Async Rust

Chapter 11. Testing

We know by now how powerful writing async systems in Rust can be. However, when building big async systems in Rust, we need to know how to test our async code. This is because the complexity of the system grows as the size of the system increases.

Testing increases the feedback of the code that we are implementing, making our code faster to write and safer. For instance, if we have a big codebase and need to alter or add functionality into a piece of code, it would be slow and dangerous if we had to spin up the entire system and run it to see if our code works. Instead, altering or adding the code we need and running the specific tests for that piece of code not only provides a faster feedback loop, but also enables us to test more edge cases, making our code safer. In this chapter, we explore various approaches to testing async and the interfaces between our code and external systems.

By the end of this chapter, you will be able to build isolated tests where you can mock interfaces and inspect the calls to those interfaces. This gives you the ability to build truly isolated atomic tests. You will also be able to test for synchronization pitfalls like deadlocks, race conditions, and channel capacity issues that block async tasks. Finally, you will also learn how to mock interactions with networks such as servers, obtain fine-grained testing control of all futures, and know when to poll them to see how your system progresses under different polling conditions. ...