# test script to generate sensor data for testing
# made to reprsent real world data

from contextlib import asynccontextmanager
from fastapi import FastAPI
from fastapi_mqtt import MQTTConfig, FastMQTT
from gmqtt import Client as MQTTClient
import fastapi_mqtt as mqtt
import uvicorn

import datetime
import logging
import random
import asyncio

mqtt_config = mqtt.MQTTConfig(
    host="localhost",
    port=1883,
    keepalive=60,
)

fast_mqtt = mqtt.FastMQTT(config=mqtt_config)

@asynccontextmanager
async def _lifespan(_app: FastAPI):
    await fast_mqtt.mqtt_startup()
    asyncio.create_task(main())
    yield
    await fast_mqtt.mqtt_shutdown()

app = FastAPI(lifespan=_lifespan)


# weights for hourly temps, based on LKTB sample on nov, 4th 2025
TIME_WEIGHTS = [
    4, 3, 3, 2, 2, 2, 2, 1, 3, 6, 8, 9, 11, 12, 12, 12, 11, 10, 9, 9, 8, 8, 5, 5
]
# weights for specific rooms, here inside/outside
ROOM_WEIGHTS = {
    "outside": 0,
    "inside": 10
}

@fast_mqtt.on_connect()
def connect(client: MQTTClient, flags, rc, properties):
    print("Connected: ", client, flags, rc, properties)

async def publish_sensor_data(inside: bool = True):
    now = datetime.datetime.now()
    base_temp = TIME_WEIGHTS[now.hour]
    temp = base_temp + ROOM_WEIGHTS["inside" if inside else "outside"] + random.gauss(0,0.5)
    logging.info(f"{now.isoformat()},{temp}")
    fast_mqtt.publish("210/temp" if inside else "000/temp", f"{now.isoformat()},{temp}", qos=0)

async def main():
    while True:
        await publish_sensor_data(inside=False)
        await publish_sensor_data(inside=True)
        await asyncio.sleep(300)

if __name__ == "__main__":
    logging.basicConfig(level=logging.INFO)
    uvicorn.run(app)