DocCreator

DocCreator is a tool designed to simplify code documentation and streamline onboarding for new contributors in open-source GitHub repositories. It leverages GPT API and LangChain to provide an intuitive interface for generating comprehensive documentation and interactive guidance. Whether you're a beginner navigating a complex codebase or a project maintainer looking to enhance collaboration, CodeDocFlow aims to make the process more accessible and efficient for everyone involved.

Overview

• app.py: Entry point of the tool
• code_parser.py: Contains logic to create AST for the provided python code
• assembler.py: Contains logic to assemble the distributed docs created by LLM
• llm_interface.py: Contains logic for Code to md doc
• embedding_generator.py: Contains logic to generate embedding for the code if required

Limitation

Exploring the limitation of this application. Encouraging contributions to resolve the limitations.

Code Parser

Here are the notable limitations along with a gist of resolution of limitations in the code parser logic (refer code_parser.py) file:

1. Limited Node Handling:

   • Limitation: The parser handles a basic set of AST nodes, excluding constructs like decorators.
   • Resolution: Extend the parser to include additional AST node types as required, ensuring a more comprehensive parsing of the Python code.

2. Simplified Node Representation:

   • Limitation: Node representations are basic and might not capture complex structures effectively.
   • Resolution: Enhance the _get_node_repr method to handle complex expressions and structures, providing a more detailed representation of the code.

3. Comment Parsing:

   • Limitation: Python's AST does not include comments, and they are currently not parsed.
   • Resolution: Implement a separate mechanism for comment extraction, such as using tokenization or regex-based methods alongside AST parsing.

4. Error Handling:

   • Limitation: The parser lacks comprehensive error handling for unexpected code structures or syntax errors.
   • Resolution: Add robust error handling and logging to manage and debug issues with parsing different types of code.

5. Scalability:

   • Limitation: Performance and memory usage are not optimized for large code bases.
   • Resolution: Optimize the parser for efficiency, possibly by implementing lazy loading or parallel processing techniques for handling large repositories.

Contribution

I encourage you to contribute to this repository or create issue if you feel so, I am open to feedbacks and suggestion upon how to make this a better and more efficient project. For any queries, feel free to reach out at @SaifSaifee_dev.