Qwerty-ftw — FRA Decision Support System

A Flask-based Decision Support System (DSS) for the Forest Rights Act (FRA) in India, focused on the Baran district of Rajasthan. The platform helps manage land rights claims by forest-dwelling communities, analyzes them against geospatial data, and determines eligibility for government schemes.

Stack

• Backend: Python 3, Flask, Flask-RESTful, Flask-SQLAlchemy, Flask-SocketIO, Flask-JWT-Extended
• Database: SQLite
• External APIs: Bhuvan (NRSC/ISRO) — LULC data & WMS groundwater layers; Groq — AI chatbot
• Real-time: WebSocket via Flask-SocketIO
• Other: UTM coordinate conversion, Gunicorn

Project Structure

Qwerty-ftw/
├── main.py                          # Flask app — routes, models, auth, file handling, WebSocket
├── dss.py                           # Decision Support System — LULC, eligibility, groundwater
├── chatBot.py                       # Chatbot conversation state manager
├── test.py                          # Bulk upload test script
├── requirements.txt                 # Python dependencies
├── .gitignore
├── baran_location_hierarchy.csv     # Administration hierarchy: District → Tehsil → RI → Halka → Village
├── instance/
│   └── fra.db                       # SQLite DB with FRA claims data
└── uploads/                         # Uploaded documents and proof images

Features

1. Role-Based Authentication

Three user roles, each registering/logging in at their own endpoint:

Role	Endpoint
DLC	/register/dlc
SDLC	/register/sdlc
Gram Sabha	/register/gram_sabha

Login returns a JWT token with the role embedded in its claims.

2. FRA Claims Management

• Add a claim — POST /claims
• List all claims — GET /claims/all
• Claims by district — GET /claims/district/<district_name>
• Upload document — POST /upload (saves to uploads/{claim_id}/)
• Upload proof image — POST /upload_proof (saves to uploads/{claim_id}/proofs/)
• Retrieve proofs — GET /get_proofs/<claim_id>
• Bulk upload — POST /bulkupload (accepts a ZIP, extracts to uploads/bulk_uploads/)

3. Decision Support System (DSS)

Land Use / Land Cover (LULC)

• District LULC — GET /lulc/<distcode> — fetches land-use statistics from Bhuvan API for a district
• AOI LULC — POST /lulc/aoi — LULC stats for an arbitrary polygon (Area of Interest)

Scheme Eligibility

Evaluates each claim against 5 government schemes using LULC and claim metadata:

Scheme	Criteria
PM-KISAN	Purpose is agriculture AND land area ≤ 2 hectares
Jal Jeevan Mission	Water body percentage < 50%
Van Dhan Yojana	Caste is ST AND forest cover > 20%
MGNREGA	Land area < 1 ha AND soil quality < 60%
DAJGUA	Purpose is agriculture/dairy AND soil quality > 70%

Endpoints:

• Single claim eligibility — GET /eligibility/<claim_id>
• District eligibility summary — GET /eligibility/summary/<district>
• AOI detailed summary — POST /aoi-dss-summary (claimant list + eligibility + LULC for a polygon)
• AOI scheme counts — POST /aoi-scheme-summary (aggregated eligible counts per scheme)

Groundwater Properties

• LGeom lookup — GET/POST /lgeom?x=<lat>&y=<lon> — queries the Bhuvan WMS RJ_LGEOM layer and returns geological formation info with a groundwater potential rating (High / Moderate / Low).

4. AI Chatbot

A WebSocket-based chatbot accessible at the root of the app. Uses Groq API (openai/gpt-oss-20b model) with a system prompt that grounds it as an FRA DSS assistant. Maintains per-session conversation history on the server.

• Connect → receives connected event
• Send message → message event; receives response event with AI reply (async, threaded)

5. Static Data

baran_location_hierarchy.csv contains the full administrative hierarchy of Baran district: District → Tehsil → RI (Circle) → Halka → Village. This powers location-based filtering of claims.

Setup & Running

1. Install dependencies

   pip install -r requirements.txt

2. Set environment variables

   Create a .env file in the project root:

   APIKEY=your_groq_api_key
   

3. Run the app

   python main.py

   Starts on http://0.0.0.0:3000 with debug mode.

API Overview

Method	Endpoint	Description
POST	/register/<role>	Register (dlc/sdlc/gram_sabha)
POST	/login/<role>	Login, returns JWT
POST	/claims	Add a new FRA claim
GET	/claims/all	List all claims
GET	/claims/district/<district>	Claims filtered by district
GET	/lulc/<distcode>	District LULC statistics
POST	/lulc/aoi	AOI LULC statistics
GET	/eligibility/<claim_id>	Scheme eligibility for a claim
GET	/eligibility/summary/<district>	District-level eligibility summary
POST	/aoi-dss-summary	AOI claim details + eligibility
POST	/aoi-scheme-summary	AOI scheme counts
GET	/lgeom	Groundwater properties at coordinate
POST	/lgeom	Groundwater properties at coordinate
POST	/upload	Upload claim document
POST	/upload_proof	Upload proof image
GET	/get_proofs/<claim_id>	Get proof images for a claim
POST	/bulkupload	Bulk upload ZIP

Notes

• Security: API tokens and JWT secret are currently hardcoded. JWT secret should move to .env and passwords should be hashed for production.
• Database: instance/fra.db is included with sample data for Baran district. The app auto-creates tables on startup.
• Languages: Claim data uses Hindi (Devanagari) for village names, purposes, and caste statuses. The eligibility rules handle both Hindi and English values.