yt_dlp/extractor/wdr.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_str,
   9     compat_urlparse,
  10 )
  11 from ..utils import (
  12     determine_ext,
  13     ExtractorError,
  14     js_to_json,
  15     strip_jsonp,
  16     try_get,
  17     unified_strdate,
  18     update_url_query,
  19     urlhandle_detect_ext,
  20     url_or_none,
  21 )
  22
  23
  24 class WDRIE(InfoExtractor):
  25     _VALID_URL = r'https?://deviceids-medp\.wdr\.de/ondemand/\d+/(?P<id>\d+)\.js'
  26     _GEO_COUNTRIES = ['DE']
  27     _TEST = {
  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 _real_extract(self, url):
  38         video_id = self._match_id(url)
  39
  40         metadata = self._download_json(
  41             url, video_id, transform_source=strip_jsonp)
  42
  43         is_live = metadata.get('mediaType') == 'live'
  44
  45         tracker_data = metadata['trackerData']
  46         title = tracker_data['trackerClipTitle']
  47         media_resource = metadata['mediaResource']
  48
  49         formats = []
  50         subtitles = {}
  51
  52         # check if the metadata contains a direct URL to a file
  53         for kind, media in media_resource.items():
  54             if kind == 'captionsHash':
  55                 for ext, url in media.items():
  56                     subtitles.setdefault('de', []).append({
  57                         'url': url,
  58                         'ext': ext,
  59                     })
  60                 continue
  61
  62             if kind not in ('dflt', 'alt'):
  63                 continue
  64             if not isinstance(media, dict):
  65                 continue
  66
  67             for tag_name, medium_url in media.items():
  68                 if tag_name not in ('videoURL', 'audioURL'):
  69                     continue
  70
  71                 ext = determine_ext(medium_url)
  72                 if ext == 'm3u8':
  73                     formats.extend(self._extract_m3u8_formats(
  74                         medium_url, video_id, 'mp4', 'm3u8_native',
  75                         m3u8_id='hls'))
  76                 elif ext == 'f4m':
  77                     manifest_url = update_url_query(
  78                         medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
  79                     formats.extend(self._extract_f4m_formats(
  80                         manifest_url, video_id, f4m_id='hds', fatal=False))
  81                 elif ext == 'smil':
  82                     formats.extend(self._extract_smil_formats(
  83                         medium_url, 'stream', fatal=False))
  84                 else:
  85                     a_format = {
  86                         'url': medium_url
  87                     }
  88                     if ext == 'unknown_video':
  89                         urlh = self._request_webpage(
  90                             medium_url, video_id, note='Determining extension')
  91                         ext = urlhandle_detect_ext(urlh)
  92                         a_format['ext'] = ext
  93                     formats.append(a_format)
  94
  95         self._sort_formats(formats)
  96
  97         caption_url = media_resource.get('captionURL')
  98         if caption_url:
  99             subtitles['de'] = [{
 100                 'url': caption_url,
 101                 'ext': 'ttml',
 102             }]
 103         captions_hash = media_resource.get('captionsHash')
 104         if isinstance(captions_hash, dict):
 105             for ext, format_url in captions_hash.items():
 106                 format_url = url_or_none(format_url)
 107                 if not format_url:
 108                     continue
 109                 subtitles.setdefault('de', []).append({
 110                     'url': format_url,
 111                     'ext': determine_ext(format_url, None) or ext,
 112                 })
 113
 114         return {
 115             'id': tracker_data.get('trackerClipId', video_id),
 116             'title': self._live_title(title) if is_live else title,
 117             'alt_title': tracker_data.get('trackerClipSubcategory'),
 118             'formats': formats,
 119             'subtitles': subtitles,
 120             'upload_date': unified_strdate(tracker_data.get('trackerClipAirTime')),
 121             'is_live': is_live,
 122         }
 123
 124
 125 class WDRPageIE(InfoExtractor):
 126     _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
 127     _PAGE_REGEX = r'/(?:mediathek/)?(?:[^/]+/)*(?P<display_id>[^/]+)\.html'
 128     _VALID_URL = r'https?://(?:www\d?\.)?(?:(?:kinder\.)?wdr\d?|sportschau)\.de' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL
 129
 130     _TESTS = [
 131         {
 132             'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
 133             # HDS download, MD5 is unstable
 134             'info_dict': {
 135                 'id': 'mdb-1058683',
 136                 'ext': 'flv',
 137                 'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
 138                 'title': 'Geheimnis Aachener Dom',
 139                 'alt_title': 'Doku am Freitag',
 140                 'upload_date': '20160304',
 141                 'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
 142                 'is_live': False,
 143                 'subtitles': {'de': [{
 144                     'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
 145                     'ext': 'ttml',
 146                 }]},
 147             },
 148             'skip': 'HTTP Error 404: Not Found',
 149         },
 150         {
 151             'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
 152             'md5': 'f4c1f96d01cf285240f53ea4309663d8',
 153             'info_dict': {
 154                 'id': 'mdb-1072000',
 155                 'ext': 'mp3',
 156                 'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
 157                 'title': 'Schriftstellerin Juli Zeh',
 158                 'alt_title': 'WDR 3 Gespräch am Samstag',
 159                 'upload_date': '20160312',
 160                 'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
 161                 'is_live': False,
 162                 'subtitles': {}
 163             },
 164             'skip': 'HTTP Error 404: Not Found',
 165         },
 166         {
 167             'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
 168             'info_dict': {
 169                 'id': 'mdb-1406149',
 170                 'ext': 'mp4',
 171                 '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}$',
 172                 'alt_title': 'WDR Fernsehen Live',
 173                 'upload_date': '20150101',
 174                 'is_live': True,
 175             },
 176             'params': {
 177                 'skip_download': True,  # m3u8 download
 178             },
 179         },
 180         {
 181             'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
 182             'playlist_mincount': 7,
 183             'info_dict': {
 184                 'id': 'aktuelle-stunde-120',
 185             },
 186         },
 187         {
 188             'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
 189             'info_dict': {
 190                 'id': 'mdb-1552552',
 191                 'ext': 'mp4',
 192                 'upload_date': 're:^[0-9]{8}$',
 193                 'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
 194             },
 195             'skip': 'The id changes from week to week because of the new episode'
 196         },
 197         {
 198             'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
 199             'md5': '803138901f6368ee497b4d195bb164f2',
 200             'info_dict': {
 201                 'id': 'mdb-186083',
 202                 'ext': 'mp4',
 203                 'upload_date': '20130919',
 204                 'title': 'Sachgeschichte - Achterbahn ',
 205             },
 206         },
 207         {
 208             'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
 209             # Live stream, MD5 unstable
 210             'info_dict': {
 211                 'id': 'mdb-869971',
 212                 'ext': 'mp4',
 213                 'title': r're:^COSMO Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 214                 'upload_date': '20160101',
 215             },
 216             'params': {
 217                 'skip_download': True,  # m3u8 download
 218             }
 219         },
 220         {
 221             'url': 'http://www.sportschau.de/handballem2018/handball-nationalmannschaft-em-stolperstein-vorrunde-100.html',
 222             'info_dict': {
 223                 'id': 'mdb-1556012',
 224                 'ext': 'mp4',
 225                 'title': 'DHB-Vizepräsident Bob Hanning - "Die Weltspitze ist extrem breit"',
 226                 'upload_date': '20180111',
 227             },
 228             'params': {
 229                 'skip_download': True,
 230             },
 231         },
 232         {
 233             'url': 'http://www.sportschau.de/handballem2018/audio-vorschau---die-handball-em-startet-mit-grossem-favoritenfeld-100.html',
 234             'only_matching': True,
 235         },
 236         {
 237             'url': 'https://kinder.wdr.de/tv/die-sendung-mit-dem-elefanten/av/video-folge---astronaut-100.html',
 238             'only_matching': True,
 239         },
 240     ]
 241
 242     def _real_extract(self, url):
 243         mobj = re.match(self._VALID_URL, url)
 244         display_id = mobj.group('display_id')
 245         webpage = self._download_webpage(url, display_id)
 246
 247         entries = []
 248
 249         # Article with several videos
 250
 251         # for wdr.de the data-extension is in a tag with the class "mediaLink"
 252         # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
 253         # for wdrmaus, in a tag with the class "videoButton" (previously a link
 254         # to the page in a multiline "videoLink"-tag)
 255         for mobj in re.finditer(
 256             r'''(?sx)class=
 257                     (?:
 258                         (["\'])(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b.*?\1[^>]+|
 259                         (["\'])videoLink\b.*?\2[\s]*>\n[^\n]*
 260                     )data-extension=(["\'])(?P<data>(?:(?!\3).)+)\3
 261                     ''', webpage):
 262             media_link_obj = self._parse_json(
 263                 mobj.group('data'), display_id, transform_source=js_to_json,
 264                 fatal=False)
 265             if not media_link_obj:
 266                 continue
 267             jsonp_url = try_get(
 268                 media_link_obj, lambda x: x['mediaObj']['url'], compat_str)
 269             if jsonp_url:
 270                 entries.append(self.url_result(jsonp_url, ie=WDRIE.ie_key()))
 271
 272         # Playlist (e.g. https://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html)
 273         if not entries:
 274             entries = [
 275                 self.url_result(
 276                     compat_urlparse.urljoin(url, mobj.group('href')),
 277                     ie=WDRPageIE.ie_key())
 278                 for mobj in re.finditer(
 279                     r'<a[^>]+\bhref=(["\'])(?P<href>(?:(?!\1).)+)\1[^>]+\bdata-extension=',
 280                     webpage) if re.match(self._PAGE_REGEX, mobj.group('href'))
 281             ]
 282
 283         return self.playlist_result(entries, playlist_id=display_id)
 284
 285
 286 class WDRElefantIE(InfoExtractor):
 287     _VALID_URL = r'https?://(?:www\.)wdrmaus\.de/elefantenseite/#(?P<id>.+)'
 288     _TEST = {
 289         'url': 'http://www.wdrmaus.de/elefantenseite/#folge_ostern_2015',
 290         'info_dict': {
 291             'title': 'Folge Oster-Spezial 2015',
 292             'id': 'mdb-1088195',
 293             'ext': 'mp4',
 294             'age_limit': None,
 295             'upload_date': '20150406'
 296         },
 297         'params': {
 298             'skip_download': True,
 299         },
 300     }
 301
 302     def _real_extract(self, url):
 303         display_id = self._match_id(url)
 304
 305         # Table of Contents seems to always be at this address, so fetch it directly.
 306         # The website fetches configurationJS.php5, which links to tableOfContentsJS.php5.
 307         table_of_contents = self._download_json(
 308             'https://www.wdrmaus.de/elefantenseite/data/tableOfContentsJS.php5',
 309             display_id)
 310         if display_id not in table_of_contents:
 311             raise ExtractorError(
 312                 'No entry in site\'s table of contents for this URL. '
 313                 'Is the fragment part of the URL (after the #) correct?',
 314                 expected=True)
 315         xml_metadata_path = table_of_contents[display_id]['xmlPath']
 316         xml_metadata = self._download_xml(
 317             'https://www.wdrmaus.de/elefantenseite/' + xml_metadata_path,
 318             display_id)
 319         zmdb_url_element = xml_metadata.find('./movie/zmdb_url')
 320         if zmdb_url_element is None:
 321             raise ExtractorError(
 322                 '%s is not a video' % display_id, expected=True)
 323         return self.url_result(zmdb_url_element.text, ie=WDRIE.ie_key())
 324
 325
 326 class WDRMobileIE(InfoExtractor):
 327     _VALID_URL = r'''(?x)
 328         https?://mobile-ondemand\.wdr\.de/
 329         .*?/fsk(?P<age_limit>[0-9]+)
 330         /[0-9]+/[0-9]+/
 331         (?P<id>[0-9]+)_(?P<title>[0-9]+)'''
 332     IE_NAME = 'wdr:mobile'
 333     _TEST = {
 334         'url': 'http://mobile-ondemand.wdr.de/CMS2010/mdb/ondemand/weltweit/fsk0/42/421735/421735_4283021.mp4',
 335         'info_dict': {
 336             'title': '4283021',
 337             'id': '421735',
 338             'ext': 'mp4',
 339             'age_limit': 0,
 340         },
 341         'skip': 'Problems with loading data.'
 342     }
 343
 344     def _real_extract(self, url):
 345         mobj = re.match(self._VALID_URL, url)
 346         return {
 347             'id': mobj.group('id'),
 348             'title': mobj.group('title'),
 349             'age_limit': int(mobj.group('age_limit')),
 350             'url': url,
 351             'http_headers': {
 352                 'User-Agent': 'mobile',
 353             },
 354         }