Switch branches/tags
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
157 lines (82 sloc) 4.32 KB

Effective Python

59 Specific Ways to Write Better Python
by Brett Slatkin, Senior Staff Software Engineer @ Google


  • Audience: Advanced programmers who want to be more effective in Python (i.e., being more Pythonic)
  • Examples use Python 3 by default


About the Author

1. Pythonic Thinking

1. Know Which Version of Python You're Using

2. Follow the PEP 8 Style Guide

3. Know the Differences Between bytes, str, and unicode

4. Write Helper Functions Instead of Complex Expressions

5. Know How to Slice Sequences

6. Avoid Using start, end, and stride in a Single Slice

7. Use List Comprehensions Instead of map and filter

8. Avoid More Than Two Expressions in List Comprehensions

9. Consider Generator Expressions for Large Comprehensions

10. Prefer enumerate Over range

11. Use zip to Process Iterators in Parallel

12. Avoid else Blocks After for and while Loops

13. Take Advantage of Each Block try/except/else/finally

2. Functions

14. Prefer Exceptions to Returning None

15. Know How Closures Interact with Variable Scope

16. Consider Generators Instead of Returning Lists

17. Be Defensive When Iterating Over Arguments

18. Reduce Visual Noise with Variable Positional Arguments

19. Provide Optional Behavior with Keyword Arguments

20. Use None and Docstrings to Specify Dynamic Default Arguments

21. Enforce Clarity with Keyword-Only Arguments

3. Classes and Inheritance

22. Prefer Helper Classes Over Bookkeeping with Dictionaries and Tuples

23. Accept Functions for Simple Interfaces Instead of Classes

24. Use @classmethod Polymorphism to Construct Objects Generically

25. Initialize Parent Classes with super

26. Use Multiple Inheritance Only for Mix-in Utility Classes

27. Prefer Public Attributes Over Private Ones

28. Inherit from for Custom Container Types

4. Metaclasses and Attributes

29. Use Plain Attributes Instead of Get and Set Methods

30. Consider @property Instead of Refactoring Attributes

31. Use Descriptors for Reusable @property Methods

32. Use __getattr__, __getattribute__, and __setattr__ for Lazy Attributes

33. Validate Subclasses with Metaclasses

34. Register Class Existence with Metaclasses

35. Annotate Class Attributes with Metaclasses

5. Concurrency and Parallelism

36. Use subprocess to Manage Child Processes

37. Use Threads for Blocking I/O, Avoid for Parallelism

38. Use Lock to Prevent Data Races in Threads

39. Use Queue to Coordinate Work Between Threads

40. Consider Coroutines to Run Many Functions Concurrently

41. Consider concurrent.futures for True Parallelism

6. Built-in Modules

42. Define Function Decorators with functools.wraps

43. Consider contextlib and with Statements for Reusable try/finally Behavior

44. Make pickle Reliable with copyreg

45. Use datetime Instead of time for Local Clocks

46. Use Built-in Algorithms and Data Structures

47. Use decimal When Precision Is Paramount

48. Know Where to Find Community-Built Modules

7. Collaboration

49. Write Docstrings for Every Function, Class, and Module

50. Use Packages to Organize Modules and Provide Stable APIs

51. Define a Root Exception to Insulate Callers from APIs

52. Know How to Break Circular Dependencies

53. Use Virtual Environments for Isolated and Reproducible Dependencies

8. Production

54. Consider Module-Scoped Code to Configure Deployment Environments

55. Use repr String for Debugging Output

56. Test Everything with unittest

57. Consider Interactive Debugging with pdb

58. Profile Before Optimizing

59. Use tracemalloc to Understand Memory Usage and Leaks