yt_dlp/extractor/hidive.py

   1 # coding: utf-8
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     int_or_none,
   8     try_get,
   9     url_or_none,
  10     urlencode_postdata,
  11 )
  12
  13
  14 class HiDiveIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?hidive\.com/stream/(?P<id>(?P<title>[^/]+)/(?P<key>[^/?#&]+))'
  16     # Using X-Forwarded-For results in 403 HTTP error for HLS fragments,
  17     # so disabling geo bypass completely
  18     _GEO_BYPASS = False
  19     _NETRC_MACHINE = 'hidive'
  20     _LOGIN_URL = 'https://www.hidive.com/account/login'
  21
  22     _TESTS = [{
  23         'url': 'https://www.hidive.com/stream/the-comic-artist-and-his-assistants/s01e001',
  24         'info_dict': {
  25             'id': 'the-comic-artist-and-his-assistants/s01e001',
  26             'ext': 'mp4',
  27             'title': 'the-comic-artist-and-his-assistants/s01e001',
  28             'series': 'the-comic-artist-and-his-assistants',
  29             'season_number': 1,
  30             'episode_number': 1,
  31         },
  32         'params': {
  33             'skip_download': True,
  34         },
  35         'skip': 'Requires Authentication',
  36     }]
  37
  38     def _perform_login(self, username, password):
  39         webpage = self._download_webpage(self._LOGIN_URL, None)
  40         form = self._search_regex(
  41             r'(?s)<form[^>]+action="/account/login"[^>]*>(.+?)</form>',
  42             webpage, 'login form')
  43         data = self._hidden_inputs(form)
  44         data.update({
  45             'Email': username,
  46             'Password': password,
  47         })
  48         self._download_webpage(
  49             self._LOGIN_URL, None, 'Logging in', data=urlencode_postdata(data))
  50
  51     def _call_api(self, video_id, title, key, data={}, **kwargs):
  52         data = {
  53             **data,
  54             'Title': title,
  55             'Key': key,
  56             'PlayerId': 'f4f895ce1ca713ba263b91caeb1daa2d08904783',
  57         }
  58         return self._download_json(
  59             'https://www.hidive.com/play/settings', video_id,
  60             data=urlencode_postdata(data), **kwargs) or {}
  61
  62     def _extract_subtitles_from_rendition(self, rendition, subtitles, parsed_urls):
  63         for cc_file in rendition.get('ccFiles', []):
  64             cc_url = url_or_none(try_get(cc_file, lambda x: x[2]))
  65             # name is used since we cant distinguish subs with same language code
  66             cc_lang = try_get(cc_file, (lambda x: x[1].replace(' ', '-').lower(), lambda x: x[0]), str)
  67             if cc_url not in parsed_urls and cc_lang:
  68                 parsed_urls.add(cc_url)
  69                 subtitles.setdefault(cc_lang, []).append({'url': cc_url})
  70
  71     def _get_subtitles(self, url, video_id, title, key, parsed_urls):
  72         webpage = self._download_webpage(url, video_id, fatal=False) or ''
  73         subtitles = {}
  74         for caption in set(re.findall(r'data-captions=\"([^\"]+)\"', webpage)):
  75             renditions = self._call_api(
  76                 video_id, title, key, {'Captions': caption}, fatal=False,
  77                 note=f'Downloading {caption} subtitle information').get('renditions') or {}
  78             for rendition_id, rendition in renditions.items():
  79                 self._extract_subtitles_from_rendition(rendition, subtitles, parsed_urls)
  80         return subtitles
  81
  82     def _real_extract(self, url):
  83         video_id, title, key = self._match_valid_url(url).group('id', 'title', 'key')
  84         settings = self._call_api(video_id, title, key)
  85
  86         restriction = settings.get('restrictionReason')
  87         if restriction == 'RegionRestricted':
  88             self.raise_geo_restricted()
  89         if restriction and restriction != 'None':
  90             raise ExtractorError(
  91                 '%s said: %s' % (self.IE_NAME, restriction), expected=True)
  92
  93         formats, parsed_urls = [], {None}
  94         for rendition_id, rendition in settings['renditions'].items():
  95             audio, version, extra = rendition_id.split('_')
  96             m3u8_url = url_or_none(try_get(rendition, lambda x: x['bitrates']['hls']))
  97             if m3u8_url not in parsed_urls:
  98                 parsed_urls.add(m3u8_url)
  99                 frmt = self._extract_m3u8_formats(
 100                     m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id=rendition_id, fatal=False)
 101                 for f in frmt:
 102                     f['language'] = audio
 103                     f['format_note'] = f'{version}, {extra}'
 104                 formats.extend(frmt)
 105         self._sort_formats(formats)
 106
 107         return {
 108             'id': video_id,
 109             'title': video_id,
 110             'subtitles': self.extract_subtitles(url, video_id, title, key, parsed_urls),
 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         }