backend/app/main.py

import logging
from contextlib import asynccontextmanager

from fastapi import FastAPI
from fastapi.middleware.cors import CORSMiddleware

from app.config import settings
from app.api import chat, ingestion, health

logging.basicConfig(
    level=logging.INFO,
    format="%(asctime)s | %(levelname)-8s | %(name)s | %(message)s",
)
logger = logging.getLogger(__name__)


@asynccontextmanager
async def lifespan(_app: FastAPI):
    # Eagerly load the embedding model and vector store so the first visitor
    # doesn't pay the model-load latency.
    from app.services.embedding_service import get_embedding_service
    from app.services.vector_store import get_vector_store
    from app.services.llm_service import get_llm_service

    get_embedding_service()
    chunks = get_vector_store().count()
    get_llm_service()
    if chunks == 0:
        logger.warning("Vector store is EMPTY — run scripts/ingest.py to build the index")
    logger.info("Warmup complete — %d chunks indexed", chunks)
    yield


app = FastAPI(
    title="Negoptim AI Backend",
    description="RAG-powered chatbot backend for Users Love IT (ULiT)",
    version="1.0.0",
    lifespan=lifespan,
)

app.add_middleware(
    CORSMiddleware,
    allow_origins=settings.cors_origin_list,
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)

app.include_router(health.router, prefix="/api", tags=["health"])
app.include_router(chat.router, prefix="/api", tags=["chat"])
app.include_router(ingestion.router, prefix="/api", tags=["ingestion"])


@app.get("/")
async def root():
    return {"service": "Negoptim AI", "docs": "/docs"}