data/build_knowledge_base.py

import json
import os
import sys

sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
import config

def generate_documents():
    print(f"Loading schemes from {config.SCHEMES_DB_PATH}...")
    with open(config.SCHEMES_DB_PATH, "r", encoding="utf-8") as f:
        schemes = json.load(f)
        
    os.makedirs(config.SCHEME_DOCS_PATH, exist_ok=True)
    
    for scheme in schemes:
        doc_content = f"""Scheme Name: {scheme['name']} ({scheme.get('name_hi', '')})
Ministry: {scheme['ministry']}
Category: {scheme['category']}
Launched: {scheme.get('launched', 'N/A')}

Description:
{scheme['description']}

Benefits:
- {chr(10) + '- '.join(scheme.get('benefits', []))}

Eligibility Criteria:
"""
        eligibility = scheme.get('eligibility', {})
        for k, v in eligibility.items():
            if v is not None:
                if isinstance(v, list):
                    doc_content += f"- {k}: {', '.join(v)}\n"
                else:
                    doc_content += f"- {k}: {v}\n"
                    
        doc_content += f"""
Application Process:
{scheme.get('application_process', '')}

Documents Required:
- {chr(10) + '- '.join(scheme.get('documents_required', []))}

Official URL: {scheme.get('official_url', 'N/A')}
"""
        
        doc_path = os.path.join(config.SCHEME_DOCS_PATH, f"{scheme['id']}.txt")
        with open(doc_path, "w", encoding="utf-8") as out:
            out.write(doc_content)
            
    print(f"Generated {len(schemes)} documents in {config.SCHEME_DOCS_PATH}")

if __name__ == "__main__":
    generate_documents()