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'''(?x)https?://
  26         (?:deviceids-medp\.wdr\.de/ondemand/\d+/|
  27            kinder\.wdr\.de/(?!mediathek/)[^#?]+-)
  28         (?P<id>\d+)\.(?:js|assetjsonp)
  29     '''
  30     _GEO_COUNTRIES = ['DE']
  31     _TEST = {
  32         'url': 'http://deviceids-medp.wdr.de/ondemand/155/1557833.js',
  33         'info_dict': {
  34             'id': 'mdb-1557833',
  35             'ext': 'mp4',
  36             'title': 'Biathlon-Staffel verpasst Podest bei Olympia-Generalprobe',
  37             'upload_date': '20180112',
  38         },
  39     }
  40
  41     def _real_extract(self, url):
  42         video_id = self._match_id(url)
  43
  44         metadata = self._download_json(
  45             url, video_id, transform_source=strip_jsonp)
  46
  47         is_live = metadata.get('mediaType') == 'live'
  48
  49         tracker_data = metadata['trackerData']
  50         title = tracker_data['trackerClipTitle']
  51         media_resource = metadata['mediaResource']
  52
  53         formats = []
  54         subtitles = {}
  55
  56         # check if the metadata contains a direct URL to a file
  57         for kind, media in media_resource.items():
  58             if kind == 'captionsHash':
  59                 for ext, url in media.items():
  60                     subtitles.setdefault('de', []).append({
  61                         'url': url,
  62                         'ext': ext,
  63                     })
  64                 continue
  65
  66             if kind not in ('dflt', 'alt'):
  67                 continue
  68             if not isinstance(media, dict):
  69                 continue
  70
  71             for tag_name, medium_url in media.items():
  72                 if tag_name not in ('videoURL', 'audioURL'):
  73                     continue
  74
  75                 ext = determine_ext(medium_url)
  76                 if ext == 'm3u8':
  77                     formats.extend(self._extract_m3u8_formats(
  78                         medium_url, video_id, 'mp4', 'm3u8_native',
  79                         m3u8_id='hls'))
  80                 elif ext == 'f4m':
  81                     manifest_url = update_url_query(
  82                         medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
  83                     formats.extend(self._extract_f4m_formats(
  84                         manifest_url, video_id, f4m_id='hds', fatal=False))
  85                 elif ext == 'smil':
  86                     formats.extend(self._extract_smil_formats(
  87                         medium_url, 'stream', fatal=False))
  88                 else:
  89                     a_format = {
  90                         'url': medium_url
  91                     }
  92                     if ext == 'unknown_video':
  93                         urlh = self._request_webpage(
  94                             medium_url, video_id, note='Determining extension')
  95                         ext = urlhandle_detect_ext(urlh)
  96                         a_format['ext'] = ext
  97                     formats.append(a_format)
  98
  99         self._sort_formats(formats)
 100
 101         caption_url = media_resource.get('captionURL')
 102         if caption_url:
 103             subtitles['de'] = [{
 104                 'url': caption_url,
 105                 'ext': 'ttml',
 106             }]
 107         captions_hash = media_resource.get('captionsHash')
 108         if isinstance(captions_hash, dict):
 109             for ext, format_url in captions_hash.items():
 110                 format_url = url_or_none(format_url)
 111                 if not format_url:
 112                     continue
 113                 subtitles.setdefault('de', []).append({
 114                     'url': format_url,
 115                     'ext': determine_ext(format_url, None) or ext,
 116                 })
 117
 118         return {
 119             'id': tracker_data.get('trackerClipId', video_id),
 120             'title': title,
 121             'alt_title': tracker_data.get('trackerClipSubcategory'),
 122             'formats': formats,
 123             'subtitles': subtitles,
 124             'upload_date': unified_strdate(tracker_data.get('trackerClipAirTime')),
 125             'is_live': is_live,
 126         }
 127
 128
 129 class WDRPageIE(InfoExtractor):
 130     _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
 131     _PAGE_REGEX = r'/(?:mediathek/)?(?:[^/]+/)*(?P<display_id>[^/]+)\.html'
 132     _VALID_URL = r'https?://(?:www\d?\.)?(?:(?:kinder\.)?wdr\d?|sportschau)\.de' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL
 133
 134     _TESTS = [
 135         {
 136             'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
 137             # HDS download, MD5 is unstable
 138             'info_dict': {
 139                 'id': 'mdb-1058683',
 140                 'ext': 'flv',
 141                 'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
 142                 'title': 'Geheimnis Aachener Dom',
 143                 'alt_title': 'Doku am Freitag',
 144                 'upload_date': '20160304',
 145                 'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
 146                 'is_live': False,
 147                 'subtitles': {'de': [{
 148                     'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
 149                     'ext': 'ttml',
 150                 }]},
 151             },
 152             'skip': 'HTTP Error 404: Not Found',
 153         },
 154         {
 155             'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
 156             'md5': 'f4c1f96d01cf285240f53ea4309663d8',
 157             'info_dict': {
 158                 'id': 'mdb-1072000',
 159                 'ext': 'mp3',
 160                 'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
 161                 'title': 'Schriftstellerin Juli Zeh',
 162                 'alt_title': 'WDR 3 Gespräch am Samstag',
 163                 'upload_date': '20160312',
 164                 'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
 165                 'is_live': False,
 166                 'subtitles': {}
 167             },
 168             'skip': 'HTTP Error 404: Not Found',
 169         },
 170         {
 171             'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
 172             'info_dict': {
 173                 'id': 'mdb-1406149',
 174                 'ext': 'mp4',
 175                 '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}$',
 176                 'alt_title': 'WDR Fernsehen Live',
 177                 'upload_date': '20150101',
 178                 'is_live': True,
 179             },
 180             'params': {
 181                 'skip_download': True,  # m3u8 download
 182             },
 183         },
 184         {
 185             'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
 186             'playlist_mincount': 7,
 187             'info_dict': {
 188                 'id': 'aktuelle-stunde-120',
 189             },
 190         },
 191         {
 192             'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
 193             'info_dict': {
 194                 'id': 'mdb-1552552',
 195                 'ext': 'mp4',
 196                 'upload_date': 're:^[0-9]{8}$',
 197                 'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
 198             },
 199             'skip': 'The id changes from week to week because of the new episode'
 200         },
 201         {
 202             'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
 203             'md5': '803138901f6368ee497b4d195bb164f2',
 204             'info_dict': {
 205                 'id': 'mdb-186083',
 206                 'ext': 'mp4',
 207                 'upload_date': '20130919',
 208                 'title': 'Sachgeschichte - Achterbahn ',
 209             },
 210         },
 211         {
 212             'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
 213             # Live stream, MD5 unstable
 214             'info_dict': {
 215                 'id': 'mdb-869971',
 216                 'ext': 'mp4',
 217                 'title': r're:^COSMO Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 218                 'upload_date': '20160101',
 219             },
 220             'params': {
 221                 'skip_download': True,  # m3u8 download
 222             }
 223         },
 224         {
 225             'url': 'http://www.sportschau.de/handballem2018/handball-nationalmannschaft-em-stolperstein-vorrunde-100.html',
 226             'info_dict': {
 227                 'id': 'mdb-1556012',
 228                 'ext': 'mp4',
 229                 'title': 'DHB-Vizepräsident Bob Hanning - "Die Weltspitze ist extrem breit"',
 230                 'upload_date': '20180111',
 231             },
 232             'params': {
 233                 'skip_download': True,
 234             },
 235         },
 236         {
 237             'url': 'http://www.sportschau.de/handballem2018/audio-vorschau---die-handball-em-startet-mit-grossem-favoritenfeld-100.html',
 238             'only_matching': True,
 239         },
 240         {
 241             'url': 'https://kinder.wdr.de/tv/die-sendung-mit-dem-elefanten/av/video-folge---astronaut-100.html',
 242             'only_matching': True,
 243         },
 244     ]
 245
 246     def _real_extract(self, url):
 247         mobj = self._match_valid_url(url)
 248         display_id = mobj.group('display_id')
 249         webpage = self._download_webpage(url, display_id)
 250
 251         entries = []
 252
 253         # Article with several videos
 254
 255         # for wdr.de the data-extension is in a tag with the class "mediaLink"
 256         # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
 257         # for wdrmaus, in a tag with the class "videoButton" (previously a link
 258         # to the page in a multiline "videoLink"-tag)
 259         for mobj in re.finditer(
 260             r'''(?sx)class=
 261                     (?:
 262                         (["\'])(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b.*?\1[^>]+|
 263                         (["\'])videoLink\b.*?\2[\s]*>\n[^\n]*
 264                     )data-extension=(["\'])(?P<data>(?:(?!\3).)+)\3
 265                     ''', webpage):
 266             media_link_obj = self._parse_json(
 267                 mobj.group('data'), display_id, transform_source=js_to_json,
 268                 fatal=False)
 269             if not media_link_obj:
 270                 continue
 271             jsonp_url = try_get(
 272                 media_link_obj, lambda x: x['mediaObj']['url'], compat_str)
 273             if jsonp_url:
 274                 entries.append(self.url_result(jsonp_url, ie=WDRIE.ie_key()))
 275
 276         # Playlist (e.g. https://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html)
 277         if not entries:
 278             entries = [
 279                 self.url_result(
 280                     compat_urlparse.urljoin(url, mobj.group('href')),
 281                     ie=WDRPageIE.ie_key())
 282                 for mobj in re.finditer(
 283                     r'<a[^>]+\bhref=(["\'])(?P<href>(?:(?!\1).)+)\1[^>]+\bdata-extension=',
 284                     webpage) if re.match(self._PAGE_REGEX, mobj.group('href'))
 285             ]
 286
 287         return self.playlist_result(entries, playlist_id=display_id)
 288
 289
 290 class WDRElefantIE(InfoExtractor):
 291     _VALID_URL = r'https?://(?:www\.)wdrmaus\.de/elefantenseite/#(?P<id>.+)'
 292     _TEST = {
 293         'url': 'http://www.wdrmaus.de/elefantenseite/#folge_ostern_2015',
 294         'info_dict': {
 295             'title': 'Folge Oster-Spezial 2015',
 296             'id': 'mdb-1088195',
 297             'ext': 'mp4',
 298             'age_limit': None,
 299             'upload_date': '20150406'
 300         },
 301         'params': {
 302             'skip_download': True,
 303         },
 304     }
 305
 306     def _real_extract(self, url):
 307         display_id = self._match_id(url)
 308
 309         # Table of Contents seems to always be at this address, so fetch it directly.
 310         # The website fetches configurationJS.php5, which links to tableOfContentsJS.php5.
 311         table_of_contents = self._download_json(
 312             'https://www.wdrmaus.de/elefantenseite/data/tableOfContentsJS.php5',
 313             display_id)
 314         if display_id not in table_of_contents:
 315             raise ExtractorError(
 316                 'No entry in site\'s table of contents for this URL. '
 317                 'Is the fragment part of the URL (after the #) correct?',
 318                 expected=True)
 319         xml_metadata_path = table_of_contents[display_id]['xmlPath']
 320         xml_metadata = self._download_xml(
 321             'https://www.wdrmaus.de/elefantenseite/' + xml_metadata_path,
 322             display_id)
 323         zmdb_url_element = xml_metadata.find('./movie/zmdb_url')
 324         if zmdb_url_element is None:
 325             raise ExtractorError(
 326                 '%s is not a video' % display_id, expected=True)
 327         return self.url_result(zmdb_url_element.text, ie=WDRIE.ie_key())
 328
 329
 330 class WDRMobileIE(InfoExtractor):
 331     _VALID_URL = r'''(?x)
 332         https?://mobile-ondemand\.wdr\.de/
 333         .*?/fsk(?P<age_limit>[0-9]+)
 334         /[0-9]+/[0-9]+/
 335         (?P<id>[0-9]+)_(?P<title>[0-9]+)'''
 336     IE_NAME = 'wdr:mobile'
 337     _TEST = {
 338         'url': 'http://mobile-ondemand.wdr.de/CMS2010/mdb/ondemand/weltweit/fsk0/42/421735/421735_4283021.mp4',
 339         'info_dict': {
 340             'title': '4283021',
 341             'id': '421735',
 342             'ext': 'mp4',
 343             'age_limit': 0,
 344         },
 345         'skip': 'Problems with loading data.'
 346     }
 347
 348     def _real_extract(self, url):
 349         mobj = self._match_valid_url(url)
 350         return {
 351             'id': mobj.group('id'),
 352             'title': mobj.group('title'),
 353             'age_limit': int(mobj.group('age_limit')),
 354             'url': url,
 355             'http_headers': {
 356                 'User-Agent': 'mobile',
 357             },
 358         }