out that `n` is of type `T`, which implements
85 // the `Minimum` trait, and hence we can call that function.
87 // There is a crucial difference to templates in C++: We actually have to declare which traits
88 // we want the type to satisfy. If we left away the `Minimum`, Rust would have complained that
89 // we cannot call `min`. Just try it!<br/>
90 // This is in strong contrast to C++, where the compiler only checks such details when the
91 // function is actually used.
92 pub fn vec_min<T: Minimum>(v: Vec<T>) -> SomethingOrNothing<T> {
93 let mut min = Nothing;
95 min = Something(match min {
97 Something(n) => e.min(n)
102 // Before going on, take a moment to ponder the flexibility of Rust's take on abstraction:
103 // We just defined our own, custom trait (interface), and then implemented that trait
104 // *for an existing type*. With the hierarchical approach of, e.g., C++ or Java,
105 // that's not possible: We cannot make an existing type suddenly also inherit from our abstract base class.
107 // In case you are worried about performance, note that Rust performs *monomorphisation*
108 // of generic functions: When you call `vec_min` with `T` being `i32`, Rust essentially goes
109 // ahead and creates a copy of the function for this particular type, filling in all the blanks.
110 // In this case, the call to `T::min` will become a call to our implementation *statically*. There is
111 // no dynamic dispatch, like there would be for Java interface methods or C++ `virtual` methods.
112 // This behavior is similar to C++ templates. The optimizer (Rust is using LLVM) then has all the
113 // information it could want to, e.g., inline function calls.
115 // ## Trait implementations
116 // To make the function usable with a `Vec<i32>`, we implement the `Minimum` trait for `i32`.
117 impl Minimum for i32 {
118 fn min(self, b: Self) -> Self {
119 std::cmp::min(self, b)
123 // We again provide a `print` function. This also shows that we can have multiple `impl` blocks
124 // for the same type (remember that `NumberOrNothing` is just a type alias for `SomethingOrNothing<i32>`),
125 // and we can provide some methods only for certain instances of a generic type.
126 impl NumberOrNothing {
129 Nothing => println!("The number is: <nothing>"),
130 Something(n) => println!("The number is: {}", n),
135 // Now we are again ready to run our code. Remember to change `main.rs` appropriately.
136 // Rust figures out automatically that we want the `T` of `vec_min` to be `i32`, and
137 // that `i32` implements `Minimum` and hence all is good.
138 fn read_vec() -> Vec<i32> {
142 let vec = read_vec();
143 let min = vec_min(vec);
147 // If this printed `3`, then you generic `vec_min` is working! So get ready for the next part.
149 // [index](main.html) | [previous](part01.html) | [next](part03.html)