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