Added pdf download

emission_forecast.py

@@ -548,7 +548,10 @@ def register_emission_routes(app):
                     applied_policies.append({
                         'id': policy['id'],
                         'name': policy['name'],
-                        'icon': policy['icon']
+                        'icon': policy['icon'],
+                        'description': policy.get('description', ''),
+                        'category': policy.get('category', 'General'),
+                        'details': policy.get('details', {})
                     })
             
             # Calculate yearly totals
@@ -745,6 +748,42 @@ Data Points: {len(year_df)} days
             import traceback
             traceback.print_exc()
             return jsonify({'error': str(e)}), 500
+            
+    @app.route('/api/policies/chat', methods=['POST'])
+    def policy_chat():
+        """
+        Chat with Gemini about selected policies.
+        """
+        try:
+            from policy_chat import PolicyChat
+            from policies_data import get_policy_by_id
+            
+            data = request.json or {}
+            policy_ids = data.get('policy_ids', [])
+            question = data.get('question', '')
+            
+            if not question:
+                return jsonify({'error': 'Question is required'}), 400
+                
+            # Get full policy details
+            policies = []
+            for pid in policy_ids:
+                p = get_policy_by_id(pid)
+                if p:
+                    policies.append(p)
+            
+            chat_engine = PolicyChat()
+            answer = chat_engine.chat(policies, question)
+            
+            return jsonify({
+                'status': 'success',
+                'answer': answer,
+                'timestamp': datetime.now().isoformat()
+            })
+        except Exception as e:
+            import traceback
+            traceback.print_exc()
+            return jsonify({'error': str(e)}), 500
 
 if __name__ == "__main__":
     print("Training XGBoost Emission Model...")

policy_chat.py

@@ -0,0 +1,55 @@
+from langchain_google_genai import ChatGoogleGenerativeAI
+from langchain_google_genai import ChatGoogleGenerativeAI
+from langchain_core.prompts import ChatPromptTemplate
+from langchain_core.runnables import RunnablePassthrough
+from langchain_core.output_parsers import StrOutputParser
+import config
+
+class PolicyChat:
+    def __init__(self):
+        if not config.GEMINI_API_KEY:
+            print("WARNING: GEMINI_API_KEY is not set for PolicyChat.")
+        
+        self.llm = ChatGoogleGenerativeAI(
+            model="gemini-2.5-flash",
+            temperature=0.5,
+            google_api_key=config.GEMINI_API_KEY
+        )
+
+        self.prompt = ChatPromptTemplate.from_template("""
+        You are an AI policy expert for an Urban Digital Twin platform.
+        Your goal is to explain specific urban policies to stakeholders (city planners, citizens).
+
+        The user is asking about the following policies which are currently selected in the simulator:
+        {policy_context}
+
+        Context about these policies:
+        - These are used to reduce CO2 emissions and improve Air Quality (AQI).
+        - The user may ask about implementation details, costs, side effects, or effectiveness.
+
+        User Question: {question}
+
+        Provide a clear, concise, and helpful answer. validation: If the user asks about something unrelated to the policies or urban planning, politely steer them back to the topic.
+        """)
+
+    def chat(self, policies, question):
+        """
+        policies: List of policy dictionaries (name, description, etc.)
+        question: User's question string
+        """
+        # Format policy context
+        policy_context = ""
+        for p in policies:
+            policy_context += f"- {p.get('name')} ({p.get('id')}): {p.get('description')}\n"
+            if 'details' in p:
+                details = p['details']
+                policy_context += f"  Cost: {details.get('implementation_cost')}, Public Acceptance: {details.get('public_acceptance')}\n"
+
+        chain = (
+            {"policy_context": lambda x: policy_context, "question": lambda x: question}
+            | self.prompt
+            | self.llm
+            | StrOutputParser()
+        )
+
+        return chain.invoke({})