python

Idiomatic Python favors readability, comprehensions for simple transforms, context managers for resource cleanup, and duck typing over inheritance hierarchies. Use truthiness checks like `if not seq` instead of `len(seq) == 0`, and let a formatter like ruff or black handle style enforcement.

Never use a mutable object like a list as a default parameter value, since it is shared across all calls. Use `None` as a sentinel and create the list inside the function body instead.

Use @dataclass for mutable data with behavior and @dataclass(frozen=True) for immutable value objects that are hashable. NamedTuple fits lightweight immutable records needing tuple compatibility, while TypedDict suits typed dictionary shapes from JSON APIs.

Ruff provides both linting and formatting in one fast tool, while black is a dedicated opinionated formatter. Pick one formatter project-wide and enforce it in CI rather than mixing tools.

isinstance(x, int) respects inheritance, so subclasses pass the check, while type(x) == int only matches the exact class. This makes isinstance the safer and more flexible choice for type checking.

Use match/case in Python 3.10+ when destructuring a single subject by structure, type, or value, such as parsing command dictionaries or class instances. Long isinstance chains are a strong signal to switch to structural pattern matching.