yt_dlp/extractor/atvat.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import datetime
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     float_or_none,
   9     jwt_encode_hs256,
  10     try_get,
  11 )
  12
  13
  14 class ATVAtIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?atv\.at/tv/(?:[^/]+/){2,3}(?P<id>.*)'
  16
  17     _TESTS = [{
  18         'url': 'https://www.atv.at/tv/bauer-sucht-frau/staffel-18/bauer-sucht-frau/bauer-sucht-frau-staffel-18-folge-3-die-hofwochen',
  19         'md5': '3c3b4aaca9f63e32b35e04a9c2515903',
  20         'info_dict': {
  21             'id': 'v-ce9cgn1e70n5-1',
  22             'ext': 'mp4',
  23             'title': 'Bauer sucht Frau - Staffel 18 Folge 3 - Die Hofwochen',
  24         }
  25     }, {
  26         'url': 'https://www.atv.at/tv/bauer-sucht-frau/staffel-18/episode-01/bauer-sucht-frau-staffel-18-vorstellungsfolge-1',
  27         'only_matching': True,
  28     }]
  29
  30     # extracted from bootstrap.js function (search for e.encryption_key and use your browser's debugger)
  31     _ACCESS_ID = 'x_atv'
  32     _ENCRYPTION_KEY = 'Hohnaekeishoogh2omaeghooquooshia'
  33
  34     def _extract_video_info(self, url, content, video):
  35         clip_id = content.get('splitId', content['id'])
  36         formats = []
  37         clip_urls = video['urls']
  38         for protocol, variant in clip_urls.items():
  39             source_url = try_get(variant, lambda x: x['clear']['url'])
  40             if not source_url:
  41                 continue
  42             if protocol == 'dash':
  43                 formats.extend(self._extract_mpd_formats(
  44                     source_url, clip_id, mpd_id=protocol, fatal=False))
  45             elif protocol == 'hls':
  46                 formats.extend(self._extract_m3u8_formats(
  47                     source_url, clip_id, 'mp4', 'm3u8_native',
  48                     m3u8_id=protocol, fatal=False))
  49             else:
  50                 formats.append({
  51                     'url': source_url,
  52                     'format_id': protocol,
  53                 })
  54         self._sort_formats(formats)
  55
  56         return {
  57             'id': clip_id,
  58             'title': content.get('title'),
  59             'duration': float_or_none(content.get('duration')),
  60             'series': content.get('tvShowTitle'),
  61             'formats': formats,
  62         }
  63
  64     def _real_extract(self, url):
  65         video_id = self._match_id(url)
  66         webpage = self._download_webpage(url, video_id)
  67         json_data = self._parse_json(
  68             self._search_regex(r'<script id="state" type="text/plain">(.*)</script>', webpage, 'json_data'),
  69             video_id=video_id)
  70
  71         video_title = json_data['views']['default']['page']['title']
  72         contentResource = json_data['views']['default']['page']['contentResource']
  73         content_id = contentResource[0]['id']
  74         content_ids = [{'id': id, 'subclip_start': content['start'], 'subclip_end': content['end']}
  75                        for id, content in enumerate(contentResource)]
  76
  77         time_of_request = datetime.datetime.now()
  78         not_before = time_of_request - datetime.timedelta(minutes=5)
  79         expire = time_of_request + datetime.timedelta(minutes=5)
  80         payload = {
  81             'content_ids': {
  82                 content_id: content_ids,
  83             },
  84             'secure_delivery': True,
  85             'iat': int(time_of_request.timestamp()),
  86             'nbf': int(not_before.timestamp()),
  87             'exp': int(expire.timestamp()),
  88         }
  89         jwt_token = jwt_encode_hs256(payload, self._ENCRYPTION_KEY, headers={'kid': self._ACCESS_ID})
  90         videos = self._download_json(
  91             'https://vas-v4.p7s1video.net/4.0/getsources',
  92             content_id, 'Downloading videos JSON', query={
  93                 'token': jwt_token.decode('utf-8')
  94             })
  95
  96         video_id, videos_data = list(videos['data'].items())[0]
  97         entries = [
  98             self._extract_video_info(url, contentResource[video['id']], video)
  99             for video in videos_data]
 100
 101         return {
 102             '_type': 'multi_video',
 103             'id': video_id,
 104             'title': video_title,
 105             'entries': entries,
 106         }