load_rock.py

from app_vars import *
from bs4 import BeautifulSoup as bs
import random as r
import requests
from PIL import Image
from io import BytesIO
import binascii

headers = { 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.3',
            "Accept" : "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7",
            "Accept-Encoding": "gzip, deflate, br",
            "Accept-Language": "en-US,en;q=0.9",
            "Cache-Control": "max-age=0",
            "Cookie": """mindat=n5e0bck7vvnn6e0o9tkb49fs75; cf_clearance=PMEAfGmJF_UGxZYgB2jsRx5KsiZyNf2plDbXs4HoN5I-1703800389-0-2-d8c97f25.71e5db39.8fc820c5-0.2.1703800389""",
            "Dnt": """1""",
            "If-Modified-Since": """Sat, 17 Apr 2021 18:41:12 GMT""",
            "If-None-Match": "607b2bc8-956c4",
            "Prefer": "safe",
            "Sec-Ch-Ua": r'"Not_A Brand";v="8", "Chromium";v="120", "Microsoft Edge";v="120"',
            "Sec-Ch-Ua-Mobile": "?0",
            "Sec-Ch-Ua-Platform": '"Windows"',
            "Sec-Fetch-Dest": "document",
            "Sec-Fetch-Mode": "navigate",
            "Sec-Fetch-Site": "none",
            "Sec-Fetch-User": "?1",
            "Upgrade-Insecure-Requests": "1",}



def get_gallery_image_list_from_rock_id(chosen_rock_id):
    system(
        f'{curl_invoker} -A "Mozilla/5.0 (platform; rv:geckoversion) Gecko/geckotrail Firefox/firefoxversion" -o "{rs}/app_data/html/index.html" https://www.mindat.org/gm/{chosen_rock_id}'
    )

    with open(f"{rs}\\app_data\\html\\index.html", "rb") as f:
        soup = bs(f, "html.parser")

    image_list = []
    for i in soup.find_all("img"):
        image_list.append("https://www.mindat.org" + i.get("src"))
    return image_list


rl = load(open(f"{rs}/app_data/rock_urls_list.json"))
rl_keys = list(rl.keys())


def rand_rock_key():
    return r.choice(rl_keys)


def url_of_rock(rock_name):
    return r.choice(rl[rock_name])


def get_rock_name_url():
    name = rand_rock_key()
    url = url_of_rock(name)
    # system(f'{curl_invoker} -A "Mozilla/5.0 (platform; rv:geckoversion) Gecko/geckotrail Firefox/firefoxversion" {url}')
    img = requests.get(url,headers = headers).content
    print(img.decode("ascii"))
    return name, Image.open(BytesIO(binascii.unhexlify(img)))