earthporn_downloader.py

# Earthporndownloader.py
# Downloads the top images from reddit.com/r/earthporn and saves them to a wallpapers folder

import requests
import json
import sys
import uuid
import re
from PIL import Image
import os.path
import os


if hasattr(sys, 'real_prefix'):
    print("in venv")

WALLPAPER_PATH = os.path.dirname(os.path.realpath(__file__))

req = requests.get("https://www.reddit.com/r/earthporn.json",
                    headers = {'User-agent' : 'Your Bot 0.1'})

if req.status_code != 200:
    print("Error downloading the earthporn.json. Exiting")
    sys.exit()

js = json.loads(req.text)

children_array = js['data']['children']

downloaded_ids = []

# Load downloaded IDs
with open('downloaded_images.txt', 'r') as downloaded_images_file :
    for id in downloaded_images_file:
        # Loads as array, remove line endings to make it cleaner/easier to match
        id = id.replace('\n', '').replace('\r', '')
        downloaded_ids.append(id)

# Now open the file to append
downloaded_images_file = open('downloaded_images.txt', 'a')

path = os.path.abspath(WALLPAPER_PATH)

for post in children_array:
    url = post['data']['url']

    id = re.match(r"https?://.*/(.*).jpg", url)

    # Check if an http url is found
    if id:
        id = id.group(1)

        # IF Image has not been downloaded before, download it now
        # Catches issues where you would download duplicates

        if(id not in downloaded_ids):
            downloaded_images_file.write(id + "\n")
            filename = os.path.join(path, str(uuid.uuid4()) + ".jpg")

            image_request = requests.get(url)
            with open(filename, 'wb') as file:
                file.write(image_request.content)

            # TODO filter by width/height/some other attribute to pick only the high quality monitors that look good for a background
            im = Image.open(filename)
            width, height = im.size

downloaded_images_file.close()