#!/usr/bin/env python3

import requests
import json
from bs4 import BeautifulSoup
import re
import urllib.parse

# Stolen from my machine, appears to work; sufficient and necessary to get
# around their firewall apparently?
UA={
    "User-Agent": 'Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/114.0'
}

response = requests.get('https://punchpizza.com/locations/', headers=UA)

soup = BeautifulSoup(response.text, 'html.parser')
location_links = soup.select('div.loctop > .wpb_wrapper > ul > li > a') # Two rows with the same id :eyeroll:

locations = []
for location_link in location_links:
    location_response = response = requests.get(urllib.parse.urljoin("https://punchpizza.com", location_link['href']), headers=UA)
    latlon = re.search(r'var punchloc = {lat: ([0-9.-]*), lng: ([0-9.-]*)};', location_response.text)
    if not latlon:
        raise Exception("No latlon found")
    locations.append({
        "type": "Feature",
        "geometry": {
            "type": "Point",
            "coordinates": [float(latlon[2]), float(latlon[1])], # yes, [lon, lat] since it's [x, y]
        },
        # TODO: addresses are kind of a mess
    })

geojson = {
    "type": "FeatureCollection",
    "features": locations,
}

print(len(locations), "locations found")

with open("data.geojson", "w") as f:
    f.write(json.dumps(geojson))