Robust Python

Chapter 16. Dependencies

It is difficult to write a program with no dependencies. Functions depend on other functions, modules depend on other modules, and programs depend on other programs. Architecture is fractal; no matter what level you’re looking at, your code can be represented as some sort of box-and-arrows diagram, like in Figure 16-1. It doesn’t matter if it’s functions, classes, modules, programs, or systems, you can draw a similar diagram to Figure 16-1 to represent the dependencies in your code.

Figure 16-1. Box-and-arrows diagram

However, if you don’t actively manage your dependencies, you soon get to what’s known as “spaghetti code,” making your box-and-arrows diagram look like Figure 16-2.

Figure 16-2. A tangled mess of dependencies

In this chapter, you are going to learn all about dependencies and how to keep them under control. You’ll learn about different types of dependencies, all of which should be managed with different techniques. You’ll learn how to graph your dependencies, and how to interpret whether you have a healthy system. You’ll learn how to truly simplify your code architecture, which will help you manage complexity and increase the robustness of your codebase.

Relationships

Dependencies are, in essence, relationships.