From: Ralf Jung Date: Mon, 15 Jun 2015 11:39:39 +0000 (+0200) Subject: refine parts 02 and 03 X-Git-Url: https://git.ralfj.de/rust-101.git/commitdiff_plain/23c658ed26c424dd6f38468108dd8ac34534fe1e?ds=inline;hp=6a83fbe44cc324f35f99da3ad290f0c0ef71260c refine parts 02 and 03 --- diff --git a/src/part02.rs b/src/part02.rs index dc885c0..233fa0f 100644 --- a/src/part02.rs +++ b/src/part02.rs @@ -19,22 +19,27 @@ pub enum SomethingOrNothing { // Instead of writing out all the variants, we can also just import them all at once. pub use self::SomethingOrNothing::*; // What this does is to define an entire family of types: We can now write -// `SomethingOrNothing` to get back our `NumberOrNothing`, but we -// can also write `SomethingOrNothing` or even `SomethingOrNothing>`. +// `SomethingOrNothing` to get back our `NumberOrNothing`. +type NumberOrNothing = SomethingOrNothing; +// However, we can also write `SomethingOrNothing` or even `SomethingOrNothing>`. // In fact, such a type is so useful that it is already present in the standard // library: It's called an *option type*, written `Option`. // Go check out its [documentation](http://doc.rust-lang.org/stable/std/option/index.html)! // (And don't worry, there's indeed lots of material mentioned there that we did not cover yet.) -// **Exercise**: Write functions converting between `SomethingOrNothing` and `Option`. You will have to use -// the names of the constructor of `Option`, which you can find in the documentation I linked above. -// -// Here's a skeleton for your solution, you only have to fill in the function bodies. -// (`panic!` is, again, a macro - this one terminates execution when it is reached). +// The types are so similar, that we can provide a generic function to construct a `SomethingOrNothing` +// from an `Option`, and vice versa. + +// **Exercise**: Implement such functions! I provided a skeleton of the solution. Here, +// `panic!` is another macro. This one terminates execution with the given message. // // Notice the syntax for giving generic implementations to generic types: Think of the first `` // as *declaring* a type variable ("I am doing something for all types `T`"), and the second `` as // *using* that variable ("The thing I do, is implement `SomethingOrNothing`"). +// +// Inside an `impl`, `Self` refers to the type we are implementing things for. Here, it is +// an alias for `SomethingOrNothing`. +// Remember that `self` is the `this` of Rust, and implicitly has type `Self`. impl SomethingOrNothing { fn new(o: Option) -> Self { panic!("Not yet implemented.") @@ -44,15 +49,11 @@ impl SomethingOrNothing { panic!("Not yet implemented.") } } -// Inside an `impl`, `Self` refers to the type we are implementing things for. Here, it is -// an alias for `SomethingOrNothing`. -// Remember that `self` is the `this` of Rust, and implicitly has type `Self`. -// // Observe how `new` does *not* have a `self` parameter. This corresponds to a `static` method // in Java or C++. In fact, `new` is the Rust convention for defining constructors: They are // nothing special, just static functions returning `Self`. - -// You can call static functions, and in particular constructors, as follows: +// +// You can call static functions, and in particular constructors, as demonstrated in `call_constructor`. fn call_constructor(x: i32) -> SomethingOrNothing { SomethingOrNothing::new(Some(x)) } @@ -64,19 +65,24 @@ fn call_constructor(x: i32) -> SomethingOrNothing { // So, as a first step towards a generic `vec_min`, we define a `Minimum` trait. // For now, just ignore the `Copy`, we will come back to this point later. // A `trait` is a lot like interfaces in Java: You define a bunch of functions -// you want to have implemented, and their argument and return types. +// you want to have implemented, and their argument and return types.
+// The function `min` takes to arguments of the same type, but I made the +// first argument the special `self` argument. I could, alternatively, have +// made `min` a static function as follows: `fn min(a: Self, b: Self) -> Self`. +// However, in Rust one typically prefers methods over static function wherever possible. pub trait Minimum : Copy { - fn min(a: Self, b: Self) -> Self; + fn min(self, b: Self) -> Self; } -// Now we can write `vec_min`, generic over a type `T` that we demand to satisfy the `Minimum` trait. -// This is called a *trait bound*. -// The only difference to the version from the previous part is that we call `T::min` (the `min` -// function provided for type `T`) instead of `std::cmp::min`. +// Next, we write `vec_min` as a generic function over a type `T` that we demand to satisfy the `Minimum` trait. +// This requirement is called a *trait bound*. +// The only difference to the version from the previous part is that we call `e.min(n)` instead +// of `std::cmp::min(n, e)`. Rust automatically figures out that `n` is of type `T`, which implements +// the `Minimum` trait, and hence we can call that function. // -// Notice a crucial difference to templates in C++: We actually have to declare which traits +// There is a crucial difference to templates in C++: We actually have to declare which traits // we want the type to satisfy. If we left away the `Minimum`, Rust would have complained that -// we cannot call `min`. Just try it! There is no reason to believe that `T` provides such an operation. +// we cannot call `min`. Just try it!
// This is in strong contrast to C++, where the compiler only checks such details when the // function is actually used. pub fn vec_min(v: Vec) -> SomethingOrNothing { @@ -84,23 +90,36 @@ pub fn vec_min(v: Vec) -> SomethingOrNothing { for e in v { min = Something(match min { Nothing => e, - Something(n) => T::min(n, e) + Something(n) => e.min(n) }); } min } +// Before going on, take a moment to ponder the flexibility of Rust's take on abstraction: +// We just defined our own, custom trait (interface), and then implemented that trait +// *for an existing type*. With the hierarchical approach of, e.g., C++ or Java, +// that's not possible: We cannot make an existing type suddenly also inherit from our abstract base class. +// +// In case you are worried about performance, note that Rust performs *monomorphisation* +// of generic functions: When you call `vec_min` with `T` being `i32`, Rust essentially goes +// ahead and creates a copy of the function for this particular type, filling in all the blanks. +// In this case, the call to `T::min` will become a call to our implementation *statically*. There is +// no dynamic dispatch, like there would be for Java interface methods or C++ `virtual` methods. +// This behavior is similar to C++ templates. The optimizer (Rust is using LLVM) then has all the +// information it could want to, e.g., inline function calls. // To make the function usable with a `Vec`, we implement the `Minimum` trait for `i32`. impl Minimum for i32 { - fn min(a: Self, b: Self) -> Self { - std::cmp::min(a, b) + fn min(self, b: Self) -> Self { + std::cmp::min(self, b) } } // In order to run our code and see the result, we again provide a `print` function. -// This also shows that we can have multiple `impl` blocks for the same type, and we +// This also shows that we can have multiple `impl` blocks for the same type (remember +// that `NumberOrNothing` is just a type alias for `SomethingOrNothing`), and we // can provide some methods only for certain instances of a generic type. -impl SomethingOrNothing { +impl NumberOrNothing{ pub fn print(self) { match self { Nothing => println!("The number is: "), @@ -112,14 +131,6 @@ impl SomethingOrNothing { // Now we are again ready to run our code. Remember to change `main.rs` appropriately. // Rust figures out automatically that we want the `T` of `vec_min` to be `i32`, and // that `i32` implements `Minimum` and hence all is good. -// -// In case you are worried about performance, note that Rust performs *monomorphisation* -// of generic functions: When you call `vec_min` with `T` being `i32`, Rust essentially goes -// ahead and creates a copy of the function for this particular type, filling in all the blanks. -// In this case, the call to `T::min` will become a call to our implementation *statically*. There is -// no dynamic dispatch, like there would be for Java interface methods or C++ `virtual` methods. -// This behavior is similar to C++ templates. The optimizer (Rust is using LLVM) then has all the -// information it could want to, e.g., inline function calls. fn read_vec() -> Vec { vec![18,5,7,3,9,27] } @@ -129,30 +140,6 @@ pub fn main() { min.print(); } -// If this printed `3`, then you generic `vec_min` is working! -// -// Before going on, take a moment to ponder the flexibility of Rust's take on abstraction: -// We just defined our own, custom trait (interface), and then implemented that trait -// *for an existing type*. With the hierarchical approach of, e.g., C++ or Java, -// that's not possible: We cannot make an existing type suddenly also inherit from our abstract base class. - -// **Exercise**: Define a trait `Print` to write a generic version of `SomethingOrNothing::print`. -// Implement that trait for `i32`, and change the code above to use it. -// I will again provide a skeleton for this solution. It also shows how to attach bounds to generic -// implementations (just compare it to the `impl` block from the previous exercise). -// You can read this as "For all types `T` satisfying the `Print` trait, I provide an implementation -// for `SomethingOrNothing`". -// -// Notice that I called the function on `SomethingOrNothing` `print2` to disambiguate from the `print` defined above. -// -// *Hint*: There is a macro `print!` for printing without appending a newline. -trait Print { - /* Add things here */ -} -impl SomethingOrNothing { - fn print2(self) { - panic!("Not yet implemented.") - } -} +// If this printed `3`, then you generic `vec_min` is working! So get ready for the next part. // [index](main.html) | [previous](part01.html) | [next](part03.html) diff --git a/src/part03.rs b/src/part03.rs index c468555..f76c4aa 100644 --- a/src/part03.rs +++ b/src/part03.rs @@ -3,18 +3,25 @@ // In part 00, I promised that we would eventually replace `read_vec` by a function // that actually asks the user to enter a bunch of numbers. Unfortunately, -// I/O is a complicated topic, so the code to do that is not pretty - but well, +// I/O is a complicated topic, so the code to do that is not exactly pretty - but well, // let's get that behind us. -// IO/ is provided by the module `std::io`, so we first import that. +// I/O is provided by the module `std::io`, so we first import that. // We also import the I/O *prelude*, which brings a bunch of commonly used I/O stuff // directly available. use std::io::prelude::*; use std::io; -// Let's now go over this function line-by-line. +// Let's now go over this function line-by-line. First, we call the constructor of `Vec` +// to create an empty vector. As mentioned in the previous part, `new` here is just +// a static function with no special treatment. While it is possible to call `new` +// for a particular type (`Vec::::new()`), the common way to make sure we +// get the right type is to annotate a type at the *variable*. It is this variable +// that we interact with for the rest of the function, so having its type available +// (and visible!) is much more useful. Without knowing the return type of `Vec::new`, +// specifying its type parameter doesn't tell us all that much. fn read_vec() -> Vec { - let mut vec = Vec::new(); + let mut vec: Vec = Vec::::new(); // The central handle to the standard input is made available by `io::stdin()`. let stdin = io::stdin(); println!("Enter a list of numbers, one per line. End with Ctrl-D."); @@ -25,24 +32,29 @@ fn read_vec() -> Vec { // it. (See [the documentation](http://doc.rust-lang.org/stable/std/io/struct.Stdin.html) for more // details.) for line in stdin.lock().lines() { - // The `line` we have here is not yet of type `String`. The problem with I/O is that it can always - // go wrong, so `line` has type `io::Result`. This is a lot like `Option` ("a `String` or + // Rust's type for (dynamic, growable) strings is `String`. However, our variable `line` + // here is not yet of that type. The problem with I/O is that it can always go wrong, so + // `line` has type `io::Result`. This is a lot like `Option` ("a `String` or // nothing"), but in the case of "nothing", there is additional information about the error. // Again, I recommend to check [the documentation](http://doc.rust-lang.org/stable/std/io/type.Result.html). // You will see that `io::Result` is actually just an alias for `Result`, so click on that to obtain // the list of all constructors and methods of the type. // We will be lazy here and just assume that nothing goes wrong: `unwrap()` returns the `String` if there is one, - // and halts the program (with an appropriate error message) otherwise. Can you find the documentation - // of `Result::unwrap()`? + // and panics the program otherwise. Since a `Result` carries some details about the error that occurred, + // there will be a somewhat reasonable error message. Still, you would not want a user to see such + // an error, so in a "real" program, we would have to do proper error handling. + // Can you find the documentation of `Result::unwrap()`? + // + // I chose the same name (`line`) for the new variable to ensure that I will never, accidentally, + // access the "old" `line` again. let line = line.unwrap(); // Now that we have our `String`, we want to make it an `i32`. `parse` is a method on `String` that // can convert a string to anything. Try finding it's documentation! // In this case, Rust *could* figure out automatically that we need an `i32` (because of the return type - // of the function), but that's a bit too much magic for my taste. So I use this opportunity to - // introduce the syntax for explicitly giving the type parameter of a generic function: `parse::` is `parse` - // with its generic type set to `i32`. + // of the function), but that's a bit too much magic for my taste. We are being more explicit here: + // `parse::` is `parse` with its generic type set to `i32`. match line.parse::() { // `parse` returns again a `Result`, and this time we use a `match` to handle errors (like, the user entering // something that is not a number). @@ -60,8 +72,8 @@ fn read_vec() -> Vec { } // So much for `read_vec`. If there are any questions left, the documentation of the respective function -// should be very helpful. I will not always provide the links, as the documentation is quite easy to navigate -// and you should get used to that. +// should be very helpful. Try finding the one for `Vec::push`. I will not always provide the links, +// as the documentation is quite easy to navigate and you should get used to that. // For the rest of the code, we just re-use part 02 by importing it with `use`. // I already sneaked a bunch of `pub` in the other module to make this possible: Only @@ -76,56 +88,23 @@ pub fn main() { min.print(); } -// After all this nit-picking about I/O details, let me show you quickly something unrelated, -// but really nice: Rust's built-in support for testing. -// Now that the user can run our program on loads of inputs, we better make sure that it is correct. -// To be able to test the result of `vec_min`, we first have to write a function that -// is able to test equality if `SimethingOrNothing`. So let's quickly do that. - -// `equals` performs pattern-matching on both `self` and `other` to test the two for being -// equal. Because we are lazy, we want to write only one `match`. so we group the two into a -// pair such that we can match on both of them at once. You can read the first arm of the match -// as testing whether `(self, other)` is `(Nothing, Nothing)`, which is the case exactly if -// both `self` and `other` are `Nothing`. Similar so for the second arm. -impl SomethingOrNothing { - pub fn equals(self, other: Self) -> bool { - match (self, other) { - (Nothing , Nothing ) => true, - (Something(n), Something(m)) => n == m, - // `_` is the syntax for "I don't care", so this is how you add a default case to your `match`. - _ => false, - } - } -} - -// Now we are almost done! Writing a test in Rust is shockingly simple. Just write a function -// that takes no arguments as returns nothing, and add `#[test]` right in front of it. -// That's called an *attribute*, and the `test` attribute, well, declares the function to -// be a test. - -// Within the function, we can then use `panic!` to indicate test failure. Helpfully, there's -// a macro `assert!` that panics if its argument becomes `false`. -// Using `assert!` and our brand-new `equals`, we can now call `vec_min` with some lists -// and make sure it returns The Right Thing. -#[test] -fn test_vec_min() { - assert!(vec_min(vec![6,325,33,532,5,7]).equals(Something(5))); - assert!(vec_min(vec![6,325,33,532]).equals(Something(6))); -} -// To execute the test, run `cargo test`. It should tell you that everything is all right. -// Now that was simple, wasn't it? -// -// **Exercise**: Add a case to `test_vec_min` that checks the behavior on empty lists. +// **Exercise**: Define a trait `Print` to write a generic version of `SomethingOrNothing::print`. +// Implement that trait for `i32`, and change the code above to use it. +// I will again provide a skeleton for this solution. It also shows how to attach bounds to generic +// implementations (just compare it to the `impl` block from the previous exercise). +// You can read this as "For all types `T` satisfying the `Print` trait, I provide an implementation +// for `SomethingOrNothing`". // -// **Exercise**: Change `vec_min` such that everything still compiles, but the test fails. +// Notice that I called the function on `SomethingOrNothing` `print2` to disambiguate from the `print` defined previously. // -// **Bonus Exercise**: Because `String::parse` is itself generic, you can change `read_vec` to -// be a generic function that works for any type, not just for `i32`. However, you will have to add -// a trait bound to `read_vec`, as not every type supports being parsed.
-// Once you made `vec_min` generic, copy your generic `print` from the previous part. Implement all -// our traits (`Minimum` and `Print`) for `f32` (32-bit floating-point numbers), and change `part_main()` -// such that your program now computes the minimum of a list of floating-point numbers.
-// *Hint*: You can figure out the trait bound `read_vec` needs from the documentation of `String::parse`. -// Furthermore, `std::cmp::min` works not just for `i32`, but also for `f32`. +// *Hint*: There is a macro `print!` for printing without appending a newline. +trait Print { + /* Add things here */ +} +impl SomethingOrNothing { + fn print2(self) { + panic!("Not yet implemented.") + } +} // [index](main.html) | [previous](part02.html) | [next](part04.html)