yt_dlp/extractor/hidive.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     int_or_none,
   5     try_get,
   6     url_or_none,
   7     urlencode_postdata,
   8 )
   9
  10
  11 class HiDiveIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?hidive\.com/stream/(?P<id>(?P<title>[^/]+)/(?P<key>[^/?#&]+))'
  13     # Using X-Forwarded-For results in 403 HTTP error for HLS fragments,
  14     # so disabling geo bypass completely
  15     _GEO_BYPASS = False
  16     _NETRC_MACHINE = 'hidive'
  17     _LOGIN_URL = 'https://www.hidive.com/account/login'
  18
  19     _TESTS = [{
  20         'url': 'https://www.hidive.com/stream/the-comic-artist-and-his-assistants/s01e001',
  21         'info_dict': {
  22             'id': 'the-comic-artist-and-his-assistants/s01e001',
  23             'ext': 'mp4',
  24             'title': 'the-comic-artist-and-his-assistants/s01e001',
  25             'series': 'the-comic-artist-and-his-assistants',
  26             'season_number': 1,
  27             'episode_number': 1,
  28         },
  29         'params': {
  30             'skip_download': True,
  31         },
  32         'skip': 'Requires Authentication',
  33     }]
  34
  35     def _perform_login(self, username, password):
  36         webpage = self._download_webpage(self._LOGIN_URL, None)
  37         form = self._search_regex(
  38             r'(?s)<form[^>]+action="/account/login"[^>]*>(.+?)</form>',
  39             webpage, 'login form', default=None)
  40         if not form:
  41             return
  42         data = self._hidden_inputs(form)
  43         data.update({
  44             'Email': username,
  45             'Password': password,
  46         })
  47         login_webpage = self._download_webpage(
  48             self._LOGIN_URL, None, 'Logging in', data=urlencode_postdata(data))
  49         # If the user has multiple profiles on their account, select one. For now pick the first profile.
  50         profile_id = self._search_regex(r'<button [^>]+?data-profile-id="(\w+)"', login_webpage, 'profile_id')
  51         if profile_id is None:
  52             return  # If only one profile, Hidive auto-selects it
  53         profile_id_hash = self._search_regex(r'\<button [^>]+?data-hash="(\w+)"', login_webpage, 'profile_id_hash')
  54         self._request_webpage(
  55             'https://www.hidive.com/ajax/chooseprofile', None,
  56             data=urlencode_postdata({
  57                 'profileId': profile_id,
  58                 'hash': profile_id_hash,
  59                 'returnUrl': '/dashboard'
  60             }))
  61
  62     def _call_api(self, video_id, title, key, data={}, **kwargs):
  63         data = {
  64             **data,
  65             'Title': title,
  66             'Key': key,
  67             'PlayerId': 'f4f895ce1ca713ba263b91caeb1daa2d08904783',
  68         }
  69         return self._download_json(
  70             'https://www.hidive.com/play/settings', video_id,
  71             data=urlencode_postdata(data), **kwargs) or {}
  72
  73     def _real_extract(self, url):
  74         video_id, title, key = self._match_valid_url(url).group('id', 'title', 'key')
  75         settings = self._call_api(video_id, title, key)
  76
  77         restriction = settings.get('restrictionReason')
  78         if restriction == 'RegionRestricted':
  79             self.raise_geo_restricted()
  80         if restriction and restriction != 'None':
  81             raise ExtractorError(
  82                 '%s said: %s' % (self.IE_NAME, restriction), expected=True)
  83
  84         formats, parsed_urls = [], {None}
  85         for rendition_id, rendition in settings['renditions'].items():
  86             audio, version, extra = rendition_id.split('_')
  87             m3u8_url = url_or_none(try_get(rendition, lambda x: x['bitrates']['hls']))
  88             if m3u8_url not in parsed_urls:
  89                 parsed_urls.add(m3u8_url)
  90                 frmt = self._extract_m3u8_formats(
  91                     m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id=rendition_id, fatal=False)
  92                 for f in frmt:
  93                     f['language'] = audio
  94                     f['format_note'] = f'{version}, {extra}'
  95                 formats.extend(frmt)
  96
  97         subtitles = {}
  98         for rendition_id, rendition in settings['renditions'].items():
  99             audio, version, extra = rendition_id.split('_')
 100             for cc_file in rendition.get('ccFiles') or []:
 101                 cc_url = url_or_none(try_get(cc_file, lambda x: x[2]))
 102                 cc_lang = try_get(cc_file, (lambda x: x[1].replace(' ', '-').lower(), lambda x: x[0]), str)
 103                 if cc_url not in parsed_urls and cc_lang:
 104                     parsed_urls.add(cc_url)
 105                     subtitles.setdefault(cc_lang, []).append({'url': cc_url})
 106
 107         return {
 108             'id': video_id,
 109             'title': video_id,
 110             'subtitles': subtitles,
 111             'formats': formats,
 112             'series': title,
 113             'season_number': int_or_none(
 114                 self._search_regex(r's(\d+)', key, 'season number', default=None)),
 115             'episode_number': int_or_none(
 116                 self._search_regex(r'e(\d+)', key, 'episode number', default=None)),
 117             'http_headers': {'Referer': url}
 118         }