{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# vector data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Import geometry data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install h3-toolkit, geopandas, polars, psycopg2-binary, sqlalchemy, geoalchemy2, connectorx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "import geopandas as gpd\n",
    "import polars as pl\n",
    "from dotenv import load_dotenv\n",
    "from sqlalchemy import create_engine\n",
    "\n",
    "load_dotenv()\n",
    "URI = os.getenv(\"POSTGRES\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "### 1.1. Use `geopandas` to fetch vector data from Postgis (slowly)  \n",
    "*CSL researcher only*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/syuanbo/Documents/GitHub/H3-ToolKits/.venv/lib/python3.12/site-packages/geopandas/io/sql.py:185: UserWarning: pandas only supports SQLAlchemy connectable (engine/connection) or database string URI or sqlite3 DBAPI2 connection. Other DBAPI2 objects are not tested. Please consider using SQLAlchemy.\n",
      "  df = pd.read_sql(\n",
      "/Users/syuanbo/Documents/GitHub/H3-ToolKits/.venv/lib/python3.12/site-packages/geopandas/io/sql.py:473: UserWarning: pandas only supports SQLAlchemy connectable (engine/connection) or database string URI or sqlite3 DBAPI2 connection. Other DBAPI2 objects are not tested. Please consider using SQLAlchemy.\n",
      "  return pd.read_sql(spatial_ref_sys_sql, con)\n"
     ]
    }
   ],
   "source": [
    "sql = \"\"\"\n",
    "with boundary as (\n",
    "    select 代碼, geometry\n",
    "    from geometry.boundary_smallest\n",
    "    where 縣市代碼 = %(city_code)s\n",
    ")\n",
    "select aps.*, boundary.geometry as geometry\n",
    "from geometry.af_ppl_stats as aps\n",
    "join boundary\n",
    "on aps.codebase = boundary.代碼\n",
    "\"\"\"\n",
    "\n",
    "engine = create_engine(URI)\n",
    "with engine.connect() as conn:\n",
    "    gdf = gpd.read_postgis(\n",
    "        sql,\n",
    "        con=conn.connection,\n",
    "        geom_col='geometry',\n",
    "        params={'city_code': '63000'} # 台北市\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>codebase</th>\n",
       "      <th>city</th>\n",
       "      <th>h_cnt</th>\n",
       "      <th>p_cnt</th>\n",
       "      <th>m_cnt</th>\n",
       "      <th>f_cnt</th>\n",
       "      <th>info_time</th>\n",
       "      <th>geometry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A6301-0097-00</td>\n",
       "      <td>臺北市</td>\n",
       "      <td>40</td>\n",
       "      <td>121</td>\n",
       "      <td>53</td>\n",
       "      <td>68</td>\n",
       "      <td>2023-12-01 00:00:00+00:00</td>\n",
       "      <td>MULTIPOLYGON (((121.54425 25.06221, 121.54466 ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>A6303-0422-00</td>\n",
       "      <td>臺北市</td>\n",
       "      <td>146</td>\n",
       "      <td>376</td>\n",
       "      <td>165</td>\n",
       "      <td>211</td>\n",
       "      <td>2023-12-01 00:00:00+00:00</td>\n",
       "      <td>MULTIPOLYGON (((121.55701 25.03439, 121.5575 2...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>A6303-1146-00</td>\n",
       "      <td>臺北市</td>\n",
       "      <td>93</td>\n",
       "      <td>229</td>\n",
       "      <td>110</td>\n",
       "      <td>119</td>\n",
       "      <td>2023-12-01 00:00:00+00:00</td>\n",
       "      <td>MULTIPOLYGON (((121.55202 25.02088, 121.55205 ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>A6311-0203-00</td>\n",
       "      <td>臺北市</td>\n",
       "      <td>58</td>\n",
       "      <td>159</td>\n",
       "      <td>80</td>\n",
       "      <td>79</td>\n",
       "      <td>2023-12-01 00:00:00+00:00</td>\n",
       "      <td>MULTIPOLYGON (((121.5243 25.11859, 121.52431 2...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>A6308-0588-00</td>\n",
       "      <td>臺北市</td>\n",
       "      <td>112</td>\n",
       "      <td>299</td>\n",
       "      <td>142</td>\n",
       "      <td>157</td>\n",
       "      <td>2023-12-01 00:00:00+00:00</td>\n",
       "      <td>MULTIPOLYGON (((121.57742 24.99215, 121.57746 ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        codebase city  h_cnt  p_cnt  m_cnt  f_cnt                 info_time  \\\n",
       "0  A6301-0097-00  臺北市     40    121     53     68 2023-12-01 00:00:00+00:00   \n",
       "1  A6303-0422-00  臺北市    146    376    165    211 2023-12-01 00:00:00+00:00   \n",
       "2  A6303-1146-00  臺北市     93    229    110    119 2023-12-01 00:00:00+00:00   \n",
       "3  A6311-0203-00  臺北市     58    159     80     79 2023-12-01 00:00:00+00:00   \n",
       "4  A6308-0588-00  臺北市    112    299    142    157 2023-12-01 00:00:00+00:00   \n",
       "\n",
       "                                            geometry  \n",
       "0  MULTIPOLYGON (((121.54425 25.06221, 121.54466 ...  \n",
       "1  MULTIPOLYGON (((121.55701 25.03439, 121.5575 2...  \n",
       "2  MULTIPOLYGON (((121.55202 25.02088, 121.55205 ...  \n",
       "3  MULTIPOLYGON (((121.5243 25.11859, 121.52431 2...  \n",
       "4  MULTIPOLYGON (((121.57742 24.99215, 121.57746 ...  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gdf.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2. Use `polars+connectorx` to fetch vector data from Postgis (faster, recommended)\n",
    "*CSL researcher only*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><style>\n",
       ".dataframe > thead > tr,\n",
       ".dataframe > tbody > tr {\n",
       "  text-align: right;\n",
       "  white-space: pre-wrap;\n",
       "}\n",
       "</style>\n",
       "<small>shape: (5, 8)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>codebase</th><th>city</th><th>h_cnt</th><th>p_cnt</th><th>m_cnt</th><th>f_cnt</th><th>info_time</th><th>geometry_wkb</th></tr><tr><td>str</td><td>str</td><td>i64</td><td>i64</td><td>i64</td><td>i64</td><td>datetime[ns, UTC]</td><td>binary</td></tr></thead><tbody><tr><td>&quot;A6303-0422-00&quot;</td><td>&quot;臺北市&quot;</td><td>146</td><td>376</td><td>165</td><td>211</td><td>2023-12-01 00:00:00 UTC</td><td>b&quot;\\x01\\x06\\x00\\x00\\x00\\x01\\x00\\x00\\x00\\x01\\x03\\x00\\x00\\x00\\x01\\x00\\x00\\x00\\x09\\x00\\x00\\x00\\x14\\x9c\\xfc\\x15\\xa6c^@\\xa5\\xa9\\xb8\\x9a\\xcd\\x089@v\\xd7\\xe3&amp;\\xaec^@\\x09\\xbfMB\\xcd\\x089@\\xef\\xe7\\x99\\x1c\\xaec&quot;...</td></tr><tr><td>&quot;A6303-1146-00&quot;</td><td>&quot;臺北市&quot;</td><td>93</td><td>229</td><td>110</td><td>119</td><td>2023-12-01 00:00:00 UTC</td><td>b&quot;\\x01\\x06\\x00\\x00\\x00\\x01\\x00\\x00\\x00\\x01\\x03\\x00\\x00\\x00\\x01\\x00\\x00\\x00\\x0b\\x00\\x00\\x006W\\xca&gt;Tc^@\\x9a\\xdb\\x15GX\\x059@\\x0d\\x06\\x09\\xdeTc^@&amp;\\x09\\x07pV\\x059@x\\xcd\\x94\\x9cVc&quot;...</td></tr><tr><td>&quot;A6311-0203-00&quot;</td><td>&quot;臺北市&quot;</td><td>58</td><td>159</td><td>80</td><td>79</td><td>2023-12-01 00:00:00 UTC</td><td>b&quot;\\x01\\x06\\x00\\x00\\x00\\x01\\x00\\x00\\x00\\x01\\x03\\x00\\x00\\x00\\x01\\x00\\x00\\x00\\x0b\\x00\\x00\\x00\\:83\\x8ea^@\\xee\\x9a\\x1a\\xc5[\\x1e9@\\xe3&#x27;-:\\x8ea^@\\xc0n2\\xc0[\\x1e9@Jz\\xd9\\xf1\\x92a&quot;...</td></tr><tr><td>&quot;A6308-0588-00&quot;</td><td>&quot;臺北市&quot;</td><td>112</td><td>299</td><td>142</td><td>157</td><td>2023-12-01 00:00:00 UTC</td><td>b&quot;\\x01\\x06\\x00\\x00\\x00\\x01\\x00\\x00\\x00\\x01\\x03\\x00\\x00\\x00\\x01\\x00\\x00\\x00\\x1d\\x00\\x00\\x00\\V,o\\xf4d^@H\\xa8\\x11\\xcc\\xfd\\xfd8@qV\\x9b\\x09\\xf5d^@\\xc3\\x93j\\x8d\\xf2\\xfd8@O\\x94G\\xf6\\xf4d&quot;...</td></tr><tr><td>&quot;A6310-1111-00&quot;</td><td>&quot;臺北市&quot;</td><td>0</td><td>0</td><td>0</td><td>0</td><td>2023-12-01 00:00:00 UTC</td><td>b&quot;\\x01\\x06\\x00\\x00\\x00\\x01\\x00\\x00\\x00\\x01\\x03\\x00\\x00\\x00\\x01\\x00\\x00\\x00&quot;\\x00\\x00\\x00Q\\x91;\\xf2\\x9ce^@\\x99\\xc8\\x19\\xd8\\xec\\x109@-\\xcf\\xdd\\xb4\\xa3e^@!\\xcfp\\x09\\xe1\\x109@m%\\xfa3\\xa5e&quot;...</td></tr></tbody></table></div>"
      ],
      "text/plain": [
       "shape: (5, 8)\n",
       "┌───────────────┬────────┬───────┬───────┬───────┬───────┬────────────────────┬────────────────────┐\n",
       "│ codebase      ┆ city   ┆ h_cnt ┆ p_cnt ┆ m_cnt ┆ f_cnt ┆ info_time          ┆ geometry_wkb       │\n",
       "│ ---           ┆ ---    ┆ ---   ┆ ---   ┆ ---   ┆ ---   ┆ ---                ┆ ---                │\n",
       "│ str           ┆ str    ┆ i64   ┆ i64   ┆ i64   ┆ i64   ┆ datetime[ns, UTC]  ┆ binary             │\n",
       "╞═══════════════╪════════╪═══════╪═══════╪═══════╪═══════╪════════════════════╪════════════════════╡\n",
       "│ A6303-0422-00 ┆ 臺北市 ┆ 146   ┆ 376   ┆ 165   ┆ 211   ┆ 2023-12-01         ┆ b\"\\x01\\x06\\x00\\x00 │\n",
       "│               ┆        ┆       ┆       ┆       ┆       ┆ 00:00:00 UTC       ┆ \\x00\\x01\\x00…      │\n",
       "│ A6303-1146-00 ┆ 臺北市 ┆ 93    ┆ 229   ┆ 110   ┆ 119   ┆ 2023-12-01         ┆ b\"\\x01\\x06\\x00\\x00 │\n",
       "│               ┆        ┆       ┆       ┆       ┆       ┆ 00:00:00 UTC       ┆ \\x00\\x01\\x00…      │\n",
       "│ A6311-0203-00 ┆ 臺北市 ┆ 58    ┆ 159   ┆ 80    ┆ 79    ┆ 2023-12-01         ┆ b\"\\x01\\x06\\x00\\x00 │\n",
       "│               ┆        ┆       ┆       ┆       ┆       ┆ 00:00:00 UTC       ┆ \\x00\\x01\\x00…      │\n",
       "│ A6308-0588-00 ┆ 臺北市 ┆ 112   ┆ 299   ┆ 142   ┆ 157   ┆ 2023-12-01         ┆ b\"\\x01\\x06\\x00\\x00 │\n",
       "│               ┆        ┆       ┆       ┆       ┆       ┆ 00:00:00 UTC       ┆ \\x00\\x01\\x00…      │\n",
       "│ A6310-1111-00 ┆ 臺北市 ┆ 0     ┆ 0     ┆ 0     ┆ 0     ┆ 2023-12-01         ┆ b\"\\x01\\x06\\x00\\x00 │\n",
       "│               ┆        ┆       ┆       ┆       ┆       ┆ 00:00:00 UTC       ┆ \\x00\\x01\\x00…      │\n",
       "└───────────────┴────────┴───────┴───────┴───────┴───────┴────────────────────┴────────────────────┘"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "city_code = 63000\n",
    "\n",
    "sql = f\"\"\"\n",
    "with boundary as (\n",
    "    select 代碼, geometry\n",
    "    from geometry.boundary_smallest\n",
    "    where 縣市代碼 = {city_code}\n",
    ")\n",
    "select aps.*, ST_AsBinary(boundary.geometry) as geometry_wkb\n",
    "from geometry.af_ppl_stats as aps\n",
    "join boundary\n",
    "on aps.codebase = boundary.代碼\n",
    "\"\"\"\n",
    "\n",
    "pdf = pl.read_database_uri(sql, URI, engine=\"connectorx\")\n",
    "pdf.head()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.3. Use `geopandas` to read geometry file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data = gpd.read_file('data/test_geom.geojson')\n",
    "data.plot(column='p_cnt',\n",
    "               cmap='OrRd',         # 使用顏色映射\n",
    "               scheme='naturalbreaks',  # 自然斷點分類\n",
    "               k=5,\n",
    "               legend=False\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Geometry to h3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `H3Toolkit` is a chainable class that can convert geometry data to h3 index.\n",
    "\n",
    "- **Step1**: use `set_aggregation_strategy()` to set the target column and the aggregation method in the dict `{ column_name : aggregation_function() }`.\n",
    "- **Step2**: use `process_from_vector()` to set the data and the h3 resolution, this function will convert the geometry to h3 index in the designated resolution and process the column(s) with the aggregation method you set.\n",
    "- **Step3**: use `get_result()` to get the result. default is a polars dataframe with the h3 index and the aggregated column(s), if you want the actual geometry, you can set `return_geometry=True` to get the geometry column, and the return type will be the geopandas dataframe.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "NOTE: this function always use the centroid of the h3 hexagon to extract the data from actual geometry object, so the result may loss some information if you choose the inproper resolution. If missing data exists, the warnging will be raised. you should choose the proper resolution to avoid this situation, ie from resolution 7 to 10. if you want the lower resolution, you can chain the result from `process_from_vector()` to `process_from_h3()` to get the lower resolution h3 index without lossing any information."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "must to use `get_result()` to get the result, otherwise return will be the H3Toolkit object."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><style>\n",
       ".dataframe > thead > tr,\n",
       ".dataframe > tbody > tr {\n",
       "  text-align: right;\n",
       "  white-space: pre-wrap;\n",
       "}\n",
       "</style>\n",
       "<small>shape: (5, 2)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>hex_id</th><th>p_cnt</th></tr><tr><td>str</td><td>f64</td></tr></thead><tbody><tr><td>&quot;8c4ba0a412a01ff&quot;</td><td>14.208333</td></tr><tr><td>&quot;8c4ba0a412a05ff&quot;</td><td>14.208333</td></tr><tr><td>&quot;8c4ba0a412a07ff&quot;</td><td>14.208333</td></tr><tr><td>&quot;8c4ba0a412a09ff&quot;</td><td>14.208333</td></tr><tr><td>&quot;8c4ba0a412a0bff&quot;</td><td>14.208333</td></tr></tbody></table></div>"
      ],
      "text/plain": [
       "shape: (5, 2)\n",
       "┌─────────────────┬───────────┐\n",
       "│ hex_id          ┆ p_cnt     │\n",
       "│ ---             ┆ ---       │\n",
       "│ str             ┆ f64       │\n",
       "╞═════════════════╪═══════════╡\n",
       "│ 8c4ba0a412a01ff ┆ 14.208333 │\n",
       "│ 8c4ba0a412a05ff ┆ 14.208333 │\n",
       "│ 8c4ba0a412a07ff ┆ 14.208333 │\n",
       "│ 8c4ba0a412a09ff ┆ 14.208333 │\n",
       "│ 8c4ba0a412a0bff ┆ 14.208333 │\n",
       "└─────────────────┴───────────┘"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from h3_toolkit import H3Toolkit\n",
    "from h3_toolkit.aggregation import EqualSplit\n",
    "\n",
    "toolkit = H3Toolkit()\n",
    "\n",
    "result = (\n",
    "    toolkit\n",
    "    .set_aggregation_strategy(\n",
    "        {\n",
    "            'p_cnt' : EqualSplit(agg_col='codebase'),\n",
    "        }\n",
    "    )\n",
    "    .process_from_vector(data, resolution=12, geometry_col='geometry')\n",
    ")\n",
    "\n",
    "output_df = result.get_result()\n",
    "\n",
    "output_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "set `return_geometry=True` to get the geometry column in the result."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hex_id</th>\n",
       "      <th>p_cnt</th>\n",
       "      <th>geometry</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8c4ba0a412a01ff</td>\n",
       "      <td>14.208333</td>\n",
       "      <td>POLYGON ((121.54706 25.04487, 121.54705 25.044...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8c4ba0a412a05ff</td>\n",
       "      <td>14.208333</td>\n",
       "      <td>POLYGON ((121.54695 25.04474, 121.54694 25.044...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8c4ba0a412a07ff</td>\n",
       "      <td>14.208333</td>\n",
       "      <td>POLYGON ((121.54713 25.04473, 121.54711 25.044...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8c4ba0a412a09ff</td>\n",
       "      <td>14.208333</td>\n",
       "      <td>POLYGON ((121.547 25.04501, 121.54698 25.04492...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8c4ba0a412a0bff</td>\n",
       "      <td>14.208333</td>\n",
       "      <td>POLYGON ((121.54717 25.045, 121.54716 25.04491...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            hex_id      p_cnt  \\\n",
       "0  8c4ba0a412a01ff  14.208333   \n",
       "1  8c4ba0a412a05ff  14.208333   \n",
       "2  8c4ba0a412a07ff  14.208333   \n",
       "3  8c4ba0a412a09ff  14.208333   \n",
       "4  8c4ba0a412a0bff  14.208333   \n",
       "\n",
       "                                            geometry  \n",
       "0  POLYGON ((121.54706 25.04487, 121.54705 25.044...  \n",
       "1  POLYGON ((121.54695 25.04474, 121.54694 25.044...  \n",
       "2  POLYGON ((121.54713 25.04473, 121.54711 25.044...  \n",
       "3  POLYGON ((121.547 25.04501, 121.54698 25.04492...  \n",
       "4  POLYGON ((121.54717 25.045, 121.54716 25.04491...  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "output_gdf = result.get_result(return_geometry=True, fill_null_value=0)\n",
    "output_gdf.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Scale current h3 resolution to the target resolution\n",
    "\n",
    "scale up the h3 index to the target resolution, the `process_from_h3()` function will do this job.\n",
    "\n",
    "NOTE: The `source_resolution` must be greater than the `target_resolution`, otherwise the result will be wrong."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><style>\n",
       ".dataframe > thead > tr,\n",
       ".dataframe > tbody > tr {\n",
       "  text-align: right;\n",
       "  white-space: pre-wrap;\n",
       "}\n",
       "</style>\n",
       "<small>shape: (5, 2)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>hex_id</th><th>p_cnt</th></tr><tr><td>str</td><td>f64</td></tr></thead><tbody><tr><td>&quot;8a4ba0a5d18ffff&quot;</td><td>23.007257</td></tr><tr><td>&quot;8a4ba0a43c4ffff&quot;</td><td>603.692149</td></tr><tr><td>&quot;8a4ba0a4e99ffff&quot;</td><td>199.093091</td></tr><tr><td>&quot;8a4ba0a5d19ffff&quot;</td><td>11.574834</td></tr><tr><td>&quot;8a4ba0a5db57fff&quot;</td><td>485.178707</td></tr></tbody></table></div>"
      ],
      "text/plain": [
       "shape: (5, 2)\n",
       "┌─────────────────┬────────────┐\n",
       "│ hex_id          ┆ p_cnt      │\n",
       "│ ---             ┆ ---        │\n",
       "│ str             ┆ f64        │\n",
       "╞═════════════════╪════════════╡\n",
       "│ 8a4ba0a5d18ffff ┆ 23.007257  │\n",
       "│ 8a4ba0a43c4ffff ┆ 603.692149 │\n",
       "│ 8a4ba0a4e99ffff ┆ 199.093091 │\n",
       "│ 8a4ba0a5d19ffff ┆ 11.574834  │\n",
       "│ 8a4ba0a5db57fff ┆ 485.178707 │\n",
       "└─────────────────┴────────────┘"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from h3_toolkit import H3Toolkit\n",
    "from h3_toolkit.aggregation import Sum\n",
    "\n",
    "toolkit = H3Toolkit()\n",
    "\n",
    "result = (\n",
    "    toolkit\n",
    "    .set_aggregation_strategy(\n",
    "        {\n",
    "            'p_cnt' : Sum(),\n",
    "        }\n",
    "    )\n",
    "    .process_from_h3(\n",
    "        output_df,\n",
    "        source_resolution=12,\n",
    "        target_resolution=10,\n",
    "        h3_col='hex_id'\n",
    "    )\n",
    ")\n",
    "\n",
    "output_df_10 = result.get_result(fill_null_value=0)\n",
    "output_df_10.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, you are successfully scale up the data from resolution 12 to 10."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The benefit of H3Toolkit is that you can chain all the functions together, it's much simple and maintainable. and make the process logic more clear!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><style>\n",
       ".dataframe > thead > tr,\n",
       ".dataframe > tbody > tr {\n",
       "  text-align: right;\n",
       "  white-space: pre-wrap;\n",
       "}\n",
       "</style>\n",
       "<small>shape: (5, 3)</small><table border=\"1\" class=\"dataframe\"><thead><tr><th>hex_id</th><th>p_cnt</th><th>h_cnt</th></tr><tr><td>str</td><td>f64</td><td>f64</td></tr></thead><tbody><tr><td>&quot;8a4ba0a4e847fff&quot;</td><td>595.238028</td><td>248.036192</td></tr><tr><td>&quot;8a4ba0a5d30ffff&quot;</td><td>2.853159</td><td>1.233115</td></tr><tr><td>&quot;8a4ba0a4301ffff&quot;</td><td>868.195243</td><td>351.344298</td></tr><tr><td>&quot;8a4ba0a4e557fff&quot;</td><td>587.340822</td><td>231.649744</td></tr><tr><td>&quot;8a4ba0a5d8f7fff&quot;</td><td>9.990734</td><td>3.819245</td></tr></tbody></table></div>"
      ],
      "text/plain": [
       "shape: (5, 3)\n",
       "┌─────────────────┬────────────┬────────────┐\n",
       "│ hex_id          ┆ p_cnt      ┆ h_cnt      │\n",
       "│ ---             ┆ ---        ┆ ---        │\n",
       "│ str             ┆ f64        ┆ f64        │\n",
       "╞═════════════════╪════════════╪════════════╡\n",
       "│ 8a4ba0a4e847fff ┆ 595.238028 ┆ 248.036192 │\n",
       "│ 8a4ba0a5d30ffff ┆ 2.853159   ┆ 1.233115   │\n",
       "│ 8a4ba0a4301ffff ┆ 868.195243 ┆ 351.344298 │\n",
       "│ 8a4ba0a4e557fff ┆ 587.340822 ┆ 231.649744 │\n",
       "│ 8a4ba0a5d8f7fff ┆ 9.990734   ┆ 3.819245   │\n",
       "└─────────────────┴────────────┴────────────┘"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "toolkit = H3Toolkit()\n",
    "\n",
    "result = (\n",
    "    toolkit\n",
    "    .set_aggregation_strategy(\n",
    "        {\n",
    "            ('p_cnt', 'h_cnt') : EqualSplit(agg_col='codebase'),\n",
    "        }\n",
    "    )\n",
    "    .process_from_vector(data, resolution=12, geometry_col='geometry')\n",
    "    .set_aggregation_strategy(\n",
    "        {\n",
    "            ('p_cnt', 'h_cnt') : Sum(),\n",
    "        }\n",
    "    )\n",
    "    .process_from_h3(\n",
    "        source_resolution=12,\n",
    "        target_resolution=10,\n",
    "        h3_col='hex_id'\n",
    "    )\n",
    "    .get_result()\n",
    ")\n",
    "\n",
    "result.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Visualization\n",
    "\n",
    "Use the `show_h3()` function from `h3_toolkit.visualization` module to visualize the h3 data interactively with pydeck.\n",
    "\n",
    "Supported classifiers (from mapclassify): NaturalBreaks, Quantiles, EqualInterval, Jenks, and more.\n",
    "Supported colormaps (from matplotlib): Oranges, viridis, RdYlGn, Blues, Greens, and many others."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from h3_toolkit.visualization import show_h3\n",
    "\n",
    "# Visualize with default classifier (NaturalBreaks) and colormap (Oranges)\n",
    "show_h3(output_df, 'p_cnt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Visualize resolution 12 data with default NaturalBreaks classifier\n",
    "toolkit.show(target_col='p_cnt')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Due to performance limitations when rendering large H3 datasets on the web, here’s a screenshot example of the visualization. For an interactive map, you can clone the notebook and run it locally in your environment\n",
    "\n",
    "\n",
    "![res12](./img/01-1.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Visualize resolution 10 data with Quantiles classifier\n",
    "show_h3(output_df_10, 'p_cnt', classifier='Quantiles', k=5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Due to performance limitations when rendering large H3 datasets on the web, here’s a screenshot example of the visualization. For an interactive map, you can clone the notebook and run it locally in your environment\n",
    "\n",
    "\n",
    "![res10](./img/01-2.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It's not recommended using `pydeck` or `kepler.gl` to visualize the h3 data if the data is too large, because the h3 data is too dense to visualize, it will be very slow and may crash.\n",
    "\n",
    "The better way to visual large dataset is using `lonboard`, but this package is not support the h3 data yet, so you should convert the h3 data to the geometry data before visualizing it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "## TODO: Use case of lonborad"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Customize the aggregation function\n",
    "\n",
    "We provided some popular aggregation methods for doing the aggregation, read the `h3_toolkit.aggregation` class to get more information.\n",
    "\n",
    "But you can also custom your specific aggregation function by yourself, just define a function that inherit from the `AggregationStrategy` class and implement the `apply()` function.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<h3_toolkit.core.H3Toolkit at 0x14396b2c0>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from h3_toolkit import H3Toolkit\n",
    "from h3_toolkit.aggregation import AggregationStrategy\n",
    "\n",
    "\n",
    "class NewAggregationStrategy(AggregationStrategy):\n",
    "    \"\"\"\n",
    "    Make sure to inherit from AggregationStrategy and implement the apply method\n",
    "    with data and target_cols attributes.\n",
    "    \"\"\"\n",
    "    def __init__(self, new_var):\n",
    "        \"\"\"\n",
    "        If you want to pass some parameters to the strategy, you can do it here.\n",
    "        \"\"\"\n",
    "        self.new_var = new_var\n",
    "\n",
    "\n",
    "    def apply(data:pl.DataFrame, target_cols:list[str]) -> pl.DataFrame:\n",
    "        \"\"\"\n",
    "        If you are not familiar with polars, you can use pandas instead.\n",
    "        just make sure to convert the type to polars dataframe before returning it.\n",
    "        >>> data = data.to_pandas()\n",
    "        >>> ...\n",
    "        >>> ...\n",
    "        >>> ...\n",
    "        >>> return pl.from_pandas(data)\n",
    "        \"\"\"\n",
    "        pass\n",
    "\n",
    "\n",
    "toolkit = H3Toolkit()\n",
    "toolkit.set_aggregation_strategy(\n",
    "    {\n",
    "        ('col_name', '...') : NewAggregationStrategy(new_var='...'),\n",
    "    }\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.12.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}