{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xml.etree.ElementTree as ET\n",
    "import requests\n",
    "import csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "xml_url = \"https://wayback.maptiles.arcgis.com/arcgis/rest/services/world_imagery/mapserver/wmts/1.0.0/wmtscapabilities.xml\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# parse the xml\n",
    "xml = requests.get(xml_url).text\n",
    "root = ET.fromstring(xml)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Namespace definitions\n",
    "namespaces = {\n",
    "    'ows': 'https://www.opengis.net/ows/1.1',\n",
    "    '': 'https://www.opengis.net/wmts/1.0'  # Default namespace\n",
    "}\n",
    "\n",
    "# Open CSV file for writing\n",
    "with open('wayback.csv', 'w', newline='') as csv_file:\n",
    "    writer = csv.writer(csv_file)\n",
    "\n",
    "    # Write header row\n",
    "    writer.writerow([\n",
    "        'Title', \n",
    "        'Identifier', \n",
    "        'LowerCorner', \n",
    "        'UpperCorner', \n",
    "        'Format', \n",
    "        'TileMatrixSetLinks', \n",
    "        'ResourceURL_Template'\n",
    "    ])\n",
    "\n",
    "    # Extract and write data\n",
    "    for layer in root.findall('.//Layer', namespaces):\n",
    "        title = layer.find('ows:Title', namespaces).text\n",
    "        identifier = layer.find('ows:Identifier', namespaces).text\n",
    "\n",
    "        bounding_box = layer.find('ows:BoundingBox/ows:LowerCorner', namespaces)\n",
    "        lower_corner = bounding_box.text if bounding_box is not None else ''\n",
    "\n",
    "        bounding_box = layer.find('ows:BoundingBox/ows:UpperCorner', namespaces)\n",
    "        upper_corner = bounding_box.text if bounding_box is not None else ''\n",
    "\n",
    "        fmt = layer.find('Format', namespaces).text if layer.find('Format', namespaces) else ''\n",
    "\n",
    "        tile_matrix_links = [tms.text for tms in layer.findall('TileMatrixSetLink/TileMatrixSet', namespaces)]\n",
    "        tile_matrix_set_links = ', '.join(tile_matrix_links)\n",
    "\n",
    "        resource_url = layer.find('ResourceURL', namespaces)\n",
    "        resource_url_template = resource_url.get('template') if resource_url is not None else ''\n",
    "\n",
    "        writer.writerow([\n",
    "            title,\n",
    "            identifier,\n",
    "            lower_corner,\n",
    "            upper_corner,\n",
    "            fmt,\n",
    "            tile_matrix_set_links,\n",
    "            resource_url_template\n",
    "        ])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Parse dates"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "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>Title</th>\n",
       "      <th>Identifier</th>\n",
       "      <th>LowerCorner</th>\n",
       "      <th>UpperCorner</th>\n",
       "      <th>Format</th>\n",
       "      <th>TileMatrixSetLinks</th>\n",
       "      <th>ResourceURL_Template</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>World Imagery (Wayback 2024-10-10)</td>\n",
       "      <td>WB_2024_R11</td>\n",
       "      <td>-2.003750722959434E7 -2.003750722959434E7</td>\n",
       "      <td>2.003750722959434E7 2.003750722959434E7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>default028mm, GoogleMapsCompatible</td>\n",
       "      <td>https://wayback.maptiles.arcgis.com/arcgis/res...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>World Imagery (Wayback 2024-09-19)</td>\n",
       "      <td>WB_2024_R10</td>\n",
       "      <td>-2.003750722959434E7 -2.003750722959434E7</td>\n",
       "      <td>2.003750722959434E7 2.003750722959434E7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>default028mm, GoogleMapsCompatible</td>\n",
       "      <td>https://wayback.maptiles.arcgis.com/arcgis/res...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                Title   Identifier  \\\n",
       "0  World Imagery (Wayback 2024-10-10)  WB_2024_R11   \n",
       "1  World Imagery (Wayback 2024-09-19)  WB_2024_R10   \n",
       "\n",
       "                                 LowerCorner  \\\n",
       "0  -2.003750722959434E7 -2.003750722959434E7   \n",
       "1  -2.003750722959434E7 -2.003750722959434E7   \n",
       "\n",
       "                               UpperCorner  Format  \\\n",
       "0  2.003750722959434E7 2.003750722959434E7     NaN   \n",
       "1  2.003750722959434E7 2.003750722959434E7     NaN   \n",
       "\n",
       "                   TileMatrixSetLinks  \\\n",
       "0  default028mm, GoogleMapsCompatible   \n",
       "1  default028mm, GoogleMapsCompatible   \n",
       "\n",
       "                                ResourceURL_Template  \n",
       "0  https://wayback.maptiles.arcgis.com/arcgis/res...  \n",
       "1  https://wayback.maptiles.arcgis.com/arcgis/res...  "
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.read_csv('wayback.csv')\n",
    "df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Title</th>\n",
       "      <th>Identifier</th>\n",
       "      <th>LowerCorner</th>\n",
       "      <th>UpperCorner</th>\n",
       "      <th>Format</th>\n",
       "      <th>TileMatrixSetLinks</th>\n",
       "      <th>ResourceURL_Template</th>\n",
       "      <th>date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>World Imagery (Wayback 2024-10-10)</td>\n",
       "      <td>WB_2024_R11</td>\n",
       "      <td>-2.003750722959434E7 -2.003750722959434E7</td>\n",
       "      <td>2.003750722959434E7 2.003750722959434E7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>default028mm, GoogleMapsCompatible</td>\n",
       "      <td>https://wayback.maptiles.arcgis.com/arcgis/res...</td>\n",
       "      <td>2024-10-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>World Imagery (Wayback 2024-09-19)</td>\n",
       "      <td>WB_2024_R10</td>\n",
       "      <td>-2.003750722959434E7 -2.003750722959434E7</td>\n",
       "      <td>2.003750722959434E7 2.003750722959434E7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>default028mm, GoogleMapsCompatible</td>\n",
       "      <td>https://wayback.maptiles.arcgis.com/arcgis/res...</td>\n",
       "      <td>2024-09-19</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                Title   Identifier  \\\n",
       "0  World Imagery (Wayback 2024-10-10)  WB_2024_R11   \n",
       "1  World Imagery (Wayback 2024-09-19)  WB_2024_R10   \n",
       "\n",
       "                                 LowerCorner  \\\n",
       "0  -2.003750722959434E7 -2.003750722959434E7   \n",
       "1  -2.003750722959434E7 -2.003750722959434E7   \n",
       "\n",
       "                               UpperCorner  Format  \\\n",
       "0  2.003750722959434E7 2.003750722959434E7     NaN   \n",
       "1  2.003750722959434E7 2.003750722959434E7     NaN   \n",
       "\n",
       "                   TileMatrixSetLinks  \\\n",
       "0  default028mm, GoogleMapsCompatible   \n",
       "1  default028mm, GoogleMapsCompatible   \n",
       "\n",
       "                                ResourceURL_Template       date  \n",
       "0  https://wayback.maptiles.arcgis.com/arcgis/res... 2024-10-10  \n",
       "1  https://wayback.maptiles.arcgis.com/arcgis/res... 2024-09-19  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['date'] = df['Title'].str.extract(r'(\\d{4}-\\d{2}-\\d{2})')\n",
    "df['date'] = pd.to_datetime(df['date'], format='%Y-%m-%d')\n",
    "df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_parquet('wayback.parquet', index=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "zeel_py310",
   "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": "3.10.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}