twitter_video.py

#!/usr/bin/env python3

import urllib.request
import json
import sys
import re
import shutil


# No, this isn't my own OAuth2 token. It isn't anyone's. It's the default token Twitter gives out to non-signedin users
AUTH_TOKEN = "AAAAAAAAAAAAAAAAAAAAANRILgAAAAAAnNwIzUejRCOuH5E6I8xnZz4puTs%3D1Zv7ttfk8LF81IUq16cHjhLTvJu4FA33AGWWjCpTnA"


def guest_token():
    req = urllib.request.Request(
        "https://api.twitter.com/1.1/guest/activate.json", method="POST"
    )
    req.add_header("Authorization", f"Bearer {AUTH_TOKEN}")
    with urllib.request.urlopen(req) as f:
        resp = f.read().decode("utf-8")
        resp = json.loads(resp)
        return resp["guest_token"]


def conversation(tweet_id):
    req = urllib.request.Request(
        f"https://api.twitter.com/2/timeline/conversation/{tweet_id}.json?tweet_mode=extended"
    )
    req.add_header("x-guest-token", guest_token())
    req.add_header("Authorization", f"Bearer {AUTH_TOKEN}")
    with urllib.request.urlopen(req) as f:
        resp = f.read().decode("utf-8")
        return json.dumps(json.loads(resp))


def find_video_urls(text):
    video_regexp = r"https:\/\/video.twimg.com\/[^\"]*?\.mp4"
    video_urls = re.findall(video_regexp, text)
    # rank by resolution, descending
    def resolution_string(src):
        resolution_regexp = r"(\d{3,5})x(\d{3,5})"
        m = re.search(resolution_regexp, src)
        if not m:
            return -1
        # e.g. 1280x720
        w, h = m.groups()
        w = int(w)
        h = int(h)
        return w + h

    # first video URL is highest resolution
    return sorted(video_urls, key=resolution_string, reverse=True)


def download_file(url, dest_filename):
    with open(dest_filename, "wb") as df:
        with urllib.request.urlopen(url) as sf:
            shutil.copyfileobj(sf, df)


if __name__ == "__main__":
    if len(sys.argv) < 2:
        print("Usage: python3 twitter_video.py <tweet URL>", file=sys.stderr)
        sys.exit(1)
    tweet_id_regexp = r"\/(\d+)\/?"
    m = re.search(tweet_id_regexp, sys.argv[1])
    if not m:
        print("That's not a recognized URL to a tweet", file=sys.stderr)
        sys.exit(1)
    tweet_id = m[1]
    payload = conversation(tweet_id)
    video_urls = find_video_urls(payload)
    if len(video_urls) == 0:
        print(
            "No videos found in this tweet. Make sure the URL is the tweet with the video. If there are, then this program could not find them and is probably outdated.",
            file=sys.stderr,
        )
        sys.exit(2)
    download_file(video_urls[0], f"{tweet_id}.mp4")