Download "Windzone" for all german zipcodes

crawl_windzone.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4f9444e2",
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests\n",
    "import pandas as pd\n",
    "import bs4\n",
    "import random\n",
    "import pathlib\n",
    "import numpy as np\n",
    "\n",
    "def read_windzone(plz):\n",
    "    # Download\n",
    "    temp_text = requests.get(base_url + plz).text\n",
    "    # Parse\n",
    "    temp_return = bs4.BeautifulSoup(temp_text,\"html.parser\").select(\"h2 span\")\n",
    "    # Return Windzone if there is one\n",
    "    if len(temp_return) == 2:\n",
    "        return plz, temp_return[1].text\n",
    "    else:\n",
    "        return plz, '0'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bdd04c14",
   "metadata": {},
   "outputs": [],
   "source": [
    "base_url = 'https://www.dehn.de/windzonen/?zip='\n",
    "windzonen_path = \"windzonen.csv\"\n",
    "\n",
    "# Download PLZ-Data from github (not official)\n",
    "plz_full = pd.read_csv('https://raw.githubusercontent.com/zauberware/postal-codes-json-xml-csv/master/data/DE/zipcodes.de.csv',\n",
    "                   dtype={'zipcode': str})\n",
    "\n",
    "# Get Unique PLZ Codes and randomize \n",
    "plz_data = plz_full.drop_duplicates(subset='zipcode')[['zipcode']].sample(frac=1,random_state=1888).reset_index(drop=True)\n",
    "\n",
    "# If doesnt exist yet --> Write csv that will be appended\n",
    "if not pathlib.Path(windzonen_path).exists():\n",
    "    plz_out = pd.DataFrame(columns = [\"zipcode\", \"windzone\"])\n",
    "    plz_out.to_csv(windzonen_path, sep = \";\", index = False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7c2db3e0",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Read existing windzonen \n",
    "existing_plz = pd.read_csv(windzonen_path, sep = \";\", dtype={'zipcode': str})\n",
    "\n",
    "# Download windzone\n",
    "for index, row in plz_data.iterrows():\n",
    "    print(index)\n",
    "    print(row[\"zipcode\"])\n",
    "    if row[\"zipcode\"] in existing_plz.zipcode.values:\n",
    "        print(\"exists already\")\n",
    "        continue\n",
    "    temp_windzone = pd.DataFrame([read_windzone(row[\"zipcode\"])], columns = [\"zipcode\", \"windzone\"] )\n",
    "    temp_windzone.to_csv(windzonen_path, mode = 'a', header = False, sep = \";\", index = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c4d6ca14",
   "metadata": {},
   "outputs": [],
   "source": [
    "existing_plz = pd.read_csv(windzonen_path, sep = \";\", dtype={'zipcode': str})\n",
    "# Check distribution of windzonen\n",
    "print(existing_plz.windzone.value_counts(dropna=False))\n",
    "# Replace 0 with np.nan and rename column\n",
    "existing_plz.windzone.replace(0.0,np.nan,inplace=True)\n",
    "existing_plz.rename(columns = {'zipcode':'plz'}, inplace = True)\n",
    "print(existing_plz.windzone.value_counts(dropna=False))\n",
    "# Save final version\n",
    "existing_plz.to_csv(\"plz_windzone.csv\", header = True, sep = \";\", index = False)"
   ]
  }
 ],
 "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
}