yt_dlp/extractor/wdr.py

   1 import re
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     determine_ext,
   8     dict_get,
   9     js_to_json,
  10     strip_jsonp,
  11     try_get,
  12     unified_strdate,
  13     update_url_query,
  14     url_or_none,
  15     urlhandle_detect_ext,
  16 )
  17
  18
  19 class WDRIE(InfoExtractor):
  20     __API_URL_TPL = '//deviceids-medp.wdr.de/ondemand/%s/%s'
  21     _VALID_URL = r'''(?x)https?://
  22         (?:deviceids-medp\.wdr\.de/ondemand/\d+/|
  23            kinder\.wdr\.de/(?!mediathek/)[^#?]+-)
  24         (?P<id>\d+)\.(?:js|assetjsonp)
  25     '''
  26     _GEO_COUNTRIES = ['DE']
  27     _TESTS = [{
  28         'url': 'http://deviceids-medp.wdr.de/ondemand/155/1557833.js',
  29         'info_dict': {
  30             'id': 'mdb-1557833',
  31             'ext': 'mp4',
  32             'title': 'Biathlon-Staffel verpasst Podest bei Olympia-Generalprobe',
  33             'upload_date': '20180112',
  34         },
  35     }]
  36
  37     def _asset_url(self, wdr_id):
  38         id_len = max(len(wdr_id), 5)
  39         return ''.join(('https:', self.__API_URL_TPL % (wdr_id[:id_len - 4], wdr_id), '.js'))
  40
  41     def _real_extract(self, url):
  42         video_id = self._match_id(url)
  43
  44         if url.startswith('wdr:'):
  45             video_id = url[4:]
  46             url = self._asset_url(video_id)
  47
  48         metadata = self._download_json(
  49             url, video_id, transform_source=strip_jsonp)
  50
  51         is_live = metadata.get('mediaType') == 'live'
  52
  53         tracker_data = metadata['trackerData']
  54         title = tracker_data['trackerClipTitle']
  55         media_resource = metadata['mediaResource']
  56
  57         formats = []
  58         subtitles = {}
  59
  60         # check if the metadata contains a direct URL to a file
  61         for kind, media in media_resource.items():
  62             if kind == 'captionsHash':
  63                 for ext, url in media.items():
  64                     subtitles.setdefault('de', []).append({
  65                         'url': url,
  66                         'ext': ext,
  67                     })
  68                 continue
  69
  70             if kind not in ('dflt', 'alt'):
  71                 continue
  72             if not isinstance(media, dict):
  73                 continue
  74
  75             for tag_name, medium_url in media.items():
  76                 if tag_name not in ('videoURL', 'audioURL'):
  77                     continue
  78
  79                 ext = determine_ext(medium_url)
  80                 if ext == 'm3u8':
  81                     formats.extend(self._extract_m3u8_formats(
  82                         medium_url, video_id, 'mp4', 'm3u8_native',
  83                         m3u8_id='hls'))
  84                 elif ext == 'f4m':
  85                     manifest_url = update_url_query(
  86                         medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
  87                     formats.extend(self._extract_f4m_formats(
  88                         manifest_url, video_id, f4m_id='hds', fatal=False))
  89                 elif ext == 'smil':
  90                     formats.extend(self._extract_smil_formats(
  91                         medium_url, 'stream', fatal=False))
  92                 else:
  93                     a_format = {
  94                         'url': medium_url,
  95                     }
  96                     if ext == 'unknown_video':
  97                         urlh = self._request_webpage(
  98                             medium_url, video_id, note='Determining extension')
  99                         ext = urlhandle_detect_ext(urlh)
 100                         a_format['ext'] = ext
 101                     formats.append(a_format)
 102
 103         caption_url = media_resource.get('captionURL')
 104         if caption_url:
 105             subtitles['de'] = [{
 106                 'url': caption_url,
 107                 'ext': 'ttml',
 108             }]
 109         captions_hash = media_resource.get('captionsHash')
 110         if isinstance(captions_hash, dict):
 111             for ext, format_url in captions_hash.items():
 112                 format_url = url_or_none(format_url)
 113                 if not format_url:
 114                     continue
 115                 subtitles.setdefault('de', []).append({
 116                     'url': format_url,
 117                     'ext': determine_ext(format_url, None) or ext,
 118                 })
 119
 120         return {
 121             'id': tracker_data.get('trackerClipId', video_id),
 122             'title': title,
 123             'alt_title': tracker_data.get('trackerClipSubcategory'),
 124             'formats': formats,
 125             'subtitles': subtitles,
 126             'upload_date': unified_strdate(tracker_data.get('trackerClipAirTime')),
 127             'is_live': is_live,
 128         }
 129
 130
 131 class WDRPageIE(WDRIE):  # XXX: Do not subclass from concrete IE
 132     _MAUS_REGEX = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/)*?(?P<maus_id>[^/?#.]+)(?:/?|/index\.php5|\.php5)$'
 133     _PAGE_REGEX = r'/(?:mediathek/)?(?:[^/]+/)*(?P<display_id>[^/]+)\.html'
 134     _VALID_URL = r'https?://(?:www\d?\.)?(?:(?:kinder\.)?wdr\d?|sportschau)\.de' + _PAGE_REGEX + '|' + _MAUS_REGEX
 135
 136     _TESTS = [
 137         {
 138             'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
 139             # HDS download, MD5 is unstable
 140             'info_dict': {
 141                 'id': 'mdb-1058683',
 142                 'ext': 'flv',
 143                 'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
 144                 'title': 'Geheimnis Aachener Dom',
 145                 'alt_title': 'Doku am Freitag',
 146                 'upload_date': '20160304',
 147                 'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
 148                 'is_live': False,
 149                 'subtitles': {'de': [{
 150                     'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
 151                     'ext': 'ttml',
 152                 }]},
 153             },
 154             'skip': 'HTTP Error 404: Not Found',
 155         },
 156         {
 157             'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
 158             'md5': 'f4c1f96d01cf285240f53ea4309663d8',
 159             'info_dict': {
 160                 'id': 'mdb-1072000',
 161                 'ext': 'mp3',
 162                 'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
 163                 'title': 'Schriftstellerin Juli Zeh',
 164                 'alt_title': 'WDR 3 Gespräch am Samstag',
 165                 'upload_date': '20160312',
 166                 'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
 167                 'is_live': False,
 168                 'subtitles': {},
 169             },
 170             'skip': 'HTTP Error 404: Not Found',
 171         },
 172         {
 173             # FIXME: Asset JSON is directly embedded in webpage
 174             'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
 175             'info_dict': {
 176                 'id': 'mdb-2296252',
 177                 'ext': 'mp4',
 178                 'title': r're:^WDR Fernsehen im Livestream (?:\(nur in Deutschland erreichbar\) )?[0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 179                 'alt_title': 'WDR Fernsehen Live',
 180                 'upload_date': '20201112',
 181                 'is_live': True,
 182             },
 183             'params': {
 184                 'skip_download': True,  # m3u8 download
 185             },
 186         },
 187         {
 188             'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
 189             'playlist_mincount': 6,
 190             'info_dict': {
 191                 'id': 'aktuelle-stunde-120',
 192             },
 193         },
 194         {
 195             'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
 196             'info_dict': {
 197                 'id': 'mdb-2627637',
 198                 'ext': 'mp4',
 199                 'upload_date': 're:^[0-9]{8}$',
 200                 'title': 're:^Die Sendung (?:mit der Maus )?vom [0-9.]{10}$',
 201             },
 202             'skip': 'The id changes from week to week because of the new episode',
 203         },
 204         {
 205             'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
 206             'md5': '803138901f6368ee497b4d195bb164f2',
 207             'info_dict': {
 208                 'id': 'mdb-186083',
 209                 'ext': 'mp4',
 210                 'upload_date': '20130919',
 211                 'title': 'Sachgeschichte - Achterbahn ',
 212             },
 213             'skip': 'HTTP Error 404: Not Found',
 214         },
 215         {
 216             'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
 217             # Live stream, MD5 unstable
 218             'info_dict': {
 219                 'id': 'mdb-869971',
 220                 'ext': 'mp4',
 221                 'title': r're:^COSMO Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 222                 'alt_title': 'COSMO Livestream',
 223                 'live_status': 'is_live',
 224                 'upload_date': '20160101',
 225             },
 226             'params': {
 227                 'skip_download': True,  # m3u8 download
 228             },
 229         },
 230         {
 231             'url': 'http://www.sportschau.de/handballem2018/handball-nationalmannschaft-em-stolperstein-vorrunde-100.html',
 232             'info_dict': {
 233                 'id': 'mdb-1556012',
 234                 'ext': 'mp4',
 235                 'title': 'DHB-Vizepräsident Bob Hanning - "Die Weltspitze ist extrem breit"',
 236                 'upload_date': '20180111',
 237             },
 238             'params': {
 239                 'skip_download': True,
 240             },
 241             'skip': 'HTTP Error 404: Not Found',
 242         },
 243         {
 244             'url': 'http://www.sportschau.de/handballem2018/audio-vorschau---die-handball-em-startet-mit-grossem-favoritenfeld-100.html',
 245             'only_matching': True,
 246         },
 247         {
 248             'url': 'https://kinder.wdr.de/tv/die-sendung-mit-dem-elefanten/av/video-folge---astronaut-100.html',
 249             'only_matching': True,
 250         },
 251         {
 252             'url': 'https://www1.wdr.de/mediathek/video/sendungen/rockpalast/video-baroness---freak-valley-festival--100.html',
 253             'info_dict': {
 254                 'id': 'mdb-2741028',
 255                 'ext': 'mp4',
 256                 'title': 'Baroness - Freak Valley Festival 2022',
 257                 'alt_title': 'Rockpalast',
 258                 'upload_date': '20220725',
 259             },
 260         },
 261     ]
 262
 263     def _real_extract(self, url):
 264         mobj = self._match_valid_url(url)
 265         display_id = dict_get(mobj.groupdict(), ('display_id', 'maus_id'), 'wdrmaus')
 266         webpage = self._download_webpage(url, display_id)
 267
 268         entries = []
 269
 270         # Article with several videos
 271
 272         # for wdr.de the data-extension-ard is in a tag with the class "mediaLink"
 273         # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
 274         # for wdrmaus, in a tag with the class "videoButton" (previously a link
 275         # to the page in a multiline "videoLink"-tag)
 276         for mobj in re.finditer(
 277             r'''(?sx)class=
 278                     (?:
 279                         (["\'])(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b.*?\1[^>]+|
 280                         (["\'])videoLink\b.*?\2[\s]*>\n[^\n]*
 281                     )data-extension(?:-ard)?=(["\'])(?P<data>(?:(?!\3).)+)\3
 282                     ''', webpage):
 283             media_link_obj = self._parse_json(
 284                 mobj.group('data'), display_id, transform_source=js_to_json,
 285                 fatal=False)
 286             if not media_link_obj:
 287                 continue
 288             jsonp_url = try_get(
 289                 media_link_obj, lambda x: x['mediaObj']['url'], str)
 290             if jsonp_url:
 291                 # metadata, or player JS with ['ref'] giving WDR id, or just media, perhaps
 292                 clip_id = media_link_obj['mediaObj'].get('ref')
 293                 if jsonp_url.endswith('.assetjsonp'):
 294                     asset = self._download_json(
 295                         jsonp_url, display_id, fatal=False, transform_source=strip_jsonp)
 296                     clip_id = try_get(asset, lambda x: x['trackerData']['trackerClipId'], str)
 297                 if clip_id:
 298                     jsonp_url = self._asset_url(clip_id[4:])
 299                 entries.append(self.url_result(jsonp_url, ie=WDRIE.ie_key()))
 300
 301         # Playlist (e.g. https://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html)
 302         if not entries:
 303             entries = [
 304                 self.url_result(
 305                     urllib.parse.urljoin(url, mobj.group('href')),
 306                     ie=WDRPageIE.ie_key())
 307                 for mobj in re.finditer(
 308                     r'<a[^>]+\bhref=(["\'])(?P<href>(?:(?!\1).)+)\1[^>]+\bdata-extension(?:-ard)?=',
 309                     webpage) if re.match(self._PAGE_REGEX, mobj.group('href'))
 310             ]
 311
 312         return self.playlist_result(entries, playlist_id=display_id)
 313
 314
 315 class WDRElefantIE(InfoExtractor):
 316     _VALID_URL = r'https?://(?:www\.)wdrmaus\.de/elefantenseite/#(?P<id>.+)'
 317     _TEST = {
 318         'url': 'http://www.wdrmaus.de/elefantenseite/#elefantenkino_wippe',
 319         # adaptive stream: unstable file MD5
 320         'info_dict': {
 321             'title': 'Wippe',
 322             'id': 'mdb-1198320',
 323             'ext': 'mp4',
 324             'upload_date': '20071003',
 325         },
 326     }
 327
 328     def _real_extract(self, url):
 329         display_id = self._match_id(url)
 330
 331         # Table of Contents seems to always be at this address, so fetch it directly.
 332         # The website fetches configurationJS.php5, which links to tableOfContentsJS.php5.
 333         table_of_contents = self._download_json(
 334             'https://www.wdrmaus.de/elefantenseite/data/tableOfContentsJS.php5',
 335             display_id)
 336         if display_id not in table_of_contents:
 337             raise ExtractorError(
 338                 'No entry in site\'s table of contents for this URL. '
 339                 'Is the fragment part of the URL (after the #) correct?',
 340                 expected=True)
 341         xml_metadata_path = table_of_contents[display_id]['xmlPath']
 342         xml_metadata = self._download_xml(
 343             'https://www.wdrmaus.de/elefantenseite/' + xml_metadata_path,
 344             display_id)
 345         zmdb_url_element = xml_metadata.find('./movie/zmdb_url')
 346         if zmdb_url_element is None:
 347             raise ExtractorError(
 348                 f'{display_id} is not a video', expected=True)
 349         return self.url_result(zmdb_url_element.text, ie=WDRIE.ie_key())
 350
 351
 352 class WDRMobileIE(InfoExtractor):
 353     _VALID_URL = r'''(?x)
 354         https?://mobile-ondemand\.wdr\.de/
 355         .*?/fsk(?P<age_limit>[0-9]+)
 356         /[0-9]+/[0-9]+/
 357         (?P<id>[0-9]+)_(?P<title>[0-9]+)'''
 358     IE_NAME = 'wdr:mobile'
 359     _WORKING = False  # no such domain
 360     _TEST = {
 361         'url': 'http://mobile-ondemand.wdr.de/CMS2010/mdb/ondemand/weltweit/fsk0/42/421735/421735_4283021.mp4',
 362         'info_dict': {
 363             'title': '4283021',
 364             'id': '421735',
 365             'ext': 'mp4',
 366             'age_limit': 0,
 367         },
 368         'skip': 'Problems with loading data.',
 369     }
 370
 371     def _real_extract(self, url):
 372         mobj = self._match_valid_url(url)
 373         return {
 374             'id': mobj.group('id'),
 375             'title': mobj.group('title'),
 376             'age_limit': int(mobj.group('age_limit')),
 377             'url': url,
 378             'http_headers': {
 379                 'User-Agent': 'mobile',
 380             },
 381         }