logo

oasis-root

Compiled tree of Oasis Linux based on own branch at <https://hacktivis.me/git/oasis/> git clone https://anongit.hacktivis.me/git/oasis-root.git

dacast.py (7370B)

    

  1. import functools
  2. import hashlib
  3. import re
  4. import time
  6. from .common import InfoExtractor
  7. from ..networking.exceptions import HTTPError
  8. from ..utils import (
  9.     ExtractorError,
  10.     classproperty,
  11.     float_or_none,
  12.     traverse_obj,
  13.     url_or_none,
  14. )
  17. class DacastBaseIE(InfoExtractor):
  18.     _URL_TYPE = None
  20.     @classproperty
  21.     def _VALID_URL(cls):
  22.         return fr'https?://iframe\.dacast\.com/{cls._URL_TYPE}/(?P<user_id>[\w-]+)/(?P<id>[\w-]+)'
  24.     @classproperty
  25.     def _EMBED_REGEX(cls):
  26.         return [rf'<iframe[^>]+\bsrc=["\'](?P<url>{cls._VALID_URL})']
  28.     _API_INFO_URL = 'https://playback.dacast.com/content/info'
  30.     @classmethod
  31.     def _get_url_from_id(cls, content_id):
  32.         user_id, media_id = content_id.split(f'-{cls._URL_TYPE}-')
  33.         return f'https://iframe.dacast.com/{cls._URL_TYPE}/{user_id}/{media_id}'
  35.     @classmethod
  36.     def _extract_embed_urls(cls, url, webpage):
  37.         yield from super()._extract_embed_urls(url, webpage)
  38.         for content_id in re.findall(
  39.                 rf'<script[^>]+\bsrc=["\']https://player\.dacast\.com/js/player\.js\?contentId=([\w-]+-{cls._URL_TYPE}-[\w-]+)["\']', webpage):
  40.             yield cls._get_url_from_id(content_id)
  43. class DacastVODIE(DacastBaseIE):
  44.     _URL_TYPE = 'vod'
  45.     _TESTS = [{
  46.         'url': 'https://iframe.dacast.com/vod/acae82153ef4d7a7344ae4eaa86af534/1c6143e3-5a06-371d-8695-19b96ea49090',
  47.         'info_dict': {
  48.             'id': '1c6143e3-5a06-371d-8695-19b96ea49090',
  49.             'ext': 'mp4',
  50.             'uploader_id': 'acae82153ef4d7a7344ae4eaa86af534',
  51.             'title': '2_4||Adnexal mass characterisation: O-RADS US and MRI||N. Bharwani, London/UK',
  52.             'thumbnail': 'https://universe-files.dacast.com/26137208-5858-65c1-5e9a-9d6b6bd2b6c2',
  53.         },
  54.         'params': {'skip_download': 'm3u8'},
  55.     }, {  # /uspaes/ in hls_url
  56.         'url': 'https://iframe.dacast.com/vod/f9823fc6-faba-b98f-0d00-4a7b50a58c5b/348c5c84-b6af-4859-bb9d-1d01009c795b',
  57.         'info_dict': {
  58.             'id': '348c5c84-b6af-4859-bb9d-1d01009c795b',
  59.             'ext': 'mp4',
  60.             'title': 'pl1-edyta-rubas-211124.mp4',
  61.             'uploader_id': 'f9823fc6-faba-b98f-0d00-4a7b50a58c5b',
  62.             'thumbnail': 'https://universe-files.dacast.com/4d0bd042-a536-752d-fc34-ad2fa44bbcbb.png',
  63.         },
  64.     }]
  65.     _WEBPAGE_TESTS = [{
  66.         'url': 'https://www.dacast.com/support/knowledgebase/how-can-i-embed-a-video-on-my-website/',
  67.         'info_dict': {
  68.             'id': 'b6674869-f08a-23c5-1d7b-81f5309e1a90',
  69.             'ext': 'mp4',
  70.             'title': '4-HowToEmbedVideo.mp4',
  71.             'uploader_id': '3b67c4a9-3886-4eb1-d0eb-39b23b14bef3',
  72.             'thumbnail': 'https://universe-files.dacast.com/d26ab48f-a52a-8783-c42e-a90290ba06b6.png',
  73.         },
  74.         'params': {'skip_download': 'm3u8'},
  75.     }, {
  76.         'url': 'https://gist.githubusercontent.com/bashonly/4ad249ef2910346fbdf3809b220f11ee/raw/87349778d4af1a80b1fcc3beb9c88108de5858f5/dacast_embeds.html',
  77.         'info_dict': {
  78.             'id': 'e7df418e-a83b-7a7f-7b5e-1a667981e8fa',
  79.             'ext': 'mp4',
  80.             'title': 'Evening Service 2-5-23',
  81.             'uploader_id': '943bb1ab3c03695ba85330d92d6d226e',
  82.             'thumbnail': 'https://universe-files.dacast.com/337472b3-e92c-2ea4-7eb7-5700da477f67',
  83.         },
  84.         'params': {'skip_download': 'm3u8'},
  85.     }]
  87.     @functools.cached_property
  88.     def _usp_signing_secret(self):
  89.         player_js = self._download_webpage(
  90.             'https://player.dacast.com/js/player.js', None, 'Downloading player JS')
  91.         # Rotates every so often, but hardcode a fallback in case of JS change/breakage before rotation
  92.         return self._search_regex(
  93.             r'\bUSP_SIGNING_SECRET\s*=\s*(["\'])(?P<secret>(?:(?!\1).)+)', player_js,
  94.             'usp signing secret', group='secret', fatal=False) or 'odnInCGqhvtyRTtIiddxtuRtawYYICZP'
  96.     def _real_extract(self, url):
  97.         user_id, video_id = self._match_valid_url(url).group('user_id', 'id')
  98.         query = {'contentId': f'{user_id}-vod-{video_id}', 'provider': 'universe'}
  99.         info = self._download_json(self._API_INFO_URL, video_id, query=query, fatal=False)
  100.         access = self._download_json(
  101.             'https://playback.dacast.com/content/access', video_id,
  102.             note='Downloading access JSON', query=query, expected_status=403)
  104.         error = access.get('error')
  105.         if error in ('Broadcaster has been blocked', 'Content is offline'):
  106.             raise ExtractorError(error, expected=True)
  107.         elif error:
  108.             raise ExtractorError(f'Dacast API says "{error}"')
  110.         hls_url = access['hls']
  111.         hls_aes = {}
  113.         if 'DRM_EXT' in hls_url:
  114.             self.report_drm(video_id)
  115.         elif '/uspaes/' in hls_url:
  116.             # Ref: https://player.dacast.com/js/player.js
  117.             ts = int(time.time())
  118.             signature = hashlib.sha1(
  119.                 f'{10413792000 - ts}{ts}{self._usp_signing_secret}'.encode()).digest().hex()
  120.             hls_aes['uri'] = f'https://keys.dacast.com/uspaes/{video_id}.key?s={signature}&ts={ts}'
  122.         for retry in self.RetryManager():
  123.             try:
  124.                 formats = self._extract_m3u8_formats(hls_url, video_id, 'mp4', m3u8_id='hls')
  125.             except ExtractorError as e:
  126.                 # CDN will randomly respond with 403
  127.                 if isinstance(e.cause, HTTPError) and e.cause.status == 403:
  128.                     retry.error = e
  129.                     continue
  130.                 raise
  132.         return {
  133.             'id': video_id,
  134.             'uploader_id': user_id,
  135.             'formats': formats,
  136.             'hls_aes': hls_aes or None,
  137.             **traverse_obj(info, ('contentInfo', {
  138.                 'title': 'title',
  139.                 'duration': ('duration', {float_or_none}),
  140.                 'thumbnail': ('thumbnailUrl', {url_or_none}),
  141.             })),
  142.         }
  145. class DacastPlaylistIE(DacastBaseIE):
  146.     _URL_TYPE = 'playlist'
  147.     _TESTS = [{
  148.         'url': 'https://iframe.dacast.com/playlist/943bb1ab3c03695ba85330d92d6d226e/b632eb053cac17a9c9a02bcfc827f2d8',
  149.         'playlist_mincount': 28,
  150.         'info_dict': {
  151.             'id': 'b632eb053cac17a9c9a02bcfc827f2d8',
  152.             'title': 'Archive Sermons',
  153.         },
  154.     }]
  155.     _WEBPAGE_TESTS = [{
  156.         'url': 'https://gist.githubusercontent.com/bashonly/7efb606f49f3c6e07ea0327de5a661d1/raw/05a16eac830245ea301fb0a585023bec71e6093c/dacast_playlist_embed.html',
  157.         'playlist_mincount': 28,
  158.         'info_dict': {
  159.             'id': 'b632eb053cac17a9c9a02bcfc827f2d8',
  160.             'title': 'Archive Sermons',
  161.         },
  162.     }]
  164.     def _real_extract(self, url):
  165.         user_id, playlist_id = self._match_valid_url(url).group('user_id', 'id')
  166.         info = self._download_json(
  167.             self._API_INFO_URL, playlist_id, note='Downloading playlist JSON', query={
  168.                 'contentId': f'{user_id}-playlist-{playlist_id}',
  169.                 'provider': 'universe',
  170.             })['contentInfo']
  172.         def entries(info):
  173.             for video in traverse_obj(info, ('features', 'playlist', 'contents', lambda _, v: v['id'])):
  174.                 yield self.url_result(
  175.                     DacastVODIE._get_url_from_id(video['id']), DacastVODIE, video['id'], video.get('title'))
  177.         return self.playlist_result(entries(info), playlist_id, info.get('title'))