Anthony Shaw - Wily Python: Writing simpler and more maintainable Python

Everyone starts with the best intentions with their Python projects, "this time it's going to be clean, simple and maintainable". But code evolves over time, requirements change and codebases can get messy and complicated quickly.

In this talk, you will learn how to use wily to measure and graph how complicated your Python code is and a series of practical techniques to simplify it. wily will show you which parts of your projects are becoming or have become hard to maintain and need a refactor. Once you know where the skeletons are, you will learn practical techniques for refactoring "complex" code and some resources to use to take your refactoring to the next level.

https://us.pycon.org/2019/schedule/presentation/162/

Transcript

1. Wily Python: Writing simpler and more maintainable Python Anthony Shaw @anthonypjshaw
2. Is my code complicated? By what measure? How do I simplify it?
3. Introduction Is my code complicated? By what measure? How do I simplify it? Beginner Level Intermediate to Advanced
4. <convince people that simpler code is good in 45 seconds>
5. Lines of code
6. Fewer lines doesn’t mean less complexity
7. “ Readability counts ” — Tim Peters
8. Cyclomatic Complexity Make it a meal? Small, large or super-size? What drink? Diet or regular? Yes Large Cola Diet * Other burgers are available +1 +1 +1 +1 Big Mac please *
11. There’s wisdom in complex code
12. Functionality More Less Less More Users ! Lines of Code "Cyclomatic Complexity " Lines of Code "Cyclomatic Complexity ! Lines of Code ! Cyclomatic Complexity " Lines of Code ! Cyclomatic Complexity
13. Halstead Metrics Warning: This chapter contains math
14. Values Variables Operands ∑ ∑ Built in syntax Operators ∑ Operators and Operands
15. Length and Vocabulary operators operands Vocabulary (h) ∃! ∃! operators operands Length (N) ∑ ∑
16. def foo(): a = 1 b = a + 1 return b Halstead in the AST
17. Volume and Difficulty Better than LoC Length (N) Vocab (h) Volume (V) !"#$ operators operands Difficulty (D) ∃! 2 operands ∃! ∑ A great compliment to CC
18. Effort Volume Difficulty Effort (E) operators operands ∃! 2 operands ∃! ∑ Length (N) Vocab (h) "#$%
19. Maintainability Index 0-25 25-50 50-75 75-100 MI = 171 − 5.2 ln , − 0.23 / − 16.2 ln(2) Halstead Volume Cyclomatic Complexity LoC Unmaintainable Cause for concern Needs improvement Super-hero code
20. Radon $ pip install radon $ radon cc my_application.py -s my_application.py F 4:0 main - B (6) $ radon mi my_application.py -s my_application.py - A (87.42)
21. $ radon hal my_application.py my_application.py: h1: 3 h2: 6 N1: 3 N2: 6 vocabulary: 9 length: 9 calculated_length: 20.264662506490406 volume: 28.529325012980813 difficulty: 1.5 effort: 42.793987519471216
23. Cache Revisions Operators Indexes Git Repo wily build
24. Demo of wily
25. Complexity has gravity app.py
26. Testing and refactoring •High test coverage should be a given •Test coverage is not an indicator of behavioral coverage •Fixing known bugs is hard
27. • Wily can benefit you in tracking complexity • Complexity Measures are long-term • Make sure you have good behavioral test coverage • Refactor often • Divide and conquer Talk notes
28. Thanks! realpython.com/python-refactoring anthonyshaw@apache.org @anthonypjshaw tonybaloney.github.io