youtube_dlc/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 )
  21
  22
  23 class WDRIE(InfoExtractor):
  24     _VALID_URL = r'https?://deviceids-medp\.wdr\.de/ondemand/\d+/(?P<id>\d+)\.js'
  25     _GEO_COUNTRIES = ['DE']
  26     _TEST = {
  27         'url': 'http://deviceids-medp.wdr.de/ondemand/155/1557833.js',
  28         'info_dict': {
  29             'id': 'mdb-1557833',
  30             'ext': 'mp4',
  31             'title': 'Biathlon-Staffel verpasst Podest bei Olympia-Generalprobe',
  32             'upload_date': '20180112',
  33         },
  34     }
  35
  36     def _real_extract(self, url):
  37         video_id = self._match_id(url)
  38
  39         metadata = self._download_json(
  40             url, video_id, transform_source=strip_jsonp)
  41
  42         is_live = metadata.get('mediaType') == 'live'
  43
  44         tracker_data = metadata['trackerData']
  45         media_resource = metadata['mediaResource']
  46
  47         formats = []
  48         subtitles = {}
  49
  50         # check if the metadata contains a direct URL to a file
  51         for kind, media_resource in media_resource.items():
  52             if kind == 'captionsHash':
  53                 for ext, url in media_resource.items():
  54                     subtitles.setdefault('de', []).append({
  55                         'url': url,
  56                         'ext': ext,
  57                     })
  58                 continue
  59
  60             if kind not in ('dflt', 'alt'):
  61                 continue
  62
  63             for tag_name, medium_url in media_resource.items():
  64                 if tag_name not in ('videoURL', 'audioURL'):
  65                     continue
  66
  67                 ext = determine_ext(medium_url)
  68                 if ext == 'm3u8':
  69                     formats.extend(self._extract_m3u8_formats(
  70                         medium_url, video_id, 'mp4', 'm3u8_native',
  71                         m3u8_id='hls'))
  72                 elif ext == 'f4m':
  73                     manifest_url = update_url_query(
  74                         medium_url, {'hdcore': '3.2.0', 'plugin': 'aasp-3.2.0.77.18'})
  75                     formats.extend(self._extract_f4m_formats(
  76                         manifest_url, video_id, f4m_id='hds', fatal=False))
  77                 elif ext == 'smil':
  78                     formats.extend(self._extract_smil_formats(
  79                         medium_url, 'stream', fatal=False))
  80                 else:
  81                     a_format = {
  82                         'url': medium_url
  83                     }
  84                     if ext == 'unknown_video':
  85                         urlh = self._request_webpage(
  86                             medium_url, video_id, note='Determining extension')
  87                         ext = urlhandle_detect_ext(urlh)
  88                         a_format['ext'] = ext
  89                     formats.append(a_format)
  90
  91         self._sort_formats(formats)
  92
  93         title = tracker_data['trackerClipTitle']
  94
  95         return {
  96             'id': tracker_data.get('trackerClipId', video_id),
  97             'title': self._live_title(title) if is_live else title,
  98             'alt_title': tracker_data.get('trackerClipSubcategory'),
  99             'formats': formats,
 100             'subtitles': subtitles,
 101             'upload_date': unified_strdate(tracker_data.get('trackerClipAirTime')),
 102             'is_live': is_live,
 103         }
 104
 105
 106 class WDRPageIE(InfoExtractor):
 107     _CURRENT_MAUS_URL = r'https?://(?:www\.)wdrmaus.de/(?:[^/]+/){1,2}[^/?#]+\.php5'
 108     _PAGE_REGEX = r'/(?:mediathek/)?(?:[^/]+/)*(?P<display_id>[^/]+)\.html'
 109     _VALID_URL = r'https?://(?:www\d?\.)?(?:wdr\d?|sportschau)\.de' + _PAGE_REGEX + '|' + _CURRENT_MAUS_URL
 110
 111     _TESTS = [
 112         {
 113             'url': 'http://www1.wdr.de/mediathek/video/sendungen/doku-am-freitag/video-geheimnis-aachener-dom-100.html',
 114             # HDS download, MD5 is unstable
 115             'info_dict': {
 116                 'id': 'mdb-1058683',
 117                 'ext': 'flv',
 118                 'display_id': 'doku-am-freitag/video-geheimnis-aachener-dom-100',
 119                 'title': 'Geheimnis Aachener Dom',
 120                 'alt_title': 'Doku am Freitag',
 121                 'upload_date': '20160304',
 122                 'description': 'md5:87be8ff14d8dfd7a7ee46f0299b52318',
 123                 'is_live': False,
 124                 'subtitles': {'de': [{
 125                     'url': 'http://ondemand-ww.wdr.de/medp/fsk0/105/1058683/1058683_12220974.xml',
 126                     'ext': 'ttml',
 127                 }]},
 128             },
 129             'skip': 'HTTP Error 404: Not Found',
 130         },
 131         {
 132             'url': 'http://www1.wdr.de/mediathek/audio/wdr3/wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100.html',
 133             'md5': 'f4c1f96d01cf285240f53ea4309663d8',
 134             'info_dict': {
 135                 'id': 'mdb-1072000',
 136                 'ext': 'mp3',
 137                 'display_id': 'wdr3-gespraech-am-samstag/audio-schriftstellerin-juli-zeh-100',
 138                 'title': 'Schriftstellerin Juli Zeh',
 139                 'alt_title': 'WDR 3 Gespräch am Samstag',
 140                 'upload_date': '20160312',
 141                 'description': 'md5:e127d320bc2b1f149be697ce044a3dd7',
 142                 'is_live': False,
 143                 'subtitles': {}
 144             },
 145             'skip': 'HTTP Error 404: Not Found',
 146         },
 147         {
 148             'url': 'http://www1.wdr.de/mediathek/video/live/index.html',
 149             'info_dict': {
 150                 'id': 'mdb-1406149',
 151                 'ext': 'mp4',
 152                 '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}$',
 153                 'alt_title': 'WDR Fernsehen Live',
 154                 'upload_date': '20150101',
 155                 'is_live': True,
 156             },
 157             'params': {
 158                 'skip_download': True,  # m3u8 download
 159             },
 160         },
 161         {
 162             'url': 'http://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html',
 163             'playlist_mincount': 7,
 164             'info_dict': {
 165                 'id': 'aktuelle-stunde-120',
 166             },
 167         },
 168         {
 169             'url': 'http://www.wdrmaus.de/aktuelle-sendung/index.php5',
 170             'info_dict': {
 171                 'id': 'mdb-1552552',
 172                 'ext': 'mp4',
 173                 'upload_date': 're:^[0-9]{8}$',
 174                 'title': 're:^Die Sendung mit der Maus vom [0-9.]{10}$',
 175             },
 176             'skip': 'The id changes from week to week because of the new episode'
 177         },
 178         {
 179             'url': 'http://www.wdrmaus.de/filme/sachgeschichten/achterbahn.php5',
 180             'md5': '803138901f6368ee497b4d195bb164f2',
 181             'info_dict': {
 182                 'id': 'mdb-186083',
 183                 'ext': 'mp4',
 184                 'upload_date': '20130919',
 185                 'title': 'Sachgeschichte - Achterbahn ',
 186             },
 187         },
 188         {
 189             'url': 'http://www1.wdr.de/radio/player/radioplayer116~_layout-popupVersion.html',
 190             # Live stream, MD5 unstable
 191             'info_dict': {
 192                 'id': 'mdb-869971',
 193                 'ext': 'mp4',
 194                 'title': r're:^COSMO Livestream [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 195                 'upload_date': '20160101',
 196             },
 197             'params': {
 198                 'skip_download': True,  # m3u8 download
 199             }
 200         },
 201         {
 202             'url': 'http://www.sportschau.de/handballem2018/handball-nationalmannschaft-em-stolperstein-vorrunde-100.html',
 203             'info_dict': {
 204                 'id': 'mdb-1556012',
 205                 'ext': 'mp4',
 206                 'title': 'DHB-Vizepräsident Bob Hanning - "Die Weltspitze ist extrem breit"',
 207                 'upload_date': '20180111',
 208             },
 209             'params': {
 210                 'skip_download': True,
 211             },
 212         },
 213         {
 214             'url': 'http://www.sportschau.de/handballem2018/audio-vorschau---die-handball-em-startet-mit-grossem-favoritenfeld-100.html',
 215             'only_matching': True,
 216         }
 217     ]
 218
 219     def _real_extract(self, url):
 220         mobj = re.match(self._VALID_URL, url)
 221         display_id = mobj.group('display_id')
 222         webpage = self._download_webpage(url, display_id)
 223
 224         entries = []
 225
 226         # Article with several videos
 227
 228         # for wdr.de the data-extension is in a tag with the class "mediaLink"
 229         # for wdr.de radio players, in a tag with the class "wdrrPlayerPlayBtn"
 230         # for wdrmaus, in a tag with the class "videoButton" (previously a link
 231         # to the page in a multiline "videoLink"-tag)
 232         for mobj in re.finditer(
 233             r'''(?sx)class=
 234                     (?:
 235                         (["\'])(?:mediaLink|wdrrPlayerPlayBtn|videoButton)\b.*?\1[^>]+|
 236                         (["\'])videoLink\b.*?\2[\s]*>\n[^\n]*
 237                     )data-extension=(["\'])(?P<data>(?:(?!\3).)+)\3
 238                     ''', webpage):
 239             media_link_obj = self._parse_json(
 240                 mobj.group('data'), display_id, transform_source=js_to_json,
 241                 fatal=False)
 242             if not media_link_obj:
 243                 continue
 244             jsonp_url = try_get(
 245                 media_link_obj, lambda x: x['mediaObj']['url'], compat_str)
 246             if jsonp_url:
 247                 entries.append(self.url_result(jsonp_url, ie=WDRIE.ie_key()))
 248
 249         # Playlist (e.g. https://www1.wdr.de/mediathek/video/sendungen/aktuelle-stunde/aktuelle-stunde-120.html)
 250         if not entries:
 251             entries = [
 252                 self.url_result(
 253                     compat_urlparse.urljoin(url, mobj.group('href')),
 254                     ie=WDRPageIE.ie_key())
 255                 for mobj in re.finditer(
 256                     r'<a[^>]+\bhref=(["\'])(?P<href>(?:(?!\1).)+)\1[^>]+\bdata-extension=',
 257                     webpage) if re.match(self._PAGE_REGEX, mobj.group('href'))
 258             ]
 259
 260         return self.playlist_result(entries, playlist_id=display_id)
 261
 262
 263 class WDRElefantIE(InfoExtractor):
 264     _VALID_URL = r'https?://(?:www\.)wdrmaus\.de/elefantenseite/#(?P<id>.+)'
 265     _TEST = {
 266         'url': 'http://www.wdrmaus.de/elefantenseite/#folge_ostern_2015',
 267         'info_dict': {
 268             'title': 'Folge Oster-Spezial 2015',
 269             'id': 'mdb-1088195',
 270             'ext': 'mp4',
 271             'age_limit': None,
 272             'upload_date': '20150406'
 273         },
 274         'params': {
 275             'skip_download': True,
 276         },
 277     }
 278
 279     def _real_extract(self, url):
 280         display_id = self._match_id(url)
 281
 282         # Table of Contents seems to always be at this address, so fetch it directly.
 283         # The website fetches configurationJS.php5, which links to tableOfContentsJS.php5.
 284         table_of_contents = self._download_json(
 285             'https://www.wdrmaus.de/elefantenseite/data/tableOfContentsJS.php5',
 286             display_id)
 287         if display_id not in table_of_contents:
 288             raise ExtractorError(
 289                 'No entry in site\'s table of contents for this URL. '
 290                 'Is the fragment part of the URL (after the #) correct?',
 291                 expected=True)
 292         xml_metadata_path = table_of_contents[display_id]['xmlPath']
 293         xml_metadata = self._download_xml(
 294             'https://www.wdrmaus.de/elefantenseite/' + xml_metadata_path,
 295             display_id)
 296         zmdb_url_element = xml_metadata.find('./movie/zmdb_url')
 297         if zmdb_url_element is None:
 298             raise ExtractorError(
 299                 '%s is not a video' % display_id, expected=True)
 300         return self.url_result(zmdb_url_element.text, ie=WDRIE.ie_key())
 301
 302
 303 class WDRMobileIE(InfoExtractor):
 304     _VALID_URL = r'''(?x)
 305         https?://mobile-ondemand\.wdr\.de/
 306         .*?/fsk(?P<age_limit>[0-9]+)
 307         /[0-9]+/[0-9]+/
 308         (?P<id>[0-9]+)_(?P<title>[0-9]+)'''
 309     IE_NAME = 'wdr:mobile'
 310     _TEST = {
 311         'url': 'http://mobile-ondemand.wdr.de/CMS2010/mdb/ondemand/weltweit/fsk0/42/421735/421735_4283021.mp4',
 312         'info_dict': {
 313             'title': '4283021',
 314             'id': '421735',
 315             'ext': 'mp4',
 316             'age_limit': 0,
 317         },
 318         'skip': 'Problems with loading data.'
 319     }
 320
 321     def _real_extract(self, url):
 322         mobj = re.match(self._VALID_URL, url)
 323         return {
 324             'id': mobj.group('id'),
 325             'title': mobj.group('title'),
 326             'age_limit': int(mobj.group('age_limit')),
 327             'url': url,
 328             'http_headers': {
 329                 'User-Agent': 'mobile',
 330             },
 331         }