Skip to content

Raanelom/Rust-tutorial

Repository files navigation

Rust-tutorial

Using the Rust-book: https://doc.rust-lang.org/book

Chapter 1: Getting Started

https://doc.rust-lang.org/book/ch01-00-getting-started.html

  • Writing the first lines of code (hello_world)
  • Using Cargo, the Rust-package manager

Today I Learned

  • rustup is a CLI for managing Rust versions and tools.
  • Rust is ahead-of-time compiled
  • Source code can be compiled using the (rustup native) rustc-command
  • Formatting code is easy using the (rustup native) command rustfmt
  • Function-like calls using !-signs as suffix indicate the use of a "macro" instead of a "function". For instance println!("hello world").
  • cargo is Rust's Swiss army knife => among others it is a package (crate) manager
  • cargo new automatically creates a git-repository
  • Cargo-configuration files follow the TOML-standard
  • Convenient commands: cargo build | run | check

Chapter 2: Programming a Guessing Game

https://doc.rust-lang.org/stable/book/ch02-00-guessing-game-tutorial.html

  • The chapter-title is pretty much self-descriptive

Today I Learned:

  • Rust automatically brings a small set of functions into scope, referred to as "the prelude": https://doc.rust-lang.org/stable/std/prelude/index.html
  • Bringing other libraries into scope can be done by using the reserved keyword use (e.g. use std::io;)
  • Rust's (module) namespace-separator is ::. It can also be used for calling "associated functions" (a.k.a. static functions). For example String::new();.
  • Variables are immutable by default <3. Use let to create a new variable (e.g. let x = "hello world";)
  • To allow a variable to be modifyable, use the explicit keyword mut (e.g. let mut x = "hello world";)
  • Passing variables to other functions explicitly requires the reference-symbol &. By default, referenced variables are immutable (even if declared as mut-variable). To allow the function to edit the referenced variable, explicitly mention mut again as follows:
    let mut some_variable = String::new();
    some_function(&mut some_variable);
  • Functions may return Result-objects (of type enum). A Result is either Ok or Err. If Err is returned, it can be caught using the function expect("alternative error message here");.
  • Print variables using inline {}, e.g.:
    println("x = {}, y = {}", x, y);
  • Library-Crates can be declared in Cargo.toml in the [dependencies]-section (+ semantic version number)
  • Rust's Crate-registry is located at crate.io
  • Cargo.lock pins Crate version-numbers, unless the newest patches are explicitly fetched using cargo update. In other words, updating from 3.5.5 to 3.5.7 requires cargo update. Otherwise, cargo.lock will preserve the installed package with version 3.5.5.
  • Rust uses match expressions to express conditions simlar to Python's lambda-expressions. It consists of "arms", and each arm consists of a "pattern". If the arm's pattern is matched, the corresponding command is executed. A match-expression is finished if one of the arm patterns evaluated to "true".
  • Rust allows "shadowing" existing variables, e.g. overriding an existing variable "var_a" of type string with a new variable "var_a" with type int.
  • Rust knows multiple signed and unsigned integer types, such as i32, u32, i64 and u64. By default, an integer is 32-bit and signed (i32).
  • Types can be expressed using :, e.g. let x: i32 = 42
  • loop { } is an endless condition-less loop, which can be exited using break
  • Match can also be used as inline expression, for example:
    let x = match guess.trim().parse() {
        Ok(num) => num,
        Err(_) => {
            // handle error here
        }
    };

Chapter 3: Common Programming Concepts

https://doc.rust-lang.org/stable/book/ch03-00-common-programming-concepts.html

Reseved keywords: https://doc.rust-lang.org/stable/book/appendix-01-keywords.html

Today I Learned:

  • Besides (mutable) variables, there's also the concept of "constants" (const). These require type-annotation.
  • Variables can't be declared in global scope. Constants can.
  • Integers range from 8-bit ([u/i]8) to 128-bit ([u/i]128)
  • isize and usize adapt to the restrictions of the OS (in most cases 32-bit vs 64-bit)
  • Possible notations:
    Type Value
    Decimal 98_222
    Hex 0xff
    Octal 0o77
    Binary 0b1111_0000
    Byte (u8 only) b'A'
  • Rust panics in debug mode if an overflow occurs. In Release-mode, the value wraps around (thus for 8-bit integers, 256 => 0, 257 => 1, etc.)
  • There are two type of reserved keywords for floating-points: f32 and f64 (default)
  • Boolean type can be used with keyword bool
  • Rust's char type is a four-byte Unicode Scalar Value
  • A "Scalar type" represents a single value (such as int, string, bool and char), whereas "Compound types" groups multilpe values into one type.
  • Rust has two Compound types: "array" [] and "tuple" (tup). Example:
    let x: (i32, f64, u8) = (500, 6.4, 1);
    let five_hundred = x.0;
    let six_point_four = x.1;
    let one = x.2;
    let test_array = [1, 2, 3];
    let two = test_array[1];
  • Array-types and length can be declared inline:
    let a: [i32; 5] = [1, 2, 3, 4, 5];
  • Repeating the same number in an array:
    let a = [3; 5];
    // a = [3, 3, 3, 3, 3]
  • Using array-indices that are out of bounds will produce an error at compile-time. E.g.:
    let a: [i32; 3] = [1, 2, 3];
    println!("{}", a[9]); // This will not compile
  • snake case is the conventional style for declaring functions.
  • It doesn't matter in which order functions are declared in a source-file.
  • Function-parameters require static typing
  • Rust distinguishes between "Statements" and "Expressions". Statements are instructions that perform some action and do not return a value. Expressions evaluate to a resulting value.
  • Rule of thumb: statements end with a semicolon, whereas expressions do not
  • Use of the return keyword is optional. A function with return type works as follows:
    fn five() -> i32 {
        5 // Return value is implicit
    }
  • Comments: //. There are no block comments in Rust.
  • if and else-expressions have the following shape:
    if some_condition {
        // handle some_condition
    } else if some_other_condition {
        // handle some_other_condition
    } else {
        // handle other cases
    }
  • Inline if/else expressions are also possible:
    // Note: arms should return the same data-type
    let some_str = if some_no > 5 { "high" } else { "low" };
  • Besides loop, Rust has while and for keywords to create repeating code blocks
  • loop-statements are able to return a value using the break-keyword + return-value.
  • for-loops can be compared with C#'s foreach:
    let a = [10, 20, 30, 40, 50];
    for element in a.iter() {
        println!("the value is: {}", element);
    }
  • A Range can be used in combination with a for-loop:
    for number in (1..4) { } // Looping over numbers 1, 2, 3

Chapter 4: Understanding Ownership

https://doc.rust-lang.org/stable/book/ch04-00-understanding-ownership.html

Today I Learned:

  • Each value in Rust has a variable that's its "owner". The relation between values and owners is strictly 1-to-1. If the owner goes out of scope, the value will be dropped.
  • Besides string literals (stored on the Stack) there is the String-type, which is stored on the heap.
  • Strings are mutated via the push_str-method:
    let mut s = String::from("hello");
    s.push_str(", world!"); // push_str() appends a literal to a String
  • The difference between these two string-types is that string-literals are known at compile-time, whereas String is only known at run-time.
  • Copying the contents of one variable into another, will invalidate the initial variable. For example:
    let s1 = String::from("some string");
    let s2 = s1; // Invalidate s1, use s2 from now on
    // This doesn't hold for integer-variables, for which the size is known at compile-time and are therefore stored on the Stack
  • Use clone() to explicityly copy heap-data to another variable.
  • Variables that store data on the heap and that are passed into a function are invalid after the function ends (by default) because it is freed at the end of the scope (= end of function). Example:
    fn main() {
        let s = String::from("hello");  // s comes into scope
    
        some_function(s); // use s in some_function() and hand over the ownership 
        // s is no longer valid here
    }
  • Variables can be passed to a function without the function taking ownership of the variable by using a reference-indicator &. This is called "borrowing":
    fn main() {
        some_var = String::from("some_value");
        some_function(&some_var);
        println!("reuse some_var: {}", some_var);
    }
    
    fn some_function(some_var: &String) {
        // "borrow" value some_var here
        ...
    }
  • References are immutable by default. They can be made mutable by using the &mut-keyword. Only one mutable reference per scope is allowed:
    let mut some_var = String::from("some_value");
    let some_var_ref = &mut some_var;
    
    some_var_ref.push_str("s");
  • It is possible to pass a part of a string as a reference, also referred to as "string slices":
    let s = String::from("some__value");
    let hello = &s[0..5];
    let world = &s[6..11];
  • String literals are of type &str (e.g. let s = "some value" <- s is of type &str); &str is an immutable reference.
  • Slices also generalise to other types, such as lists.

Chapter 5: Using Structs to Structure Related Data

https://doc.rust-lang.org/stable/book/ch05-00-structs.html

Today I Learned:

  • A struct defines the structure of an object. It can hold together multiple related values.
  • It's possible to access fields using the dot notation (e.g. user.email)
  • Fields can only be updated if the entire struct is marked mutable
  • We can use "field init shorthand" to provide values to fields, e.g.:
    // Default assignment notation
    fn build_user(email: String, username: String) -> User {
        User {
            email: email,
            username: username,
            active: true,
            sign_in_count: 1,
        }
    }
    // Field init shorthand notation
    fn build_user(email: String, username: String) -> User {
        User {
            email, // Note that we don't explicitly link the value here
            username, // Here neither
            active: true,
            sign_in_count: 1,
        }
    }
  • This only works if the parameter-name matches the field name
  • To copy over some fields from one object to another, use the .. notation:
    let user2 = User {
        email: String::from("[email protected]"), // Set a unique e-mailadress
        ..user1 // Copy the remaining fields from user 1
    }
  • If you're in doubt whether to pick a Tuple or Struct, choose the Tuple Struct: a named Tuple:
    struct Color(i32, i32, i32);
  • To print entire structs, one could use pretty-print-options using {:?} or {:#?} instead of {}and by annotating structs with#[derive(Debug)]`.
  • There's a difference between methods and functions; methods are defined within the context of a struct, whereas functions are not. The first parameter of a method is always self. Methods are defined within an impl-block
  • self can also be borrowed or referenced, just like any other parameter
  • impl-blocks can also hold "associated functions", i.e. a function without the "self"-parameter. Calling associated functions can be done using the ::-syntax.
  • It's possible to have multiple impl-blocks

Chapter 6: Enums and Pattern Matching

https://doc.rust-lang.org/stable/book/ch06-00-enums.html

  • Enums can be used for enumerating several options, e.g.:
    enum IpAddrKind {
        V4,
        V6
    }
  • It's also possible to pass parameters (any amount/kind) to enums:
    enum IpAddr {
        V4(String),
        V6(String)
    }
  • Just like structs, use impl to add methods to an enum
  • Rust doesn't have null values! :party:
  • Matching enums and Options is also supported by Rust:
    fn plus_one(x: Option<i32>) -> Option<i32> {
        match x {
            None => None, // Do nothing if value is None
            Some(i) => Some(i + 1), // Increment by 1 if value is Some
        }
    }
  • Matches are exhaustive, thus if you miss one, Rust won't compile
  • The "default" match is handled by _
  • if let is another way of expressing a match

Chapter 7: Managing Growing Projects with Packages, Crates, and Modules

https://doc.rust-lang.org/stable/book/ch07-00-managing-growing-projects-with-packages-crates-and-modules.html

  • Larger projects can be structured using Rust's building blocks:
    • Packages: A Cargo feature that lets you build, test, and share crates
    • Crates: A tree of modules that produces a library or executable
    • Modules and use: Let you control the organization, scope, and privacy of paths
    • Paths: A way of naming an item, such as a struct, function, or module
  • A package consists of at least one crate, which could either be a library or a binary crate. A package can only have zero or one library crates, and any number of binary crates.
  • The purpose of a module is mainly to describe the structure of the code and organise it. It also determines the accessability of modules and its content (e.g. a function) by using the keywords pub and priv.
  • Paths can be used to describe the location of items in the module tree. It can either be absolute or relative:
    // Absolute path
    crate::front_of_house::hosting::add_to_waitlist();
    
    // Relative path
    front_of_house::hosting::add_to_waitlist();
  • We can use the as keyword to name module imports: use std::io::Result as IoResult;
  • Imports from the same package can be nested (no *, but use std::{cmp::Ordering, io};). A wildcard * brings all items into scope.
  • The module hierarchy can be separated into files and folders. A nested module is then saved in a separate, similarly named file in a folder with the name of the parent module.

Chapter 8: Common Collections

https://doc.rust-lang.org/stable/book/ch08-00-common-collections.html

  • Besides the built-in data structures such as list and array, Rust also contains the structures vector, string and hash map.
  • A Vector can hold any type of data.
  • It can either hold data of a single type (default), or use enums to store different types.
  • Strings can be concatenated using the + operator. However, this requires borrowing the second argument.
  • Another way to concat two or more strings is by using the format!-macro. It doesn't take ownership of any of its parameters.
  • Slicing a string in Rust requires a range of indices, not a single index as in other languages. This is because some characters take up more than a single byte, such as characters in the string some_str="Здравствуйте". Slicing this string like some_str[0..4] returns the first two characters (because they both take up 2 bytes of space).
  • It's possible to use the string-operations chars() and bytes() to iterate over a string.
  • Conclusion: Strings are complicated

TODO: continue @ https://doc.rust-lang.org/stable/book/ch08-02-strings.html

About

RUST-tutorial

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages