Why test?#
Testing is to programming what sport is to most people: everybody agrees it is important but nobody has time for it. In this chapter, we will introduce the minimum you need to perform tests with pytest and even enjoy it!
Why not using unittest, the Python built-in testing library?
Although it is an additional dependency, pytest is often preferred because it leads to cleaner and easier to read tests, while unittest can lead to a lot of boilerplate code.
Testing#
Testing consist in writing small functions that test:
- whether small units of your code function as expected
- whether these small units integrate well together
- whether your code takes care of edge cases
- whether your code's inputs and outputs are correctly treated
Running your tests allow checking that new changes to the code base did not break anything in your code (at least what your are testing for!).
There are side-effects to writing tests!
As you write tests, you get to experience what it takes to use your library and this might lead you to refactor parts of your code; refactoring is an important part of a software life-cycle!
