Download Youtube Audio with multi threading

download.py

import pandas as pd
import youtube_dl
import os

from youtube_dl.utils import DownloadError

from concurrent.futures import ThreadPoolExecutor, wait, as_completed


def download_youtube(youtube_id, idx):
    ydl_opts = {
        'format': 'm4a',
        'outtmpl': './audio/%(id)s.%(ext)s'
    }
    with youtube_dl.YoutubeDL(ydl_opts) as ydl:

        try:
	        ydl.download([youtube_id])
        except DownloadError as e:
            print("skip")
            print(str(e))

    print("[{}] Done: {}".format(idx, youtube_id))
    return youtube_id


df = pd.read_csv('../notebook/new_lyric.csv')

print(len(df))
df.dropna(subset=['youtube'], inplace=True)
print(len(df))

pool = ThreadPoolExecutor(4)
futures = []

for idx, row in df.iterrows():

    youtube_id = row['youtube']
    title = row["title"]
    print("[{}] {}: {}".format(idx, title, youtube_id))

    filename = "./audio/{}.m4a".format(youtube_id)

    if os.path.isfile(filename):
        print("{}: Skip file exist".format(youtube_id))
        continue

    futures.append(pool.submit(download_youtube, youtube_id, idx))

for x in as_completed(futures):
    print("{} completed".format(x.result()))