yt_dlp/extractor/glomex.py

   1 import re
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     determine_ext,
   8     extract_attributes,
   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, f'Downloading {video_id_type} JSON',
  53             f'Unable to download {video_id_type} JSON',
  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(f'no videos found for {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
  86         images = (video.get('images') or []) + [video.get('image') or {}]
  87         thumbnails = [{
  88             'id': image.get('id'),
  89             'url': f'{image["url"]}/profile:player-960x540',
  90             'width': 960,
  91             'height': 540,
  92         } for image in images if image.get('url')]
  93         self._remove_duplicate_formats(thumbnails)
  94
  95         return {
  96             'id': video.get('clip_id') or video_id,
  97             'title': video.get('title'),
  98             'description': video.get('description'),
  99             'thumbnails': thumbnails,
 100             'duration': int_or_none(video.get('clip_duration')),
 101             'timestamp': video.get('created_at'),
 102             'formats': formats,
 103             'subtitles': subs,
 104         }
 105
 106
 107 class GlomexIE(GlomexBaseIE):
 108     IE_NAME = 'glomex'
 109     IE_DESC = 'Glomex videos'
 110     _VALID_URL = r'https?://video\.glomex\.com/[^/]+/(?P<id>v-[^-]+)'
 111     _INTEGRATION_ID = '19syy24xjn1oqlpc'
 112
 113     _TESTS = [{
 114         'url': 'https://video.glomex.com/sport/v-cb24uwg77hgh-nach-2-0-sieg-guardiola-mit-mancity-vor-naechstem-titel',
 115         'md5': 'cec33a943c4240c9cb33abea8c26242e',
 116         'info_dict': {
 117             'id': 'v-cb24uwg77hgh',
 118             'ext': 'mp4',
 119             'title': 'md5:38a90cedcfadd72982c81acf13556e0c',
 120             'description': 'md5:1ea6b6caff1443fcbbba159e432eedb8',
 121             'duration': 29600,
 122             'timestamp': 1619895017,
 123             'upload_date': '20210501',
 124         },
 125     }]
 126
 127     def _real_extract(self, url):
 128         video_id = self._match_id(url)
 129         return self.url_result(
 130             GlomexEmbedIE.build_player_url(video_id, self._INTEGRATION_ID, url),
 131             GlomexEmbedIE.ie_key(), video_id)
 132
 133
 134 class GlomexEmbedIE(GlomexBaseIE):
 135     IE_NAME = 'glomex:embed'
 136     IE_DESC = 'Glomex embedded videos'
 137     _BASE_PLAYER_URL = '//player.glomex.com/integration/1/iframe-player.html'
 138     _BASE_PLAYER_URL_RE = re.escape(_BASE_PLAYER_URL).replace('/1/', r'/[^/]/')
 139     _VALID_URL = rf'https?:{_BASE_PLAYER_URL_RE}\?([^#]+&)?playlistId=(?P<id>[^#&]+)'
 140
 141     _TESTS = [{
 142         'url': 'https://player.glomex.com/integration/1/iframe-player.html?integrationId=4059a013k56vb2yd&playlistId=v-cfa6lye0dkdd-sf',
 143         'md5': '68f259b98cc01918ac34180142fce287',
 144         'info_dict': {
 145             'id': 'v-cfa6lye0dkdd-sf',
 146             'ext': 'mp4',
 147             'timestamp': 1635337199,
 148             'duration': 133080,
 149             'upload_date': '20211027',
 150             'description': 'md5:e741185fc309310ff5d0c789b437be66',
 151             'title': 'md5:35647293513a6c92363817a0fb0a7961',
 152         },
 153     }, {
 154         'url': 'https://player.glomex.com/integration/1/iframe-player.html?origin=fullpage&integrationId=19syy24xjn1oqlpc&playlistId=rl-vcb49w1fb592p&playlistIndex=0',
 155         'info_dict': {
 156             'id': 'rl-vcb49w1fb592p',
 157         },
 158         'playlist_count': 100,
 159     }, {
 160         'url': 'https://player.glomex.com/integration/1/iframe-player.html?playlistId=cl-bgqaata6aw8x&integrationId=19syy24xjn1oqlpc',
 161         'info_dict': {
 162             'id': 'cl-bgqaata6aw8x',
 163         },
 164         'playlist_mincount': 2,
 165     }]
 166
 167     @classmethod
 168     def build_player_url(cls, video_id, integration, origin_url=None):
 169         query_string = urllib.parse.urlencode({
 170             'playlistId': video_id,
 171             'integrationId': integration,
 172         })
 173         return cls._smuggle_origin_url(f'https:{cls._BASE_PLAYER_URL}?{query_string}', origin_url)
 174
 175     @classmethod
 176     def _extract_embed_urls(cls, url, webpage):
 177         # https://docs.glomex.com/publisher/video-player-integration/javascript-api/
 178         quot_re = r'["\']'
 179
 180         regex = fr'''(?x)
 181             <iframe[^>]+?src=(?P<q>{quot_re})(?P<url>
 182                 (?:https?:)?{cls._BASE_PLAYER_URL_RE}\?(?:(?!(?P=q)).)+
 183             )(?P=q)'''
 184         for mobj in re.finditer(regex, webpage):
 185             embed_url = unescapeHTML(mobj.group('url'))
 186             if cls.suitable(embed_url):
 187                 yield cls._smuggle_origin_url(embed_url, url)
 188
 189         regex = fr'''(?x)
 190             <glomex-player [^>]+?>|
 191             <div[^>]* data-glomex-player=(?P<q>{quot_re})true(?P=q)[^>]*>'''
 192         for mobj in re.finditer(regex, webpage):
 193             attrs = extract_attributes(mobj.group(0))
 194             if attrs.get('data-integration-id') and attrs.get('data-playlist-id'):
 195                 yield cls.build_player_url(attrs['data-playlist-id'], attrs['data-integration-id'], url)
 196
 197         # naive parsing of inline scripts for hard-coded integration parameters
 198         regex = fr'''(?x)
 199             (?P<is_js>dataset\.)?%s\s*(?(is_js)=|:)\s*
 200             (?P<q>{quot_re})(?P<id>(?:(?!(?P=q)).)+)(?P=q)\s'''
 201         for mobj in re.finditer(r'(?x)<script[^<]*>.+?</script>', webpage):
 202             script = mobj.group(0)
 203             integration_id = re.search(regex % 'integrationId', script)
 204             if not integration_id:
 205                 continue
 206             playlist_id = re.search(regex % 'playlistId', script)
 207             if playlist_id:
 208                 yield cls.build_player_url(playlist_id, integration_id, url)
 209
 210     def _real_extract(self, url):
 211         url, origin_url = self._unsmuggle_origin_url(url)
 212         playlist_id = self._match_id(url)
 213         integration = parse_qs(url).get('integrationId', [None])[0]
 214         if not integration:
 215             raise ExtractorError('No integrationId in URL', expected=True)
 216         return self._download_and_extract_api_data(playlist_id, integration, origin_url)