{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pytz\n",
    "from datetime import datetime\n",
    "import json, csv, time\n",
    "import paho.mqtt.client as mqtt\n",
    "from datetime import datetime, timedelta"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Set the MQTT broker connection details\n",
    "MQTT_BROKER = \"hp.henrypump.cloud\"\n",
    "MQTT_PORT = 1883\n",
    "\n",
    "# Create an MQTT client instance\n",
    "client = mqtt.Client(client_id=\"faskens-bp-compressor\")\n",
    "client.username_pw_set(\"faskensmqtt\", \"faskensmqtt@1903\" )\n",
    "client.connect(MQTT_BROKER, MQTT_PORT)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# Define a function to convert datetime to timestamp and round to nearest 10 minutes\n",
    "def convert_datetime_to_timestamp(dt_str):\n",
    "    cst_tz = pytz.timezone('America/Chicago')\n",
    "    dt_cst = cst_tz.localize(datetime.strptime(dt_str, '%Y-%m-%d/%H:%M:%S'))\n",
    "    \n",
    "    # Calculate the remainder of minutes in the hour\n",
    "    minute_remainder = dt_cst.minute % 10\n",
    "    \n",
    "    if minute_remainder >= 5:\n",
    "        # Round up to the next 10-minute interval\n",
    "        dt_cst += timedelta(minutes=10 - minute_remainder)\n",
    "    else:\n",
    "        # Round down to the previous 10-minute interval\n",
    "        dt_cst -= timedelta(minutes=minute_remainder)\n",
    "    \n",
    "    # Ensure the resulting datetime is valid (no 23:60:00 or similar issues)\n",
    "    while dt_cst.minute % 10 != 0:\n",
    "        if dt_cst.minute > 50:\n",
    "            dt_cst += timedelta(hours=1)\n",
    "            dt_cst -= timedelta(minutes=dt_cst.minute % 60)\n",
    "        else:\n",
    "            dt_cst -= timedelta(minutes=dt_cst.minute % 10)\n",
    "    \n",
    "    return int(dt_cst.timestamp()*1000)\n",
    "\n",
    "# Transform the data and send it to the MQTT broker in chunks of 20\n",
    "transformed_data = []\n",
    "with open('/Users/nico/Downloads/history_data_default.csv', 'r') as csvfile:\n",
    "    reader = csv.DictReader(csvfile)\n",
    "    for row in reader:\n",
    "        dt = row['Time']\n",
    "        value = json.loads(row[' Value'])[\"plcpond\"][\"air_comp_val\"]\n",
    "        if value[0] == 1:\n",
    "            transformed_data.append({\"ts\": convert_datetime_to_timestamp(dt), \"values\": {\"air_comp_val\": value[1]}})\n",
    "\n",
    "\n",
    "\n",
    "chunk_size = 20\n",
    "for i in range(0, len(transformed_data), chunk_size):\n",
    "    chunk = transformed_data[i:i + chunk_size]\n",
    "    for x in chunk:\n",
    "        client.publish(\"v1/devices/me/telemetry\", json.dumps(x))\n",
    "    time.sleep(1)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "client.disconnect()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "aws",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}