Rust – more than a ‘memory safe’ language

Rust – introduction, and history

Rust is a general-purpose programming language tailored for the development of high-load systems. It is used to write web applications, browser engines, blockchains, and other complex platforms that process requests from millions of users. Let’s analyze its main features.

1. Compilation. Before running, the Rust code is immediately translated into machine language, which turns into a set of zeros and ones. Rust differs, for example, from JavaScript, where the code is first run through the interpreter and only then goes to the processor. The trick to compiling is that it makes programs run faster and consume less memory.
2. Multiparadigm. A programming paradigm defines the style in which developers write code. Rust simultaneously supports several paradigms: functional, declarative, and object-oriented programming. Companies often use the latter approach, but Rust can work in any mode.
3. High level of abstraction. Commands in Rust are simple English words: input, let, match, and so on. Therefore, the code on it is easy to read and check. At the same time, the language supports using some low-level tools for finer work with hardware.

The main feature of Rust is a combination of speed and reliability. On the one hand, the language gives the programmer complete access to memory; on the other hand, it ensures him against errors and “shots in the foot.” Therefore, Rust is used to write software for which stable operation under challenging conditions is essential: multi-threaded systems, software for complex calculations, and so on.

At the beginning of the 2000s, there were two approaches to creating computer languages: one emphasized speed, and the other emphasized security. For example, C++ programs worked quickly but could fail due to a random programmer error. And Java, on the contrary, reduced the risks for the developer, but because of this, the programs lost momentum.

The idea behind Rust was to combine the power of C++ with the stability of Java and make the most of both approaches.

Initially, the language developed as an author’s project of Mozilla programmer Graydon Hoare and ultimately came under the company’s control. The Firefox browser was then undergoing an engine update, and Rust helped to implement some important things – for example, loading sites in multi-threaded mode.

By the way, the language got its name in honor of the fungi of the order Rust ( Rust Fungi). These mushrooms have exceptional survivability – according to the author, Rust programs should have become like that.

The strength of Rust is safe and economical memory management. When you write C++ code, you need to work with it manually: place pointers, clean up the program from garbage, and watch for leaks. You are solely responsible if the program crashes due to some bug or vulnerability.

Rust has an entirely different approach – all work with memory happens under the hood of the language according to strictly defined rules:

1. Any value (for example, a number) has an owner variable.
2. A deal can only have one owner at a time.
3. If you move the owner to another part of the code, the value is reset to zero.

If one of these rules is violated, the compiler will generate an error and refuse to build the program. If there is a problem in the code, you will know about it immediately and can fix it on the fly. Finally, if a Rust program is built and runs, it will run smoothly even under stressful conditions.

Rust – a ‘memory safe’ language

Rust uses the concept of ownership to ensure high performance and memory safety. More formally, this is an example of affine typing. All Rust code follows specific rules that allow the compiler to manage memory without wasting runtime. It’s all about the innovative concept of ownership, within which memory is controlled using the following rules:

1. a value (i.e., a region of memory that contains a simple integer or more complex structure) has an owner (i.e., a function);
2. at any given moment in time, the value has only one owner;
3. when execution leaves the scope of the owner (i.e., exits the function), the value is destroyed (freed);
4. the property is passed from one function to another (in this case, the original function loses access to the value);
5. the function “borrows” access to the value from another function, which at the same time remains the owner (this, for example, is how to borrow a book to read);
6. there can be any number of such read-only borrowings;
7. at any given time, there can be only one borrow for reading and writing;
8. the presence of a read-write borrow precludes any read-only borrow;
9. the borrow must refer to an active value (not yet destroyed, in which case there is no null/nill pointer).

This set of rules allows the compiler to statically:

1. determine when to allocate and deallocate memory, and thus write code directly in the generated binary file;
2. ensure the absence of dangling pointers;
3. ensure that there is no “race condition” (when there are simultaneous data writes in an indefinite order to the same memory area);
4. ensure that no memory overflows while references or borrows are smart pointers, according to the last rule.

What is written in Rust

Most often, Rust is used in those projects where stability and reliability under high load and the program’s overall speed are needed.

Rust is suitable for developing OS, web servers, system monitoring programs, and web engines and creating scalable front-end and back-end parts. For example, here are the most notable projects where Rust was the primary programming language:

1. Dropbox is the server part that is responsible for the synchronization.
2. Coursera – Most of the front and back end is written in Rust.
3. Mozilla: Firefox and cache (distributed cache for the compiler).
4. OpenDNS is a service for using public DNS services.
5. Servo is a multi-threaded browser engine.
6. Twitter – uses Rust for the high-load part of the service.

What is Rust used for?

1. Cryptocurrencies and blockchain. For blockchains to make secure transactions between millions of different devices, you need fast and stable code – and programming in Rust allows you to do this. For example, it runs one of the most popular clients for the Ethereum platform – Parity.
2. OS. Rust can write a full-fledged OS from scratch with drivers, a file manager, a browser, and other stuffing. In practice, few people want to compete with Apple and Microsoft, so there have been few successful projects. You can only remember the OS for smart devices Tock and the secure Redox operating system.
3. Websites and web applications. Because Rust is resistant to loads, it can be used to develop server-side site logic. For example, the Dropbox cloud service rewrote its file storage system on it – and these are tens of thousands of server machines where gigabytes of new data are constantly uploaded.
4. Browser Engines. It’s not for nothing that Rust was invented in Mozilla – the language was initially sharpened precisely for browser development. It has features for quickly rendering web pages in multi-threaded mode – you can split the site load into several tasks and distribute them between processor cores.
5. Machine learning. Rust supports many libraries for working with big data: training algorithms, data visualization, linear algebra, working with vectors, image processing; you name it. If you see yourself in Data Science in the future, look closely at Rust.

Pros of Rust

1. Speed and security. As we wrote above, Rust programs are fast and break only in hopeless situations.
2. Static typing. In Rust, variable type checking occurs at the compilation stage – this is a plus for the reliability of the language.
3. C++ compatible. In Rust programs, you can quickly call blocks of code from C and C++ and vice versa.
4. Caring compiler. If an error occurs in the program, the compiler will tell you what went wrong and advise you on how to fix it.
5. Cross-platform. Rust runs on almost any device, from computers and smartphones to ATMs and smartwatches.
6. Multithreading. Rust programs can perform several operations simultaneously using the processor’s power.
7. High salaries for programmers.
8. Documentation. Rust is very well documented, not only in the language itself but also in its standard library and, in general, the libraries that are in the public domain.

Cons of Rust 

1. Limited OOP support. For example, it is only possible to implement full-fledged inheritance of code and classes with crutches.
2. Frequent updates. The syntax of Rust changes with each new version, so developers have to relearn constantly. On the other hand, the language ends up being safer and easier to use.
3. Low popularity. Companies are mainly looking for experienced specialists, and there are very few offers for juniors.
4. The difficulty of mastering. Mastering this language requires a solid amount of knowledge and skills.
5. Reading difficulty. Developer spends an average of 90% of their working time reading code (either their own or someone else’s) and only 10% writing it.

Who is the Rust programming language suitable for?

Where and by whom is the Rust programming language used, and what is it best suited for? These are the following groups.

Large development teams

Rust has everything to be shared by many programmers of different skill levels. Usually, there are many minor errors in low-level code, which, when working with other languages, can only be detected with the help of global testing and detailed, rigorous analysis (and you need to involve the most experienced specialists for this).

In Rust, the compiler does not miss these subtle concurrency errors. Thanks to this, developers can focus more on the logic of the program instead of looking for problem areas.

Rust also has an extensive system programming toolkit:

– Cargo. This built-in tool allows you to add, compile and manage dependencies using the Rust ecosystem.

– Rust allows all developers to code in a consistent, standard style.

– Rust Language Server provides IDE (Integrated Development Environment) support with automatic code maintenance and a built-in error manager.

There are other tools in the Rust ecosystem that, along with those listed, allow developers to excel when creating system codes.

Students, future programmers

Knowledge of the language will be sufficient for students and those who study system concepts. Rust provides an understanding of what operating system development is all about.

The Rust team is working hard to make this programming language as accessible as possible for beginners and, in general, for everyone interested in learning it.

Large and small companies

They use Rust to perform a variety of tasks. This applies to command line tools, web services, embedded components, DevOps tools, bioinformatics, and cryptocurrencies. It also implies working with audio and video files (analysis and transcoding), search engines, and applications. Plus, machine learning and the main parts of the Firefox browser are also affected

Anyone who cares about speed and stability

This refers to the speed of programs written in Rust and the speed of development of these programs. Rust’s compiler checks guarantee stability (through feature additions and refactorings). Codes created using languages that do not have these checks are fragile, and developers are often hesitant to change them.

Rust can compile high-level functions into low-level code as fast as hand-written code. Plus, it aims for zero-cost abstractions. All this allows you to create regulations that are not only safe but also fast.

Conclusion

Despite its young age (a little more than ten years old), Rust has become an essential language in the software development environment. Rust is the right solution for mission-critical, functionally complex projects where performance is measured in nanoseconds and, most important – where memory safety is needed.