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     ExtractorError,
  11     int_or_none,
  12     parse_qs,
  13     smuggle_url,
  14     unescapeHTML,
  15     unsmuggle_url,
  16 )
  17
  18
  19 class GlomexBaseIE(InfoExtractor):
  20     _DEFAULT_ORIGIN_URL = 'https://player.glomex.com/'
  21     _API_URL = 'https://integration-cloudfront-eu-west-1.mes.glomex.cloud/'
  22
  23     @staticmethod
  24     def _smuggle_origin_url(url, origin_url):
  25         if origin_url is None:
  26             return url
  27         return smuggle_url(url, {'origin': origin_url})
  28
  29     @classmethod
  30     def _unsmuggle_origin_url(cls, url, fallback_origin_url=None):
  31         defaults = {'origin': fallback_origin_url or cls._DEFAULT_ORIGIN_URL}
  32         unsmuggled_url, data = unsmuggle_url(url, default=defaults)
  33         return unsmuggled_url, data['origin']
  34
  35     def _get_videoid_type(self, video_id):
  36         _VIDEOID_TYPES = {
  37             'v': 'video',
  38             'pl': 'playlist',
  39             'rl': 'related videos playlist',
  40             'cl': 'curated playlist',
  41         }
  42         prefix = video_id.split('-')[0]
  43         return _VIDEOID_TYPES.get(prefix, 'unknown type')
  44
  45     def _download_api_data(self, video_id, integration, current_url=None):
  46         query = {
  47             'integration_id': integration,
  48             'playlist_id': video_id,
  49             'current_url': current_url or self._DEFAULT_ORIGIN_URL,
  50         }
  51         video_id_type = self._get_videoid_type(video_id)
  52         return self._download_json(
  53             self._API_URL,
  54             video_id, 'Downloading %s JSON' % video_id_type,
  55             'Unable to download %s JSON' % video_id_type,
  56             query=query)
  57
  58     def _download_and_extract_api_data(self, video_id, integration, current_url):
  59         api_data = self._download_api_data(video_id, integration, current_url)
  60         videos = api_data['videos']
  61         if not videos:
  62             raise ExtractorError('no videos found for %s' % video_id)
  63         videos = [self._extract_api_data(video, video_id) for video in videos]
  64         return videos[0] if len(videos) == 1 else self.playlist_result(videos, video_id)
  65
  66     def _extract_api_data(self, video, video_id):
  67         if video.get('error_code') == 'contentGeoblocked':
  68             self.raise_geo_restricted(countries=video['geo_locations'])
  69
  70         formats, subs = [], {}
  71         for format_id, format_url in video['source'].items():
  72             ext = determine_ext(format_url)
  73             if ext == 'm3u8':
  74                 formats_, subs_ = self._extract_m3u8_formats_and_subtitles(
  75                     format_url, video_id, 'mp4', m3u8_id=format_id,
  76                     fatal=False)
  77                 formats.extend(formats_)
  78                 self._merge_subtitles(subs_, target=subs)
  79             else:
  80                 formats.append({
  81                     'url': format_url,
  82                     'format_id': format_id,
  83                 })
  84         if video.get('language'):
  85             for fmt in formats:
  86                 fmt['language'] = video['language']
  87         self._sort_formats(formats)
  88
  89         images = (video.get('images') or []) + [video.get('image') or {}]
  90         thumbnails = [{
  91             'id': image.get('id'),
  92             'url': f'{image["url"]}/profile:player-960x540',
  93             'width': 960,
  94             'height': 540,
  95         } for image in images if image.get('url')]
  96         self._remove_duplicate_formats(thumbnails)
  97
  98         return {
  99             'id': video.get('clip_id') or video_id,
 100             'title': video.get('title'),
 101             'description': video.get('description'),
 102             'thumbnails': thumbnails,
 103             'duration': int_or_none(video.get('clip_duration')),
 104             'timestamp': video.get('created_at'),
 105             'formats': formats,
 106             'subtitles': subs,
 107         }
 108
 109
 110 class GlomexIE(GlomexBaseIE):
 111     IE_NAME = 'glomex'
 112     IE_DESC = 'Glomex videos'
 113     _VALID_URL = r'https?://video\.glomex\.com/[^/]+/(?P<id>v-[^-]+)'
 114     _INTEGRATION_ID = '19syy24xjn1oqlpc'
 115
 116     _TESTS = [{
 117         'url': 'https://video.glomex.com/sport/v-cb24uwg77hgh-nach-2-0-sieg-guardiola-mit-mancity-vor-naechstem-titel',
 118         'md5': 'cec33a943c4240c9cb33abea8c26242e',
 119         'info_dict': {
 120             'id': 'v-cb24uwg77hgh',
 121             'ext': 'mp4',
 122             'title': 'md5:38a90cedcfadd72982c81acf13556e0c',
 123             'description': 'md5:1ea6b6caff1443fcbbba159e432eedb8',
 124             'duration': 29600,
 125             'timestamp': 1619895017,
 126             'upload_date': '20210501',
 127         },
 128     }]
 129
 130     def _real_extract(self, url):
 131         video_id = self._match_id(url)
 132         return self.url_result(
 133             GlomexEmbedIE.build_player_url(video_id, self._INTEGRATION_ID, url),
 134             GlomexEmbedIE.ie_key(), video_id)
 135
 136
 137 class GlomexEmbedIE(GlomexBaseIE):
 138     IE_NAME = 'glomex:embed'
 139     IE_DESC = 'Glomex embedded videos'
 140     _BASE_PLAYER_URL = '//player.glomex.com/integration/1/iframe-player.html'
 141     _BASE_PLAYER_URL_RE = re.escape(_BASE_PLAYER_URL).replace('/1/', r'/[^/]/')
 142     _VALID_URL = rf'https?:{_BASE_PLAYER_URL_RE}\?([^#]+&)?playlistId=(?P<id>[^#&]+)'
 143
 144     _TESTS = [{
 145         'url': 'https://player.glomex.com/integration/1/iframe-player.html?integrationId=4059a013k56vb2yd&playlistId=v-cfa6lye0dkdd-sf',
 146         'md5': '68f259b98cc01918ac34180142fce287',
 147         'info_dict': {
 148             'id': 'v-cfa6lye0dkdd-sf',
 149             'ext': 'mp4',
 150             'timestamp': 1635337199,
 151             'duration': 133080,
 152             'upload_date': '20211027',
 153             'description': 'md5:e741185fc309310ff5d0c789b437be66',
 154             'title': 'md5:35647293513a6c92363817a0fb0a7961',
 155         },
 156     }, {
 157         'url': 'https://player.glomex.com/integration/1/iframe-player.html?origin=fullpage&integrationId=19syy24xjn1oqlpc&playlistId=rl-vcb49w1fb592p&playlistIndex=0',
 158         'info_dict': {
 159             'id': 'rl-vcb49w1fb592p',
 160         },
 161         'playlist_count': 100,
 162     }, {
 163         'url': 'https://player.glomex.com/integration/1/iframe-player.html?playlistId=cl-bgqaata6aw8x&integrationId=19syy24xjn1oqlpc',
 164         'info_dict': {
 165             'id': 'cl-bgqaata6aw8x',
 166         },
 167         'playlist_mincount': 2,
 168     }]
 169
 170     @classmethod
 171     def build_player_url(cls, video_id, integration, origin_url=None):
 172         query_string = urllib.parse.urlencode({
 173             'playlistId': video_id,
 174             'integrationId': integration,
 175         })
 176         return cls._smuggle_origin_url(f'https:{cls._BASE_PLAYER_URL}?{query_string}', origin_url)
 177
 178     @classmethod
 179     def _extract_urls(cls, webpage, origin_url):
 180         # in comparison with _VALID_URL:
 181         # * make the scheme optional
 182         # * simplify the query string part; after extracting iframe src, the URL will be matched again
 183         VALID_SRC = rf'(?:https?:)?{cls._BASE_PLAYER_URL_RE}\?(?:(?!(?P=_q1)).)+'
 184
 185         # https://docs.glomex.com/publisher/video-player-integration/javascript-api/
 186         EMBED_RE = r'''(?x)(?:
 187             <iframe[^>]+?src=(?P<_q1>%(quot_re)s)(?P<url>%(url_re)s)(?P=_q1)|
 188             <(?P<html_tag>glomex-player|div)(?:
 189                 data-integration-id=(?P<_q2>%(quot_re)s)(?P<integration_html>(?:(?!(?P=_q2)).)+)(?P=_q2)|
 190                 data-playlist-id=(?P<_q3>%(quot_re)s)(?P<id_html>(?:(?!(?P=_q3)).)+)(?P=_q3)|
 191                 data-glomex-player=(?P<_q4>%(quot_re)s)(?P<glomex_player>true)(?P=_q4)|
 192                 [^>]*?
 193             )+>|
 194             # naive parsing of inline scripts for hard-coded integration parameters
 195             <(?P<script_tag>script)[^<]*?>(?:
 196                 (?P<_stjs1>dataset\.)?integrationId\s*(?(_stjs1)=|:)\s*
 197                     (?P<_q5>%(quot_re)s)(?P<integration_js>(?:(?!(?P=_q5)).)+)(?P=_q5)\s*(?(_stjs1);|,)?|
 198                 (?P<_stjs2>dataset\.)?playlistId\s*(?(_stjs2)=|:)\s*
 199                     (?P<_q6>%(quot_re)s)(?P<id_js>(?:(?!(?P=_q6)).)+)(?P=_q6)\s*(?(_stjs2);|,)?|
 200                 (?:\s|.)*?
 201             )+</script>
 202         )''' % {'quot_re': r'["\']', 'url_re': VALID_SRC}
 203
 204         for mobj in re.finditer(EMBED_RE, webpage):
 205             mdict = mobj.groupdict()
 206             if mdict.get('url'):
 207                 url = unescapeHTML(mdict['url'])
 208                 if not cls.suitable(url):
 209                     continue
 210                 yield cls._smuggle_origin_url(url, origin_url)
 211             elif mdict.get('html_tag'):
 212                 if mdict['html_tag'] == 'div' and not mdict.get('glomex_player'):
 213                     continue
 214                 if not mdict.get('video_id_html') or not mdict.get('integration_html'):
 215                     continue
 216                 yield cls.build_player_url(mdict['video_id_html'], mdict['integration_html'], origin_url)
 217             elif mdict.get('script_tag'):
 218                 if not mdict.get('video_id_js') or not mdict.get('integration_js'):
 219                     continue
 220                 yield cls.build_player_url(mdict['video_id_js'], mdict['integration_js'], origin_url)
 221
 222     def _real_extract(self, url):
 223         url, origin_url = self._unsmuggle_origin_url(url)
 224         playlist_id = self._match_id(url)
 225         integration = parse_qs(url).get('integrationId', [None])[0]
 226         if not integration:
 227             raise ExtractorError('No integrationId in URL', expected=True)
 228         return self._download_and_extract_api_data(playlist_id, integration, origin_url)