---

name: python-style description: Python coding style guide and best practices. Use when writing Python code, reviewing Python files, or checking code style. Triggers on "python style", "python code", "파이썬 스타일", "코드 스타일".

Python Style Guide

This skill provides Python coding style guidelines and best practices.

Core Principles

• Follow PEP 8 (line length: 100 chars)
• Use type hints for arguments and return values
• Use Google-style docstrings
• Prefer f-strings, list/dict comprehensions, context managers
• Prefer pathlib over os.path
• Use dataclasses or pydantic for data structures

Type Hints

Copy# Good
def process_data(items: list[str], threshold: float = 0.5) -> dict[str, int]:
    ...

# Bad
def process_data(items, threshold=0.5):
    ...

Docstrings (Google Style)

Copydef fetch_data(source_id: str, limit: int = 100) -> list[dict]:
    """Fetch data from the specified source.

    Args:
        source_id: The unique identifier for the data source.
        limit: Maximum number of records to fetch.

    Returns:
        A list of dictionaries containing the fetched records.

    Raises:
        ValueError: If source_id is empty.
        ConnectionError: If unable to connect to the source.
    """
    ...

Preferred Patterns

F-strings over format()

Copy# Good
message = f"Processing {count} items for user {user_id}"

# Bad
message = "Processing {} items for user {}".format(count, user_id)

List/Dict Comprehensions

Copy# Good
squares = [x**2 for x in range(10)]
user_map = {user.id: user for user in users}

# Bad
squares = []
for x in range(10):
    squares.append(x**2)

Context Managers

Copy# Good
with open(file_path) as f:
    content = f.read()

# Bad
f = open(file_path)
content = f.read()
f.close()

Pathlib over os.path

Copyfrom pathlib import Path

# Good
config_path = Path(__file__).parent / "config" / "settings.yaml"

# Bad
import os
config_path = os.path.join(os.path.dirname(__file__), "config", "settings.yaml")

Data Structures

Use Dataclasses

Copyfrom dataclasses import dataclass

@dataclass
class User:
    id: str
    name: str
    email: str
    is_active: bool = True

Use Pydantic for Validation

Copyfrom pydantic import BaseModel, EmailStr

class UserCreate(BaseModel):
    name: str
    email: EmailStr
    age: int | None = None

Common Commands

Linting and Formatting (ruff)

Copy# Check for issues
ruff check src/ tests/

# Auto-fix issues
ruff check --fix src/ tests/

# Format code
ruff format src/ tests/

Running Tests (pytest)

Copy# Run all tests
pytest tests/ -v

# Run specific test file
pytest tests/test_module.py -v

# Run with coverage
pytest tests/ --cov=src --cov-report=term-missing

Import Organization

Copy# Standard library
import os
from datetime import datetime
from pathlib import Path

# Third-party
import pandas as pd
from pydantic import BaseModel

# Local
from src.module import function
from src.utils import helper

Error Handling

Copy# Good - specific exceptions with context
try:
    result = fetch_data(source_id)
except ConnectionError as e:
    logger.error(f"Failed to connect to source {source_id}: {e}")
    raise
except ValueError as e:
    logger.warning(f"Invalid data from source {source_id}: {e}")
    return None

# Bad - generic exception
try:
    result = fetch_data(source_id)
except Exception:
    pass

Logging

Copyimport logging

logger = logging.getLogger(__name__)

# Good - structured logging with context
logger.info("Processing started", extra={"source_id": source_id, "count": len(items)})
logger.error(f"Failed to process item {item_id}: {error}", exc_info=True)

# Bad - print statements
print(f"Processing {source_id}")