Find the Art - A solution to finding the art page for Permanent Redirect by Donald Hanson

find_the_art.py

#!/usr/bin/env python3


"""
A simple program to scrape the contents of permanent redirects to
"The Art" in order to track it down, before it can move on...
"""


import re
import requests
import sys
import time

from requests.exceptions import ConnectionError, SSLError


LINK_REGEX = re.compile(r">permanent-redirect.xyz/pages/([0-9]+)<")


def main():
    path_suffix = int(time.time())

    url_template = "https://permanent-redirect.xyz/pages/{}"

    print("Attempting to get a foot on the ladder...")
    while True:
        url = url_template.format(path_suffix)
        print("Trying {}".format(url))
        try:
            res = requests.get(url)
        except (ConnectionError, SSLError):
            print("Error during request, just keep moving on")
            continue

        status_code = res.status_code
        if status_code == 200:
            break

        path_suffix -= 1

    print("\nOk, we're on. Now climb it to the top...")
    retries = 3
    while retries > 0:
        url = url_template.format(path_suffix)
        print("Trying {}".format(url))
        try:
            res = requests.get(url)
        except (ConnectionError, SSLError):
            print("Error: SSLError during request")
            retries -= 1
            continue

        status_code = res.status_code
        if status_code != 200:
            print("Error: GET failed with {}".format(status_code))
            retries -= 1
            continue

        body = res.text
        if "301 Permanent Redirect" not in body:
            print(body)
            sys.exit(0)

        match = LINK_REGEX.search(body)
        if not match:
            print("Error: Could not find next link in page")
            print(body)
            sys.exit(1)

        path_suffix = match.group(1)


if __name__ == "__main__":
    main()

It aint pretty, but I don't want to spend too much time on it. Feel free to improve on my approach. The main insight is that the paths all end with Unix epoch timestamps of when they were created, so get a redirect page close to now seconds since 1970 and follow the trail from there. Not extensively tested, so YMMV.