playlist.py

# -*- coding: utf-8 -*-
import argparse
import pickle
import os
import re
from subprocess import Popen

import requests
from bs4 import BeautifulSoup


BASE_URL = "https://www.youtube.com/playlist?list={list_id:s}"
WATCH_URL = "https://www.youtube.com/watch?v={video_id:s}"
WATCH_LINK_FMT = "/watch\?v=([^\&]+)"
TEMP_PLAYLIST_FILE = "/tmp/temp_playlist_response.pkl"


def download_playlist(list_id):
    response = requests.get(BASE_URL.format(list_id=list_id))
    if response.status_code != 200:
        raise RuntimeError("Something is wrong")
    with open(TEMP_PLAYLIST_FILE, 'w') as f:
        pickle.dump(response, f)


def parse_playlist():
    with open(TEMP_PLAYLIST_FILE, 'r') as f:
        obj = pickle.load(f)

    soup = BeautifulSoup(obj.text, "html.parser")
    playlist_video_links = soup.find_all("a", attrs={
        "class": "pl-video-title-link",
        "href": re.compile('^/watch'),
    })

    for link_elem in playlist_video_links:
        href_attr = link_elem.get('href')
        video_id = re.match(WATCH_LINK_FMT, href_attr).group(1)

        p = Popen(["youtube-dl", WATCH_URL.format(video_id=video_id)])
        pid, sts = os.waitpid(p.pid, 0)


if __name__ == '__main__':
    p = argparse.ArgumentParser()
    p.add_argument('--list', '-l', type=str, required=True)
    args = p.parse_args()

    download_playlist(args.list)
    parse_playlist()