Python Typing Generator

Purpose & User Intent: The tool analyzes Python source and docstrings to generate precise typing hints. It targets developers, library maintainers, data engineers, and teams migrating to typing who need consistent type contracts across large codebases. Use during refactoring, API surface design, or onboarding to standardize typing.

Core Logic & Features: Required features include AST-based inference from function definitions, parameters, return types, and docstrings; support for PEP 484 typing constructs (List, Dict, Optional, Union, Callable); export formats for inline annotations or standalone stub files (.pyi); basic TypeVar and generics support; cross-module resolution to reduce unresolved imports. Optional features include configurable inference depth, integration with existing type hints, and type summary generation. The logic favors explicit annotations when present, otherwise infers probable types and falls back to Any or Optional for indeterminate cases. Outputs can be injected into code or delivered as patch-ready stubs. A validation pass checks typing consistency and reports potential conflicts.

Inputs & Outputs: Inputs are a Python code snippet or file path, a mode selector (inline vs stub), and optional guidance on inference depth. Outputs include annotated code blocks or .pyi files, formatted to align with PEP 8 and typing standards. The tool also emits a concise type summary, a list of inferred types, and warnings for unresolved imports or dynamic constructs.

Algorithms & Calculations: The tool parses the AST, collects symbol definitions, and derives parameter and return types from signatures and docstrings. It translates docstring types to typing constructs, resolves imports, and propagates inferred types across modules. The final outputs are either direct edits of the source or separate stub declarations. Edge cases include dynamic attributes, reflection, or heavy metaprogramming where typing is uncertain; in such cases the tool signals conservative typings and recommends explicit annotations.

Industry/Region & Localization; Assumptions & Exclusions: Follows standard Python typing conventions (PEP 484) and US/EU coding norms; no locale-specific formatting is required. Assumes valid Python input, does not execute code, and excludes runtime refactoring or full static analysis beyond typing guidance.