maps.chandlerswift.com/layers/chains/menards/get_data.py
Chandler Swift 7957523c3c
Distribute data with layers instead of data/ dir
This was originally done to make the gitignoring easier, but ended up
being somewhat more complex when trying to include files, so they're
moving out closer to the point of use.
2023-07-25 19:14:05 -05:00

59 lines
1.6 KiB
Python
Executable file

#!/usr/bin/env python3
import requests
import re
import json
# Stolen from my machine, appears to work; sufficient and necessary to get
# around their firewall apparently? Last time this didn't work, so I wonder if
# their protections get stronger as I make more requests. Hopefully not!
UA={
"User-Agent": 'Mozilla/5.0 (X11; Linux x86_64; rv:109.0) Gecko/20100101 Firefox/114.0'
}
res = requests.get('https://www.menards.com/store-details/locator.html', headers=UA)
content = res.text
# # For debugging, only fetch the above once; then cache it:
# with open('temp.txt', 'w') as f:
# f.write(content)
# ...and re-open it each time
# with open('temp.txt') as f:
# content = f.read()
try:
raw_initial_store_data = re.search(r'data-initial-stores="([^"]*)"', content)[1]
except:
print(content)
raise SystemExit
menardses = json.loads(raw_initial_store_data.replace(""", '"'))
stores = []
for menards in menardses:
stores.append({
"type": "Feature",
"geometry": {
"type": "Point",
"coordinates": [float(menards['longitude']), float(menards['latitude'])], # yes, [lon, lat] since it's [x, y]
},
"properties": {
'address': menards['street'].title(),
'city': menards['city'].title(),
'state': menards['state'],
'zip': menards['zip'],
'website': f"https://www.menards.com/store-details/store.html?store={menards['number']}",
},
})
geojson = {
"type": "FeatureCollection",
"features": stores,
}
with open("data.geojson", "w") as f:
f.write(json.dumps(geojson))
print(f"{len(menardses)} locations found")