# Python Code Review

Review {{input}} for code quality, correctness, and best practices.

## Types and Style
- Add type hints on all function signatures (PEP 484); use `from __future__ import annotations` for forward refs
- Follow PEP 8: snake_case for functions/variables, PascalCase for classes, UPPER_CASE for constants
- Avoid mutable default arguments (`def f(x=[])` is a classic bug — use `None` and set inside)
- Prefer f-strings over `.format()` or `%` formatting

## Error Handling
- Catch specific exceptions, not bare `except:` or `except Exception:`
- Use context managers (`with`) for all file I/O, locks, and DB connections
- Never suppress exceptions silently; at minimum log them with context
- Distinguish between recoverable errors (return/raise) and programming errors (assert)

## Reliability
- Handle edge cases: empty sequences, None inputs, zero denominators, empty strings
- Avoid shared mutable state in module scope
- For I/O-bound work prefer async or threading; for CPU-bound work prefer multiprocessing
- Validate all config and environment variables at startup

## Security
- No secrets or PII in logs, exception messages, or tracebacks exposed to users
- Parameterise all DB queries — no f-string interpolation in SQL
- Sanitise file paths; avoid `os.system()` / `subprocess` with user-controlled strings
- Avoid `pickle` for untrusted data; prefer `json` or `msgpack`

## Code Quality
- Functions should do one thing; < 30 lines each
- No dead code, `# TODO` left unaddressed, or commented-out blocks
- Tests use pytest; cover happy path and key error conditions

---
List findings as **Critical** (security/data loss/crashes) -> **Major** (bugs/wrong behaviour) -> **Minor** (style/improvements). Cite each location and suggest the concrete fix.