yt_dlp/extractor/glomex.py

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