xkcd-countdown.py

#!/usr/bin/env python3

# A small script to collect the xkcd countdown frames from munvoseli's
# xkcd-countdown page or from explainxkcd and render them as a video.

# Requires beautifulsoup4 and requests
# Requires ffmpeg to be installed

import argparse
from pathlib import Path
import re
import subprocess

import bs4
import requests

FPS = 10
TMP = "tmp"
OUT = "out.mp4"

TMP_PATH = Path(TMP)
OUT_PATH = Path(OUT)

def image_names():
    i = 1
    while True:
        yield f"{i:04}.png"
        i += 1

def crawl_munvoseli():
    r = requests.get("https://raw.githubusercontent.com/munvoseli/xkcd-countdown/master/framestr.js")
    for line in re.fullmatch(r".*`(.*)`.*", r.text, flags=re.DOTALL).group(1).splitlines():
        img_hash = re.fullmatch(r"(\S+)\s.*", line).group(1)
        yield f"https://xkcd.com/count-wimRikmef/imgs/{img_hash}.png"

def crawl_explainxkcd():
    r = requests.get("https://www.explainxkcd.com/wiki/index.php/Countdown_in_header_text/images")
    soup = bs4.BeautifulSoup(r.text, "html.parser")
    image_ul = soup.find(id = "Images").parent.find_next_sibling("ul")

    for image in image_ul.find_all("img"):
        yield f"https://www.explainxkcd.com/{image['src']}"

def render_result():
    OUT_PATH.unlink(missing_ok=True)
    subprocess.run(["ffmpeg", "-r", f"{FPS}", "-i", "tmp/%04d.png", "-pix_fmt", "yuv420p", OUT_PATH], capture_output=True)

def main():
    parser = argparse.ArgumentParser()
    parser.add_argument("--explainxkcd", action="store_true",
            help="Use explainxkcd.com instead of munvoseli.github.com/xkcd-countdown")
    args = parser.parse_args()

    crawl = crawl_explainxkcd if args.explainxkcd else crawl_munvoseli

    tmp = Path("tmp")
    tmp.mkdir(parents=True, exist_ok=True)

    for (url, name) in zip(crawl(), image_names()):
        path = TMP_PATH / name

        if path.exists():
            print(f"Skipping {name}")
            continue

        print(f"Downloading {name} from {url}")
        r = requests.get(url)
        with open(path, "wb") as f:
            for chunk in r.iter_content(chunk_size=1024):
                f.write(chunk)

    print(f"Rendering clip to {OUT_PATH}")
    render_result()

if __name__ == "__main__":
    main()