{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "eb4d98e6",
   "metadata": {},
   "source": [
    "### In this notebook we will use Valhalla to snap the latitude/longitude points from VED to roads using OpenStreetMap data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "d99459b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import requests\n",
    "import json\n",
    "from tqdm.notebook import tqdm"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b3e0b1f",
   "metadata": {},
   "source": [
    "Create the headers and the meili request body."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "20178448",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_meili_request_body(df_trip):\n",
    "    \"\"\"Returns the body for meili for its request.\n",
    "    \"\"\"\n",
    "    coordinates = df_trip.to_json(orient='records')\n",
    "    head = '{\"shape\":'\n",
    "    tail = \"\"\",\"search_radius\": 5, \"shape_match\":\"map_snap\", \"costing\":\"auto\", \"format\":\"osrm\"}\"\"\"\n",
    "    body = head + coordinates + tail\n",
    "    return body"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "357a90c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "headers = {'Content-type': 'application/json'}\n",
    "trace_url = \"http://localhost:8002/trace_attributes\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "154c0f88",
   "metadata": {},
   "source": [
    "Read all the file names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "f1097900",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "da818fc0e91c42fcb09ca4b0fadd0d45",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/54 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data_path = \"../data/ved/dynamic/\"\n",
    "files = [os.path.join(data_path, file) for file in tqdm(os.listdir(data_path)) if file.endswith(\".csv\")]\n",
    "files.sort()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f9b0acf",
   "metadata": {},
   "source": [
    "The functions that send the requests to the meili\n",
    "Match types: {'matched': 0, 'interpolated': 1, 'unmatched': 2}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "91bffff4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def process_matched_response(matched_points, group_df):\n",
    "    \"\"\" Returns a list of dictionaries for the matched coordinates.\n",
    "        Dictionary keys: matched_lat, matched_lon, and match_type. \n",
    "        If a points was not matched, returns the original coordinates. \"\"\"\n",
    "    \n",
    "    processed_matches = []\n",
    "    for (i, matched_point) in enumerate(matched_points):\n",
    "        match_type = {'matched': 0, 'interpolated': 1, 'unmatched': 2}\n",
    "        \n",
    "        matched_dict = {'matched_lat': group_df.iloc[i, 1] \n",
    "                        if matched_point['type'] == 'unmatched' \n",
    "                        else matched_point['lat'], \n",
    "                        'matched_lon': group_df.iloc[i, 0] \n",
    "                        if matched_point['type'] == 'unmatched' \n",
    "                        else matched_point['lon'], \n",
    "                        'match_type': match_type[matched_point['type']]}\n",
    "\n",
    "        processed_matches.append(matched_dict)\n",
    "    \n",
    "    return processed_matches\n",
    "      "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "fb36bca3",
   "metadata": {},
   "outputs": [],
   "source": [
    "def map_match(file):\n",
    "    \"\"\" Takes a csv file and sends requests to Valhalla Meili service to snap the coordinates\n",
    "    in the file to roads. \"\"\"\n",
    "    signal_df = pd.read_csv(file)\n",
    "    dataframes = []\n",
    "    \n",
    "    # Group the dataframe by the columns VehId and Trip\n",
    "    for _, group_df in signal_df.groupby(['VehId', 'Trip']):\n",
    "        group = group_df.rename(columns={'Longitude[deg]':'lon','Latitude[deg]':'lat'})[['lon', 'lat']]\n",
    "        meili_request_body = create_meili_request_body(group)\n",
    "        r = requests.post(trace_url, data=str(meili_request_body), headers=headers)\n",
    "        \n",
    "        if r.status_code == 200:\n",
    "            res = r.json()\n",
    "            matched_dicts = process_matched_response(res['matched_points'], group)\n",
    "            matched_df = pd.DataFrame.from_dict(matched_dicts)\n",
    "            dataframes.append(matched_df.set_index(group_df.index.copy()))\n",
    "            \n",
    "        else:\n",
    "            print('Error', r.status_code)\n",
    "\n",
    "    conc_dfs = pd.concat(dataframes)\n",
    "    signal_df = pd.merge(signal_df, conc_dfs, left_index=True, right_index=True)\n",
    "    return signal_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1d6be11d",
   "metadata": {},
   "source": [
    "Process all files and save the results to another directory."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d1466e8d",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "18f1e3506e9d40eeb9f8966359fd6065",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/54 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def get_save_path(filename):\n",
    "    return f\"../data/ved-valhalla/{filename[len(data_path)-1:]}\"\n",
    "\n",
    "for file in tqdm(files):\n",
    "    matched_df = map_match(file)\n",
    "    matched_df.to_csv(get_save_path(file), na_rep=np.nan, index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7865645b",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}