Coding Standard Python

Python Coding Standards (PEP 8)

When reviewing or generating Python code, follow these rules:

File Naming

• Source files: Use snake_case (e.g., user_service.py, api_client.py)
• Package directories: Use snake_case (e.g., data_processing/, utils/)
• Test files: Use test_ prefix (e.g., test_user_service.py)
• Config files: Use snake_case (e.g., config_settings.py)

Variable Naming

• Variables: snake_case (e.g., user_name, is_active, total_count)
• Constants: UPPER_SNAKE_CASE (e.g., MAX_RETRIES, API_BASE_URL)
• Boolean variables: Prefix with is_, has_, can_, should_ (e.g., is_loading, has_error)
• Protected variables: Single underscore prefix (e.g., _internal_data)
• Private variables: Double underscore prefix (e.g., __private_data)

Function Naming

• Functions: snake_case (e.g., calculate_total(), fetch_user_data())
• Private functions: Prefix with underscore (e.g., _validate_input(), _process_data())
• Dunder methods: Double underscores (e.g., __init__, __str__, __repr__)

Class Naming

• Classes: PascalCase (e.g., UserService, DataProcessor, ApiClient)
• Exception classes: PascalCase with Error or Exception suffix (e.g., ValidationError)
• Abstract classes: PascalCase, optionally prefix with Base or Abstract (e.g., BaseHandler)

Method Naming

• Instance methods: snake_case (e.g., get_user(), process_data())
• Class methods: snake_case with @classmethod decorator
• Static methods: snake_case with @staticmethod decorator
• Properties: snake_case with @property decorator

Module Organization

• Imports at the top: standard library, third-party, local imports (separated by blank lines)
• Module-level dunder names after imports (__all__, __version__)
• One class per file for large classes; multiple related classes okay for small ones
• Use __all__ to define public API