src/main.py

# main.py
import asyncio
from pydantic import BaseModel
from fastapi import FastAPI
from src.chatbot.engine import call_llm
from fastapi.middleware.cors import CORSMiddleware
import logging
import uuid

# Logging setup
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger("ylf-api")

app = FastAPI(title="YLF AI Platform")

# Allows the frontend to talk to backend
app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"],
    allow_methods=["*"],
    allow_headers=["*"],
)


@app.get("/")
def home():
    return {"message": "API is running 🚀"}


class ChatRequest(BaseModel):
    message: str
    mode: str = "socratic"
    # Optional session_id; if not provided, a new one will be generated
    session_id: str = None


@app.post("/chat")
async def chat_endpoint(request: ChatRequest):
    # Optimize session_id generation to be compatible with Engine logs
    session_id = request.session_id or f"user_{uuid.uuid4().hex[:8]}"

    try:
        # Execute the enhanced Engine (contains Fallback and Smart Routing logic)
        # Using to_thread because the Engine performs blocking HTTP requests
        answer = await asyncio.to_thread(
            call_llm,
            user_query=request.message,
            mode=request.mode,
            session_id=session_id
        )

        # Critical check: Verify if the Engine returned a total failure message
        if "All models failed" in answer:
            logger.error(f"Critical Engine Failure: {answer}")
            return {
                "answer": "Sorry, our AI engines are currently under heavy load. Please try again in a minute.",
                "session_id": session_id,
                "status": "temporary_failure"
            }

        return {"answer": answer, "session_id": session_id, "status": "success"}

    except Exception as e:
        logger.error(f"API Logic Error: {str(e)}")
        return {"error": "Internal API Error", "session_id": session_id}