Download landcover tile to local filesystem

download_landcover_data.py

import os

import requests

from take_home_test_starter.config import Config

BASE_URL = "https://s3-eu-west-1.amazonaws.com/vito.landcover.global/v3.0.1/2019/"

LATITUDES = ["S40", "S20", "N00", "N20", "N40", "N60", "N80"]

LONGTITUDES = [
    "W180",
    "W160",
    "W140",
    "W120",
    "W100",
    "W080",
    "W060",
    "W040",
    "W020",
    "E000",
    "E020",
    "E040",
    "E060",
    "E080",
    "E100",
    "E120",
    "E140",
    "E160",
]

KINDS = ["Discrete"]

DIR = ".landcover"

def build_url(tile_id: str, tile_kind: str) -> str:
    return f"{BASE_URL}{tile_id}/{tile_id}_PROBAV_LC100_global_v3.0.1_2019-nrt_{tile_kind}-Classification-map_EPSG-4326.tif"

def download_file(tile_id: str, kind: str) -> str:
    file_path = None
    response = requests.get(build_url(tile_id, kind), timeout=30)
    if response.status_code == 200:
        file_path = os.path.join(DIR, f"{tile_id}.tif")
        with open(file_path, "wb") as file:
            file.write(response.content)
            # Remove empty files
            if os.path.getsize(file_path) == 0:
                os.remove(file_path)
                file_path = None
            else:
                file_path
    return file_path
    
def download_landcover_data() -> list[str]:
    if not os.path.exists(DIR):
        os.makedirs(DIR)
    files = []
    # Loop through the combinations of latitude, longitude, and kind and download each file
    for latitude in LATITUDES:
        for longitude in LONGTITUDES:
            for kind in KINDS:
                file = download_file(f"{longitude}{latitude}", kind)
                if file:
                    files.append(file)
    return files

if __name__ == "__main__":
    download_landcover_data()