Book Image

Practical System Programming for Rust Developers

By : Prabhu Eshwarla
Book Image

Practical System Programming for Rust Developers

By: Prabhu Eshwarla

Overview of this book

Modern programming languages such as Python, JavaScript, and Java have become increasingly accepted for application-level programming, but for systems programming, C and C++ are predominantly used due to the need for low-level control of system resources. Rust promises the best of both worlds: the type safety of Java, and the speed and expressiveness of C++, while also including memory safety without a garbage collector. This book is a comprehensive introduction if you’re new to Rust and systems programming and are looking to build reliable and efficient systems software without C or C++. The book takes a unique approach by starting each topic with Linux kernel concepts and APIs relevant to that topic. You’ll also explore how system resources can be controlled from Rust. As you progress, you’ll delve into advanced topics. You’ll cover network programming, focusing on aspects such as working with low-level network primitives and protocols in Rust, before going on to learn how to use and compile Rust with WebAssembly. Later chapters will take you through practical code examples and projects to help you build on your knowledge. By the end of this Rust programming book, you will be equipped with practical skills to write systems software tools, libraries, and utilities in Rust.
Table of Contents (17 chapters)
1
Section 1: Getting Started with System Programming in Rust
6
Section 2: Managing and Controlling System Resources in Rust
12
Section 3: Advanced Topics

Dealing with errors

Error handling deals with the question: how do we communicate program errors to users?

In our project, errors can occur due to two main reasons—there could be a programming error, or an error could occur due to invalid inputs. Let's first discuss the Rust approach to error handling.

In Rust, errors are first-class citizens in that an error is a data type in itself, just like an integer, string, or vector. Because error is a data type, type checking can happen at compile time. The Rust standard library has a std::error::Error trait implemented by all errors in the Rust standard library. Rust does not use exception handling, but a unique approach where a computation can return a Result type:

enum Result<T, E> {   Ok(T),   Err(E),}

Result<T, E> is an enum with two variants, where Ok(T) represents success and Err(E) represents the error returned. Pattern matching is used to handle the two types of return...