yt_dlp/extractor/glomex.py

   1 import re
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     determine_ext,
   7     extract_attributes,
   8     ExtractorError,
   9     int_or_none,
  10     parse_qs,
  11     smuggle_url,
  12     unescapeHTML,
  13     unsmuggle_url,
  14 )
  15
  16
  17 class GlomexBaseIE(InfoExtractor):
  18     _DEFAULT_ORIGIN_URL = 'https://player.glomex.com/'
  19     _API_URL = 'https://integration-cloudfront-eu-west-1.mes.glomex.cloud/'
  20
  21     @staticmethod
  22     def _smuggle_origin_url(url, origin_url):
  23         if origin_url is None:
  24             return url
  25         return smuggle_url(url, {'origin': origin_url})
  26
  27     @classmethod
  28     def _unsmuggle_origin_url(cls, url, fallback_origin_url=None):
  29         defaults = {'origin': fallback_origin_url or cls._DEFAULT_ORIGIN_URL}
  30         unsmuggled_url, data = unsmuggle_url(url, default=defaults)
  31         return unsmuggled_url, data['origin']
  32
  33     def _get_videoid_type(self, video_id):
  34         _VIDEOID_TYPES = {
  35             'v': 'video',
  36             'pl': 'playlist',
  37             'rl': 'related videos playlist',
  38             'cl': 'curated playlist',
  39         }
  40         prefix = video_id.split('-')[0]
  41         return _VIDEOID_TYPES.get(prefix, 'unknown type')
  42
  43     def _download_api_data(self, video_id, integration, current_url=None):
  44         query = {
  45             'integration_id': integration,
  46             'playlist_id': video_id,
  47             'current_url': current_url or self._DEFAULT_ORIGIN_URL,
  48         }
  49         video_id_type = self._get_videoid_type(video_id)
  50         return self._download_json(
  51             self._API_URL,
  52             video_id, 'Downloading %s JSON' % video_id_type,
  53             'Unable to download %s JSON' % video_id_type,
  54             query=query)
  55
  56     def _download_and_extract_api_data(self, video_id, integration, current_url):
  57         api_data = self._download_api_data(video_id, integration, current_url)
  58         videos = api_data['videos']
  59         if not videos:
  60             raise ExtractorError('no videos found for %s' % video_id)
  61         videos = [self._extract_api_data(video, video_id) for video in videos]
  62         return videos[0] if len(videos) == 1 else self.playlist_result(videos, video_id)
  63
  64     def _extract_api_data(self, video, video_id):
  65         if video.get('error_code') == 'contentGeoblocked':
  66             self.raise_geo_restricted(countries=video['geo_locations'])
  67
  68         formats, subs = [], {}
  69         for format_id, format_url in video['source'].items():
  70             ext = determine_ext(format_url)
  71             if ext == 'm3u8':
  72                 formats_, subs_ = self._extract_m3u8_formats_and_subtitles(
  73                     format_url, video_id, 'mp4', m3u8_id=format_id,
  74                     fatal=False)
  75                 formats.extend(formats_)
  76                 self._merge_subtitles(subs_, target=subs)
  77             else:
  78                 formats.append({
  79                     'url': format_url,
  80                     'format_id': format_id,
  81                 })
  82         if video.get('language'):
  83             for fmt in formats:
  84                 fmt['language'] = video['language']
  85         self._sort_formats(formats)
  86
  87         images = (video.get('images') or []) + [video.get('image') or {}]
  88         thumbnails = [{
  89             'id': image.get('id'),
  90             'url': f'{image["url"]}/profile:player-960x540',
  91             'width': 960,
  92             'height': 540,
  93         } for image in images if image.get('url')]
  94         self._remove_duplicate_formats(thumbnails)
  95
  96         return {
  97             'id': video.get('clip_id') or video_id,
  98             'title': video.get('title'),
  99             'description': video.get('description'),
 100             'thumbnails': thumbnails,
 101             'duration': int_or_none(video.get('clip_duration')),
 102             'timestamp': video.get('created_at'),
 103             'formats': formats,
 104             'subtitles': subs,
 105         }
 106
 107
 108 class GlomexIE(GlomexBaseIE):
 109     IE_NAME = 'glomex'
 110     IE_DESC = 'Glomex videos'
 111     _VALID_URL = r'https?://video\.glomex\.com/[^/]+/(?P<id>v-[^-]+)'
 112     _INTEGRATION_ID = '19syy24xjn1oqlpc'
 113
 114     _TESTS = [{
 115         'url': 'https://video.glomex.com/sport/v-cb24uwg77hgh-nach-2-0-sieg-guardiola-mit-mancity-vor-naechstem-titel',
 116         'md5': 'cec33a943c4240c9cb33abea8c26242e',
 117         'info_dict': {
 118             'id': 'v-cb24uwg77hgh',
 119             'ext': 'mp4',
 120             'title': 'md5:38a90cedcfadd72982c81acf13556e0c',
 121             'description': 'md5:1ea6b6caff1443fcbbba159e432eedb8',
 122             'duration': 29600,
 123             'timestamp': 1619895017,
 124             'upload_date': '20210501',
 125         },
 126     }]
 127
 128     def _real_extract(self, url):
 129         video_id = self._match_id(url)
 130         return self.url_result(
 131             GlomexEmbedIE.build_player_url(video_id, self._INTEGRATION_ID, url),
 132             GlomexEmbedIE.ie_key(), video_id)
 133
 134
 135 class GlomexEmbedIE(GlomexBaseIE):
 136     IE_NAME = 'glomex:embed'
 137     IE_DESC = 'Glomex embedded videos'
 138     _BASE_PLAYER_URL = '//player.glomex.com/integration/1/iframe-player.html'
 139     _BASE_PLAYER_URL_RE = re.escape(_BASE_PLAYER_URL).replace('/1/', r'/[^/]/')
 140     _VALID_URL = rf'https?:{_BASE_PLAYER_URL_RE}\?([^#]+&)?playlistId=(?P<id>[^#&]+)'
 141
 142     _TESTS = [{
 143         'url': 'https://player.glomex.com/integration/1/iframe-player.html?integrationId=4059a013k56vb2yd&playlistId=v-cfa6lye0dkdd-sf',
 144         'md5': '68f259b98cc01918ac34180142fce287',
 145         'info_dict': {
 146             'id': 'v-cfa6lye0dkdd-sf',
 147             'ext': 'mp4',
 148             'timestamp': 1635337199,
 149             'duration': 133080,
 150             'upload_date': '20211027',
 151             'description': 'md5:e741185fc309310ff5d0c789b437be66',
 152             'title': 'md5:35647293513a6c92363817a0fb0a7961',
 153         },
 154     }, {
 155         'url': 'https://player.glomex.com/integration/1/iframe-player.html?origin=fullpage&integrationId=19syy24xjn1oqlpc&playlistId=rl-vcb49w1fb592p&playlistIndex=0',
 156         'info_dict': {
 157             'id': 'rl-vcb49w1fb592p',
 158         },
 159         'playlist_count': 100,
 160     }, {
 161         'url': 'https://player.glomex.com/integration/1/iframe-player.html?playlistId=cl-bgqaata6aw8x&integrationId=19syy24xjn1oqlpc',
 162         'info_dict': {
 163             'id': 'cl-bgqaata6aw8x',
 164         },
 165         'playlist_mincount': 2,
 166     }]
 167
 168     @classmethod
 169     def build_player_url(cls, video_id, integration, origin_url=None):
 170         query_string = urllib.parse.urlencode({
 171             'playlistId': video_id,
 172             'integrationId': integration,
 173         })
 174         return cls._smuggle_origin_url(f'https:{cls._BASE_PLAYER_URL}?{query_string}', origin_url)
 175
 176     @classmethod
 177     def _extract_urls(cls, webpage, origin_url):
 178         # https://docs.glomex.com/publisher/video-player-integration/javascript-api/
 179         quot_re = r'["\']'
 180
 181         regex = fr'''(?x)
 182             <iframe[^>]+?src=(?P<q>{quot_re})(?P<url>
 183                 (?:https?:)?{cls._BASE_PLAYER_URL_RE}\?(?:(?!(?P=q)).)+
 184             )(?P=q)'''
 185         for mobj in re.finditer(regex, webpage):
 186             url = unescapeHTML(mobj.group('url'))
 187             if cls.suitable(url):
 188                 yield cls._smuggle_origin_url(url, origin_url)
 189
 190         regex = fr'''(?x)
 191             <glomex-player [^>]+?>|
 192             <div[^>]* data-glomex-player=(?P<q>{quot_re})true(?P=q)[^>]*>'''
 193         for mobj in re.finditer(regex, webpage):
 194             attrs = extract_attributes(mobj.group(0))
 195             if attrs.get('data-integration-id') and attrs.get('data-playlist-id'):
 196                 yield cls.build_player_url(attrs['data-playlist-id'], attrs['data-integration-id'], origin_url)
 197
 198         # naive parsing of inline scripts for hard-coded integration parameters
 199         regex = fr'''(?x)
 200             (?P<is_js>dataset\.)?%s\s*(?(is_js)=|:)\s*
 201             (?P<q>{quot_re})(?P<id>(?:(?!(?P=q)).)+)(?P=q)\s'''
 202         for mobj in re.finditer(r'(?x)<script[^<]*>.+?</script>', webpage):
 203             script = mobj.group(0)
 204             integration_id = re.search(regex % 'integrationId', script)
 205             if not integration_id:
 206                 continue
 207             playlist_id = re.search(regex % 'playlistId', script)
 208             if playlist_id:
 209                 yield cls.build_player_url(playlist_id, integration_id, origin_url)
 210
 211     def _real_extract(self, url):
 212         url, origin_url = self._unsmuggle_origin_url(url)
 213         playlist_id = self._match_id(url)
 214         integration = parse_qs(url).get('integrationId', [None])[0]
 215         if not integration:
 216             raise ExtractorError('No integrationId in URL', expected=True)
 217         return self._download_and_extract_api_data(playlist_id, integration, origin_url)