kpmuks2709/cobol-rag-lineage

COBOL RAG Lineage Q&A System

A RAG (Retrieval-Augmented Generation) system for querying COBOL codebase using natural language.

Setup

1. Install Dependencies

pip install -r requirements.txt

2. Set OpenAI API Key

export OPENAI_API_KEY='your-api-key-here'

Or on Windows:

set OPENAI_API_KEY=your-api-key-here

3. Initialize the System (First Time Only)

python RAG_lineage_POC_v2.py

This will:

• —Generate descriptions for all COBOL files
• —Create embeddings and FAISS index
• —Save metadata

Note: This step may take several minutes depending on the number of COBOL files.

Running the Application

Quick Start (Recommended)

On Mac/Linux:

./run.sh

On Windows:

run.bat

Or manually:

python app.py

Access Options

1. 1.Local Access: http://localhost:7860 (only on your machine)

1. 1.Public Shareable Link: The app automatically creates a public link when you run it. You'll see output like:

   Running on local URL:  http://127.0.0.1:7860
   Running on public URL: https://xxxxx.gradio.live

Share the public URL with others to give them access. The link is active as long as your server is running.

Usage

1. 1.Initialize System (if not already done):
2. 2.Go to "Initialize System" tab
3. 3.Click "Initialize RAG System" button
4. 4.Wait for completion

1. 1.Query COBOL Code:
2. 2.Go to "Query COBOL Code" tab
3. 3.Enter your question
4. 4.Click "Submit Query"
5. 5.View the response

Example Queries

• —"What are the input and output variables in transaction processing programs?"
• —"Which files handle user authentication?"
• —"Show me files that process account data"
• —"What are the main functions in the COBOL codebase?"

Project Structure

Lineage_RAG_POC_V2/
├── app.py                      # Gradio web interface
├── RAG_lineage_POC_v2.py      # Main RAG logic
├── requirements.txt            # Python dependencies
├── cobol_files/               # COBOL source files
├── faiss_cobol_v2.index       # FAISS vector index (generated)
├── cobol_metadata_v2.json     # Metadata (generated)
└── Cobol_code_desc_v2.csv     # Descriptions (generated)

Troubleshooting

• —Import Error: Make sure all dependencies are installed: pip install -r requirements.txt
• —FAISS Index Not Found: Run initialization first: python RAG_lineage_POC_v2.py
• —API Key Error: Ensure OPENAI_API_KEY environment variable is set
• —Port Already in Use: Change the port in app.py: server_port=7861