### Run this on startup
import requests
import pandas as pd
from datasets import Dataset, DatasetDict
from datetime import datetime
from tqdm import tqdm


url = "https://dumpstermap.herokuapp.com/dumpsters/withinbounds/-180/-90/180/90/"
response = requests.get(url)
entry = response.json()
dumpsters = entry["features"]

cols = ["Latitude", "Longitude", "dumpster_created", "voting", "comment", "voting_created", "name"]

dumpsters_df = pd.DataFrame(columns=cols)
progress_text = "% of spots fetched"

num_dumpsters = len(dumpsters)


for dumpster in tqdm(dumpsters):
    url = f"https://dumpstermap.herokuapp.com/dumpsters/{dumpster['id']}"
    response = requests.get(url)
    if response.status_code != 200:
        print("Problem fetching from:", url, response.status_code)
    try:
        entry = response.json()

        rows = []
        lat = entry['geometry']['coordinates'][1]
        lon = entry['geometry']['coordinates'][0]
        dumpster_created = entry['properties']['created']

        for vote in entry['properties']['voting_set']:
            rows.append({
                "Latitude": lat,
                "Longitude": lon,
                "dumpster_created": dumpster_created,
                "voting": vote['value'],
                "comment": vote['comment'],
                "voting_created": vote['created_date'],
                "name": vote['name']
            })

        if rows:
            dumpsters_df = pd.concat([dumpsters_df, pd.DataFrame(rows)], ignore_index=True)
    except Exception as e:
        print(e)

dataset = Dataset.from_pandas(dumpsters_df)
dataset_dict = DatasetDict({datetime.now().strftime("%Y.%m.%d"): dataset})
dataset_dict.push_to_hub("Hitchwiki/dumpster_diving_spots")