{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gplace\n",
    "\n",
    "location = \"13.744677,100.5295593\"  # Latitude and Longitude\n",
    "keyword = \"ร้านกาแฟ\"\n",
    "result = gplace.nearby_search(keyword, location)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "from typing import TypedDict, Optional\n",
    "\n",
    "class NearbyDenseCommunityInput(TypedDict):\n",
    "    location_name: str\n",
    "    radius: int"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "def find_place_from_text(location:str):\n",
    "    \"\"\"Finds a place and related data from the query text\"\"\"\n",
    "    \n",
    "    result = gplace.find_place_from_text(location)\n",
    "    r = result['candidates'][0]\n",
    "    return f\"\"\"\n",
    "    address: {r['formatted_address']}\\n\n",
    "    location: {r['geometry']['location']}\\n\n",
    "    name: {r['name']}\\n\n",
    "    opening hours: {r['opening_hours']}\\n\n",
    "    rating: {r['rating']}\\n\n",
    "    \"\"\"\n",
    "    \n",
    "def nearby_search(keyword:str, location:str, radius=2000, place_type=None):\n",
    "    \"\"\"Searches for many places nearby the location based on a keyword. using keyword like \\\"coffee shop\\\", \\\"restaurants\\\". radius is the range to search from the location\"\"\"\n",
    "    location = gplace.find_location(location, radius=radius)\n",
    "    result = gplace.nearby_search(keyword, location, radius)\n",
    "    \n",
    "    strout = \"\"\n",
    "    for r in result:\n",
    "        # Use .get() to handle missing keys\n",
    "        address = r.get('vicinity', 'N/A')\n",
    "        location_info = r.get('geometry', {}).get('location', 'N/A')\n",
    "        name = r.get('name', 'N/A')\n",
    "        opening_hours = r.get('opening_hours', 'N/A')\n",
    "        rating = r.get('rating', 'N/A')\n",
    "        plus_code = r.get('plus_code', {}).get('global_code', 'N/A')\n",
    "        \n",
    "        strout += f\"\"\"\n",
    "        address: {address}\\n\n",
    "        location: {location_info}\\n\n",
    "        name: {name}\\n\n",
    "        opening hours: {opening_hours}\\n\n",
    "        rating: {rating}\\n\n",
    "        plus code: {plus_code}\\n\\n\n",
    "        \"\"\"\n",
    "    return strout\n",
    "\n",
    "def nearby_dense_community(input_dict: NearbyDenseCommunityInput) -> str:\n",
    "    \"\"\" getting nearby dense community such as (community mall, hotel, school, etc), by location name, radius(in meters)\n",
    "    return list of location community nearby, name, community type.\n",
    "    \"\"\"\n",
    "    location = input_dict['location_name']\n",
    "    radius = input_dict['radius']\n",
    "    \n",
    "    location_coords = gplace.find_location(location, radius=radius)\n",
    "    result = gplace.nearby_dense_community(location_coords, radius)\n",
    "    \n",
    "    strout = \"\"\n",
    "    for r in result:\n",
    "        # Use .get() to handle missing keys\n",
    "        address = r.get('vicinity', 'N/A')\n",
    "        location_types = r.get('types', 'N/A')\n",
    "        name = r.get('name', 'N/A')\n",
    "        opening_hours = r.get('opening_hours', 'N/A')\n",
    "        rating = r.get('rating', 'N/A')\n",
    "        plus_code = r.get('plus_code', {}).get('global_code', 'N/A')\n",
    "        \n",
    "        strout += f\"\"\"\n",
    "        name: {name}\\n\n",
    "        types: {location_types}\\n\n",
    "        \"\"\"\n",
    "    return strout\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# gplace_tools.py\n",
    "from langgraph.prebuilt import ToolNode\n",
    "from langchain_core.tools import tool\n",
    "from langchain_core.tools import Tool\n",
    "from langchain_google_community import GoogleSearchAPIWrapper\n",
    "from langchain_community.document_loaders import WebBaseLoader\n",
    "\n",
    "import utils\n",
    "\n",
    "utils.load_env()\n",
    "\n",
    "search = GoogleSearchAPIWrapper()\n",
    "\n",
    "find_place_from_text = tool(find_place_from_text)\n",
    "nearby_search = tool(nearby_search)\n",
    "google_search = Tool(\n",
    "    name=\"google_search\",\n",
    "    description=\"Search Google for recent results.\",\n",
    "    func=search.run,\n",
    ")\n",
    "web_loader = Tool(\n",
    "    name=\"google_search\",\n",
    "    description=\"Search Google for recent results.\",\n",
    "    func=WebBaseLoader,\n",
    ")\n",
    "\n",
    "tools = [find_place_from_text, nearby_search]\n",
    "\n",
    "# Create ToolNodes for each tool\n",
    "tool_node = ToolNode(tools)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\"\\n        name: Bangkok\\n\\n        types: ['locality', 'political']\\n\\n        \\n        name: Metropoint Bangkok Hotel\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: The Grand Fourwings Convention Hotel\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Hua Mak Indoor Stadium\\n\\n        types: ['point_of_interest', 'establishment']\\n\\n        \\n        name: B2 Bangkok Srinagarindra Boutique & Budget Hotel\\n\\n        types: ['clothing_store', 'lodging', 'point_of_interest', 'store', 'establishment']\\n\\n        \\n        name: HappyLand Mansion\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Bangkok Swimming by Kru Jin\\n\\n        types: ['point_of_interest', 'establishment']\\n\\n        \\n        name: Aunchaleena grand Hotel\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Anda Hotel\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Grand Mandarin Residence\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Wallada Place Hotel\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: NIDA Rooms Plubpla Bangkapi 591\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Bangkok Interplace\\n\\n        types: ['lodging', 'restaurant', 'food', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Vejthani Hospital\\n\\n        types: ['hospital', 'doctor', 'point_of_interest', 'health', 'establishment']\\n\\n        \\n        name: โรงแรม ชาลีน่า ปริ้นเซส Chaleena princess\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Royal Pimand\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Baron Residence Hotel\\n\\n        types: ['lodging', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Ridwanun Islam Mosque\\n\\n        types: ['mosque', 'place_of_worship', 'point_of_interest', 'establishment']\\n\\n        \\n        name: Thep Phanom Building\\n\\n        types: ['point_of_interest', 'establishment']\\n\\n        \\n        name: Bang Kapi District\\n\\n        types: ['sublocality_level_1', 'sublocality', 'political']\\n\\n        \""
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nearby_dense_community({'location_name': 'ลุมพินี เซ็นเตอร์ ลาดพร้าว', 'radius': 8000})"
   ]
  }
 ],
 "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}