Project Samarth — Enhanced Prototype

This enhanced package adds live data fetching from data.gov.in (with local CSV fallback), robust crop/state matching (including fuzzy matching), cross-state district comparison, policy advisory scaffolding, and improved trend/correlation analysis.

Key files added

• config.py — Put your API keys and resource IDs here.
• data_fetcher.py — Fetch datasets from data.gov.in with local caching and fallback.
• matching.py — Name cleaning and fuzzy matching utilities.
• analysis.py — Trend analysis, growth rates, correlation utilities.
• enhanced_handler.py — An upgraded handle_question() skeleton tying everything together.
• requirements.txt — Suggested Python packages.
• example_run.py — Small script that shows how to call the main handler.
• CHANGELOG.md — Notes about what was added.

Usage

1. Install dependencies:

   pip install -r requirements.txt

2. Edit config.py to include your API key and resource IDs (already prefilled).
3. Run python example_run.py to see an example query flow.
4. Integrate enhanced_handler.handle_question into your existing codebase — it falls back to local CSVs if live fetching is unavailable.

Notes

• This environment couldn't call the API while packaging; the code is ready to run locally or on a server with internet access.
• Caching folder: ./cache/ — fetched CSVs are stored there to avoid repeated downloads.