yt_dlp/extractor/dacast.py

   1 import hashlib
   2 import re
   3 import time
   4
   5 from .common import InfoExtractor
   6 from ..networking.exceptions import HTTPError
   7 from ..utils import (
   8     ExtractorError,
   9     classproperty,
  10     float_or_none,
  11     traverse_obj,
  12     url_or_none,
  13 )
  14
  15
  16 class DacastBaseIE(InfoExtractor):
  17     _URL_TYPE = None
  18
  19     @classproperty
  20     def _VALID_URL(cls):
  21         return fr'https?://iframe\.dacast\.com/{cls._URL_TYPE}/(?P<user_id>[\w-]+)/(?P<id>[\w-]+)'
  22
  23     @classproperty
  24     def _EMBED_REGEX(cls):
  25         return [rf'<iframe[^>]+\bsrc=["\'](?P<url>{cls._VALID_URL})']
  26
  27     _API_INFO_URL = 'https://playback.dacast.com/content/info'
  28
  29     @classmethod
  30     def _get_url_from_id(cls, content_id):
  31         user_id, media_id = content_id.split(f'-{cls._URL_TYPE}-')
  32         return f'https://iframe.dacast.com/{cls._URL_TYPE}/{user_id}/{media_id}'
  33
  34     @classmethod
  35     def _extract_embed_urls(cls, url, webpage):
  36         yield from super()._extract_embed_urls(url, webpage)
  37         for content_id in re.findall(
  38                 rf'<script[^>]+\bsrc=["\']https://player\.dacast\.com/js/player\.js\?contentId=([\w-]+-{cls._URL_TYPE}-[\w-]+)["\']', webpage):
  39             yield cls._get_url_from_id(content_id)
  40
  41
  42 class DacastVODIE(DacastBaseIE):
  43     _URL_TYPE = 'vod'
  44     _TESTS = [{
  45         'url': 'https://iframe.dacast.com/vod/acae82153ef4d7a7344ae4eaa86af534/1c6143e3-5a06-371d-8695-19b96ea49090',
  46         'info_dict': {
  47             'id': '1c6143e3-5a06-371d-8695-19b96ea49090',
  48             'ext': 'mp4',
  49             'uploader_id': 'acae82153ef4d7a7344ae4eaa86af534',
  50             'title': '2_4||Adnexal mass characterisation: O-RADS US and MRI||N. Bharwani, London/UK',
  51             'thumbnail': 'https://universe-files.dacast.com/26137208-5858-65c1-5e9a-9d6b6bd2b6c2',
  52         },
  53         'params': {'skip_download': 'm3u8'},
  54     }]
  55     _WEBPAGE_TESTS = [{
  56         'url': 'https://www.dacast.com/support/knowledgebase/how-can-i-embed-a-video-on-my-website/',
  57         'info_dict': {
  58             'id': 'b6674869-f08a-23c5-1d7b-81f5309e1a90',
  59             'ext': 'mp4',
  60             'title': '4-HowToEmbedVideo.mp4',
  61             'uploader_id': '3b67c4a9-3886-4eb1-d0eb-39b23b14bef3',
  62             'thumbnail': 'https://universe-files.dacast.com/d26ab48f-a52a-8783-c42e-a90290ba06b6.png',
  63         },
  64         'params': {'skip_download': 'm3u8'},
  65     }, {
  66         'url': 'https://gist.githubusercontent.com/bashonly/4ad249ef2910346fbdf3809b220f11ee/raw/87349778d4af1a80b1fcc3beb9c88108de5858f5/dacast_embeds.html',
  67         'info_dict': {
  68             'id': 'e7df418e-a83b-7a7f-7b5e-1a667981e8fa',
  69             'ext': 'mp4',
  70             'title': 'Evening Service 2-5-23',
  71             'uploader_id': '943bb1ab3c03695ba85330d92d6d226e',
  72             'thumbnail': 'https://universe-files.dacast.com/337472b3-e92c-2ea4-7eb7-5700da477f67',
  73         },
  74         'params': {'skip_download': 'm3u8'},
  75     }]
  76
  77     def _real_extract(self, url):
  78         user_id, video_id = self._match_valid_url(url).group('user_id', 'id')
  79         query = {'contentId': f'{user_id}-vod-{video_id}', 'provider': 'universe'}
  80         info = self._download_json(self._API_INFO_URL, video_id, query=query, fatal=False)
  81         access = self._download_json(
  82             'https://playback.dacast.com/content/access', video_id,
  83             note='Downloading access JSON', query=query, expected_status=403)
  84
  85         error = access.get('error')
  86         if error in ('Broadcaster has been blocked', 'Content is offline'):
  87             raise ExtractorError(error, expected=True)
  88         elif error:
  89             raise ExtractorError(f'Dacast API says "{error}"')
  90
  91         hls_url = access['hls']
  92         hls_aes = {}
  93
  94         if 'DRM_EXT' in hls_url:
  95             self.report_drm(video_id)
  96         elif '/uspaes/' in hls_url:
  97             # From https://player.dacast.com/js/player.js
  98             ts = int(time.time())
  99             signature = hashlib.sha1(
 100                 f'{10413792000 - ts}{ts}YfaKtquEEpDeusCKbvYszIEZnWmBcSvw').digest().hex()
 101             hls_aes['uri'] = f'https://keys.dacast.com/uspaes/{video_id}.key?s={signature}&ts={ts}'
 102
 103         for retry in self.RetryManager():
 104             try:
 105                 formats = self._extract_m3u8_formats(hls_url, video_id, 'mp4', m3u8_id='hls')
 106             except ExtractorError as e:
 107                 # CDN will randomly respond with 403
 108                 if isinstance(e.cause, HTTPError) and e.cause.status == 403:
 109                     retry.error = e
 110                     continue
 111                 raise
 112
 113         return {
 114             'id': video_id,
 115             'uploader_id': user_id,
 116             'formats': formats,
 117             'hls_aes': hls_aes or None,
 118             **traverse_obj(info, ('contentInfo', {
 119                 'title': 'title',
 120                 'duration': ('duration', {float_or_none}),
 121                 'thumbnail': ('thumbnailUrl', {url_or_none}),
 122             })),
 123         }
 124
 125
 126 class DacastPlaylistIE(DacastBaseIE):
 127     _URL_TYPE = 'playlist'
 128     _TESTS = [{
 129         'url': 'https://iframe.dacast.com/playlist/943bb1ab3c03695ba85330d92d6d226e/b632eb053cac17a9c9a02bcfc827f2d8',
 130         'playlist_mincount': 28,
 131         'info_dict': {
 132             'id': 'b632eb053cac17a9c9a02bcfc827f2d8',
 133             'title': 'Archive Sermons',
 134         },
 135     }]
 136     _WEBPAGE_TESTS = [{
 137         'url': 'https://gist.githubusercontent.com/bashonly/7efb606f49f3c6e07ea0327de5a661d1/raw/05a16eac830245ea301fb0a585023bec71e6093c/dacast_playlist_embed.html',
 138         'playlist_mincount': 28,
 139         'info_dict': {
 140             'id': 'b632eb053cac17a9c9a02bcfc827f2d8',
 141             'title': 'Archive Sermons',
 142         },
 143     }]
 144
 145     def _real_extract(self, url):
 146         user_id, playlist_id = self._match_valid_url(url).group('user_id', 'id')
 147         info = self._download_json(
 148             self._API_INFO_URL, playlist_id, note='Downloading playlist JSON', query={
 149                 'contentId': f'{user_id}-playlist-{playlist_id}',
 150                 'provider': 'universe',
 151             })['contentInfo']
 152
 153         def entries(info):
 154             for video in traverse_obj(info, ('features', 'playlist', 'contents', lambda _, v: v['id'])):
 155                 yield self.url_result(
 156                     DacastVODIE._get_url_from_id(video['id']), DacastVODIE, video['id'], video.get('title'))
 157
 158         return self.playlist_result(entries(info), playlist_id, info.get('title'))