download_philarmonie_paris.py

"""
Script to dowload video from the ARTE recording of the Paris's Philarmonie.

Example:
python3 download.py "https://www.arte.tv/fr/videos/087078-007-A/kazuki-yamada-dirige-berlioz-et-jarrell/" 
python3 download.py "https://www.arte.tv/fr/videos/087078-007-A/kazuki-yamada-dirige-berlioz-et-jarrell/" my_output_name.mp4
"""
import os
import sys
import json
import argparse
import requests

from time import time


HEADERS = {'User-Agent': "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/75.0.3770.100 Safari/537.36"}

def download(url, filename):
    #source : https://sumit-ghosh.com/articles/python-download-progress-bar/
    with open(filename, 'wb') as f:
        response = requests.get(url, stream=True)
        total = response.headers.get('content-length')

        if total is None:
            f.write(response.content)
        else:
            downloaded = 0
            total = int(total)
            for data in response.iter_content(chunk_size=max(int(total / 1000), 1024 * 1024)):
                downloaded += len(data)
                f.write(data)
                done = int(50 * downloaded/total)
                sys.stdout.write('\r[{}{}]'.format('█' * done, '.' * (50-done)))
                sys.stdout.flush()
    sys.stdout.write('\n')

if __name__ == "__main__":
    if sys.version_info.major < 3 and sys.version_info.minor < 7:
        print("This script requires python 3.7 at least !")

    if not len(sys.argv[1:]):
        print("Usage : python3 video_page_url [out_path].")

    url = sys.argv[1]
    out_name = "" if not len(sys.argv) > 2 else sys.argv[2]

    # Get the API info url code from the video page url
    video_id = url.split("/videos/")[1].split('/')[0]
    video_infos_url = f"https://api.arte.tv/api/player/v1/config/fr/{video_id}?autostart=1&lifeCycle=1&lang=fr_FR&autostart=1&mute=0"

    # Start a session
    session = requests.Session()

    # Download videos info via the API
    r = session.get(video_infos_url, headers=HEADERS)
    if not r.ok:
        raise Exception(f"Failed to find infos... :'( error code : {r.status_code}")

    # Parse API result
    api_infos = json.loads(r.content.decode())

    # Find the highest bitrate mp4 video url
    video_candidates = api_infos['videoJsonPlayer']['VSR']
    video_candidates_keys = sorted(
        filter(
            lambda k: video_candidates[k]['mediaType'] == 'mp4',
            video_candidates
        ), 
        key=lambda k: -video_candidates[k]['bitrate']
    )
    if not len(video_candidates_keys):
        raise Exception("No mp4 url found... :'(")
    video_candidate_key = video_candidates_keys[0]
    video_url = video_candidates[video_candidate_key]['url']

    # Attribute an output file name
    if not out_name:
        valid_chars = '-_.() abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789'
        if 'VTI' in api_infos['videoJsonPlayer']:
            out_name = api_infos['videoJsonPlayer']['VTI']
            out_name = ''.join(c for c in out_name if c in valid_chars)
        elif 'VID' in api_infos['videoJsonPlayer']:
            out_name = api_infos['videoJsonPlayer']['VID']
            out_name = ''.join(c for c in out_name if c in valid_chars)
        else:
            out_name = f"{time()}"
        out_name = out_name + '.mp4'

    # Download and save content
    download(video_url, out_name)