yt_dlp/extractor/wdr.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..compat import (
   5     compat_str,
   6     compat_urlparse,
   7 )
   8 from ..utils import (
   9     ExtractorError,
  10     determine_ext,
  11     dict_get,
  12     js_to_json,
  13     strip_jsonp,
  14     try_get,
  15     unified_strdate,
  16     update_url_query,
  17     url_or_none,
  18     urlhandle_detect_ext,
  19 )
  20
  21
  22 class WDRIE(InfoExtractor):
  23     __API_URL_TPL = '//deviceids-medp.wdr.de/ondemand/%s/%s'
  24     _VALID_URL = r'''(?x)https?://
  25         (?:deviceids-medp\.wdr\.de/ondemand/\d+/|
  26            kinder\.wdr\.de/(?!mediathek/)[^#?]+-)
  27         (?P<id>\d+)\.(?:js|assetjsonp)
  28     '''
  29     _GEO_COUNTRIES = ['DE']
  30     _TESTS = [{
  31         'url': 'http://deviceids-medp.wdr.de/ondemand/155/1557833.js',
  32         'info_dict': {
  33             'id': 'mdb-1557833',
  34             'ext': 'mp4',
  35             'title': 'Biathlon-Staffel verpasst Podest bei Olympia-Generalprobe',
  36             'upload_date': '20180112',
  37         },
  38     }]
  39
  40     def _asset_url(self, wdr_id):
  41         id_len = max(len(wdr_id), 5)
  42         return ''.join(('https:', self.__API_URL_TPL % (wdr_id[:id_len - 4], wdr_id, ), '.js'))
  43
  44     def _real_extract(self, url):
  45         video_id = self._match_id(url)
  46
  47         if url.startswith('wdr:'):
  48             video_id = url[4:]
  49             url = self._asset_url(video_id)
  50
  51         metadata = self._download_json(
  52             url, video_id, transform_source=strip_jsonp)
  53
  54         is_live = metadata.get('mediaType') == 'live'
  55
  56         tracker_data = metadata['trackerData']
  57         title = tracker_data['trackerClipTitle']
  58         media_resource = metadata['mediaResource']
  59
  60         formats = []
  61         subtitles = {}
  62
  63         # check if the metadata contains a direct URL to a file
  64         for kind, media in media_resource.items():
  65             if kind == 'captionsHash':
  66                 for ext, url in media.items():
  67                     subtitles.setdefault('de', []).append({
  68                         'url': url,
  69                         'ext': ext,
  70                     })
  71                 continue
  72
  73             if kind not in ('dflt', 'alt'):
  74                 continue
  75             if not isinstance(media, dict):
  76                 continue
  77
  78             for tag_name, medium_url in media.items():
  79                 if tag_name not in ('videoURL', 'audioURL'):
  80                     continue
  81
  82                 ext = determine_ext(medium_url)
  83                 if ext == 'm3u8':
  84                     formats.extend(self._extract_m3u8_formats(
  85                         medium_url, video_id, 'mp4', 'm3u8_native',
  86                         m3u8_id='hls'))
  87                 elif ext == 'f4m':
  88                     manifest_url = update_url_query(
  89                         medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
  90                     formats.extend(self._extract_f4m_formats(
  91                         manifest_url, video_id, f4m_id='hds', fatal=False))
  92                 elif ext == 'smil':
  93                     formats.extend(self._extract_smil_formats(
  94                         medium_url, 'stream', fatal=False))
  95                 else:
  96                     a_format = {
  97                         'url': medium_url
  98                     }
  99                     if ext == 'unknown_video':
 100                         urlh = self._request_webpage(
 101                             medium_url, video_id, note='Determining extension')
 102                         ext = urlhandle_detect_ext(urlh)
 103                         a_format['ext'] = ext
 104                     formats.append(a_format)
 105
 106         caption_url = media_resource.get('captionURL')
 107         if caption_url:
 108             subtitles['de'] = [{
 109                 'url': caption_url,
 110                 'ext': 'ttml',
 111             }]
 112         captions_hash = media_resource.get('captionsHash')
 113         if isinstance(captions_hash, dict):
 114             for ext, format_url in captions_hash.items():
 115                 format_url = url_or_none(format_url)
 116                 if not format_url:
 117                     continue
 118                 subtitles.setdefault('de', []).append({
 119                     'url': format_url,
 120                     'ext': determine_ext(format_url, None) or ext,
 121                 })
 122
 123         return {
 124             'id': tracker_data.get('trackerClipId', video_id),
 125             'title': title,
 126             'alt_title': tracker_data.get('trackerClipSubcategory'),
 127             'formats': formats,
 128             'subtitles': subtitles,
 129             'upload_date': unified_strdate(tracker_data.get('trackerClipAirTime')),
 130             'is_live': is_live,
 131         }
 132
 133
 134 class WDRPageIE(WDRIE):  # XXX: Do not subclass from concrete IE
 135     _MAUS_REGEX = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/)*?(?P<maus_id>[^/?#.]+)(?:/?|/index\.php5|\.php5)$'
 136     _PAGE_REGEX = r'/(?:mediathek/)?(?:[^/]+/)*(?P<display_id>[^/]+)\.html'
 137     _VALID_URL = r'https?://(?:www\d?\.)?(?:(?:kinder\.)?wdr\d?|sportschau)\.de' + _PAGE_REGEX + '|' + _MAUS_REGEX
 138
 139     _TESTS = [
 140         {
 141             'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
 142             # HDS download, MD5 is unstable
 143             'info_dict': {
 144                 'id': 'mdb-1058683',
 145                 'ext': 'flv',
 146                 'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
 147                 'title': 'Geheimnis Aachener Dom',
 148                 'alt_title': 'Doku am Freitag',
 149                 'upload_date': '20160304',
 150                 'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
 151                 'is_live': False,
 152                 'subtitles': {'de': [{
 153                     'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
 154                     'ext': 'ttml',
 155                 }]},
 156             },
 157             'skip': 'HTTP Error 404: Not Found',
 158         },
 159         {
 160             'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
 161             'md5': 'f4c1f96d01cf285240f53ea4309663d8',
 162             'info_dict': {
 163                 'id': 'mdb-1072000',
 164                 'ext': 'mp3',
 165                 'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
 166                 'title': 'Schriftstellerin Juli Zeh',
 167                 'alt_title': 'WDR 3 Gespräch am Samstag',
 168                 'upload_date': '20160312',
 169                 'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
 170                 'is_live': False,
 171                 'subtitles': {}
 172             },
 173             'skip': 'HTTP Error 404: Not Found',
 174         },
 175         {
 176             # FIXME: Asset JSON is directly embedded in webpage
 177             'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
 178             'info_dict': {
 179                 'id': 'mdb-2296252',
 180                 'ext': 'mp4',
 181                 '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}$',
 182                 'alt_title': 'WDR Fernsehen Live',
 183                 'upload_date': '20201112',
 184                 'is_live': True,
 185             },
 186             'params': {
 187                 'skip_download': True,  # m3u8 download
 188             },
 189         },
 190         {
 191             'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
 192             'playlist_mincount': 6,
 193             'info_dict': {
 194                 'id': 'aktuelle-stunde-120',
 195             },
 196         },
 197         {
 198             'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
 199             'info_dict': {
 200                 'id': 'mdb-2627637',
 201                 'ext': 'mp4',
 202                 'upload_date': 're:^[0-9]{8}$',
 203                 'title': 're:^Die Sendung (?:mit der Maus )?vom [0-9.]{10}$',
 204             },
 205             'skip': 'The id changes from week to week because of the new episode'
 206         },
 207         {
 208             'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
 209             'md5': '803138901f6368ee497b4d195bb164f2',
 210             'info_dict': {
 211                 'id': 'mdb-186083',
 212                 'ext': 'mp4',
 213                 'upload_date': '20130919',
 214                 'title': 'Sachgeschichte - Achterbahn ',
 215             },
 216             'skip': 'HTTP Error 404: Not Found',
 217         },
 218         {
 219             'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
 220             # Live stream, MD5 unstable
 221             'info_dict': {
 222                 'id': 'mdb-869971',
 223                 'ext': 'mp4',
 224                 'title': r're:^COSMO Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 225                 'alt_title': 'COSMO Livestream',
 226                 'live_status': 'is_live',
 227                 'upload_date': '20160101',
 228             },
 229             'params': {
 230                 'skip_download': True,  # m3u8 download
 231             }
 232         },
 233         {
 234             'url': 'http://www.sportschau.de/handballem2018/handball-nationalmannschaft-em-stolperstein-vorrunde-100.html',
 235             'info_dict': {
 236                 'id': 'mdb-1556012',
 237                 'ext': 'mp4',
 238                 'title': 'DHB-Vizepräsident Bob Hanning - "Die Weltspitze ist extrem breit"',
 239                 'upload_date': '20180111',
 240             },
 241             'params': {
 242                 'skip_download': True,
 243             },
 244             'skip': 'HTTP Error 404: Not Found',
 245         },
 246         {
 247             'url': 'http://www.sportschau.de/handballem2018/audio-vorschau---die-handball-em-startet-mit-grossem-favoritenfeld-100.html',
 248             'only_matching': True,
 249         },
 250         {
 251             'url': 'https://kinder.wdr.de/tv/die-sendung-mit-dem-elefanten/av/video-folge---astronaut-100.html',
 252             'only_matching': True,
 253         },
 254         {
 255             'url': 'https://www1.wdr.de/mediathek/video/sendungen/rockpalast/video-baroness---freak-valley-festival--100.html',
 256             'info_dict': {
 257                 'id': 'mdb-2741028',
 258                 'ext': 'mp4',
 259                 'title': 'Baroness - Freak Valley Festival 2022',
 260                 'alt_title': 'Rockpalast',
 261                 'upload_date': '20220725',
 262             },
 263         }
 264     ]
 265
 266     def _real_extract(self, url):
 267         mobj = self._match_valid_url(url)
 268         display_id = dict_get(mobj.groupdict(), ('display_id', 'maus_id'), 'wdrmaus')
 269         webpage = self._download_webpage(url, display_id)
 270
 271         entries = []
 272
 273         # Article with several videos
 274
 275         # for wdr.de the data-extension-ard is in a tag with the class "mediaLink"
 276         # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
 277         # for wdrmaus, in a tag with the class "videoButton" (previously a link
 278         # to the page in a multiline "videoLink"-tag)
 279         for mobj in re.finditer(
 280             r'''(?sx)class=
 281                     (?:
 282                         (["\'])(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b.*?\1[^>]+|
 283                         (["\'])videoLink\b.*?\2[\s]*>\n[^\n]*
 284                     )data-extension(?:-ard)?=(["\'])(?P<data>(?:(?!\3).)+)\3
 285                     ''', webpage):
 286             media_link_obj = self._parse_json(
 287                 mobj.group('data'), display_id, transform_source=js_to_json,
 288                 fatal=False)
 289             if not media_link_obj:
 290                 continue
 291             jsonp_url = try_get(
 292                 media_link_obj, lambda x: x['mediaObj']['url'], compat_str)
 293             if jsonp_url:
 294                 # metadata, or player JS with ['ref'] giving WDR id, or just media, perhaps
 295                 clip_id = media_link_obj['mediaObj'].get('ref')
 296                 if jsonp_url.endswith('.assetjsonp'):
 297                     asset = self._download_json(
 298                         jsonp_url, display_id, fatal=False, transform_source=strip_jsonp)
 299                     clip_id = try_get(asset, lambda x: x['trackerData']['trackerClipId'], compat_str)
 300                 if clip_id:
 301                     jsonp_url = self._asset_url(clip_id[4:])
 302                 entries.append(self.url_result(jsonp_url, ie=WDRIE.ie_key()))
 303
 304         # Playlist (e.g. https://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html)
 305         if not entries:
 306             entries = [
 307                 self.url_result(
 308                     compat_urlparse.urljoin(url, mobj.group('href')),
 309                     ie=WDRPageIE.ie_key())
 310                 for mobj in re.finditer(
 311                     r'<a[^>]+\bhref=(["\'])(?P<href>(?:(?!\1).)+)\1[^>]+\bdata-extension(?:-ard)?=',
 312                     webpage) if re.match(self._PAGE_REGEX, mobj.group('href'))
 313             ]
 314
 315         return self.playlist_result(entries, playlist_id=display_id)
 316
 317
 318 class WDRElefantIE(InfoExtractor):
 319     _VALID_URL = r'https?://(?:www\.)wdrmaus\.de/elefantenseite/#(?P<id>.+)'
 320     _TEST = {
 321         'url': 'http://www.wdrmaus.de/elefantenseite/#elefantenkino_wippe',
 322         # adaptive stream: unstable file MD5
 323         'info_dict': {
 324             'title': 'Wippe',
 325             'id': 'mdb-1198320',
 326             'ext': 'mp4',
 327             'upload_date': '20071003'
 328         },
 329     }
 330
 331     def _real_extract(self, url):
 332         display_id = self._match_id(url)
 333
 334         # Table of Contents seems to always be at this address, so fetch it directly.
 335         # The website fetches configurationJS.php5, which links to tableOfContentsJS.php5.
 336         table_of_contents = self._download_json(
 337             'https://www.wdrmaus.de/elefantenseite/data/tableOfContentsJS.php5',
 338             display_id)
 339         if display_id not in table_of_contents:
 340             raise ExtractorError(
 341                 'No entry in site\'s table of contents for this URL. '
 342                 'Is the fragment part of the URL (after the #) correct?',
 343                 expected=True)
 344         xml_metadata_path = table_of_contents[display_id]['xmlPath']
 345         xml_metadata = self._download_xml(
 346             'https://www.wdrmaus.de/elefantenseite/' + xml_metadata_path,
 347             display_id)
 348         zmdb_url_element = xml_metadata.find('./movie/zmdb_url')
 349         if zmdb_url_element is None:
 350             raise ExtractorError(
 351                 '%s is not a video' % display_id, expected=True)
 352         return self.url_result(zmdb_url_element.text, ie=WDRIE.ie_key())
 353
 354
 355 class WDRMobileIE(InfoExtractor):
 356     _VALID_URL = r'''(?x)
 357         https?://mobile-ondemand\.wdr\.de/
 358         .*?/fsk(?P<age_limit>[0-9]+)
 359         /[0-9]+/[0-9]+/
 360         (?P<id>[0-9]+)_(?P<title>[0-9]+)'''
 361     IE_NAME = 'wdr:mobile'
 362     _WORKING = False  # no such domain
 363     _TEST = {
 364         'url': 'http://mobile-ondemand.wdr.de/CMS2010/mdb/ondemand/weltweit/fsk0/42/421735/421735_4283021.mp4',
 365         'info_dict': {
 366             'title': '4283021',
 367             'id': '421735',
 368             'ext': 'mp4',
 369             'age_limit': 0,
 370         },
 371         'skip': 'Problems with loading data.'
 372     }
 373
 374     def _real_extract(self, url):
 375         mobj = self._match_valid_url(url)
 376         return {
 377             'id': mobj.group('id'),
 378             'title': mobj.group('title'),
 379             'age_limit': int(mobj.group('age_limit')),
 380             'url': url,
 381             'http_headers': {
 382                 'User-Agent': 'mobile',
 383             },
 384         }