Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ard.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from .generic import GenericIE
   8 from ..utils import (
   9     determine_ext,
  10     ExtractorError,
  11     int_or_none,
  12     parse_duration,
  13     qualities,
  14     str_or_none,
  15     try_get,
  16     unified_strdate,
  17     unified_timestamp,
  18     update_url_query,
  19     url_or_none,
  20     xpath_text,
  21 )
  22 from ..compat import compat_etree_fromstring
  23
  24
  25 class ARDMediathekIE(InfoExtractor):
  26     IE_NAME = 'ARD:mediathek'
  27     _VALID_URL = r'^https?://(?:(?:(?:www|classic)\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de|one\.ard\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
  28
  29     _TESTS = [{
  30         # available till 26.07.2022
  31         'url': 'http://www.ardmediathek.de/tv/S%C3%9CDLICHT/Was-ist-die-Kunst-der-Zukunft-liebe-Ann/BR-Fernsehen/Video?bcastId=34633636&documentId=44726822',
  32         'info_dict': {
  33             'id': '44726822',
  34             'ext': 'mp4',
  35             'title': 'Was ist die Kunst der Zukunft, liebe Anna McCarthy?',
  36             'description': 'md5:4ada28b3e3b5df01647310e41f3a62f5',
  37             'duration': 1740,
  38         },
  39         'params': {
  40             # m3u8 download
  41             'skip_download': True,
  42         }
  43     }, {
  44         'url': 'https://one.ard.de/tv/Mord-mit-Aussicht/Mord-mit-Aussicht-6-39-T%C3%B6dliche-Nach/ONE/Video?bcastId=46384294&documentId=55586872',
  45         'only_matching': True,
  46     }, {
  47         # audio
  48         'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
  49         'only_matching': True,
  50     }, {
  51         'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
  52         'only_matching': True,
  53     }, {
  54         # audio
  55         'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158',
  56         'only_matching': True,
  57     }, {
  58         'url': 'https://classic.ardmediathek.de/tv/Panda-Gorilla-Co/Panda-Gorilla-Co-Folge-274/Das-Erste/Video?bcastId=16355486&documentId=58234698',
  59         'only_matching': True,
  60     }]
  61
  62     @classmethod
  63     def suitable(cls, url):
  64         return False if ARDBetaMediathekIE.suitable(url) else super(ARDMediathekIE, cls).suitable(url)
  65
  66     def _extract_media_info(self, media_info_url, webpage, video_id):
  67         media_info = self._download_json(
  68             media_info_url, video_id, 'Downloading media JSON')
  69
  70         formats = self._extract_formats(media_info, video_id)
  71
  72         if not formats:
  73             if '"fsk"' in webpage:
  74                 raise ExtractorError(
  75                     'This video is only available after 20:00', expected=True)
  76             elif media_info.get('_geoblocked'):
  77                 raise ExtractorError('This video is not available due to geo restriction', expected=True)
  78
  79         self._sort_formats(formats)
  80
  81         duration = int_or_none(media_info.get('_duration'))
  82         thumbnail = media_info.get('_previewImage')
  83         is_live = media_info.get('_isLive') is True
  84
  85         subtitles = {}
  86         subtitle_url = media_info.get('_subtitleUrl')
  87         if subtitle_url:
  88             subtitles['de'] = [{
  89                 'ext': 'ttml',
  90                 'url': subtitle_url,
  91             }]
  92
  93         return {
  94             'id': video_id,
  95             'duration': duration,
  96             'thumbnail': thumbnail,
  97             'is_live': is_live,
  98             'formats': formats,
  99             'subtitles': subtitles,
 100         }
 101
 102     def _extract_formats(self, media_info, video_id):
 103         type_ = media_info.get('_type')
 104         media_array = media_info.get('_mediaArray', [])
 105         formats = []
 106         for num, media in enumerate(media_array):
 107             for stream in media.get('_mediaStreamArray', []):
 108                 stream_urls = stream.get('_stream')
 109                 if not stream_urls:
 110                     continue
 111                 if not isinstance(stream_urls, list):
 112                     stream_urls = [stream_urls]
 113                 quality = stream.get('_quality')
 114                 server = stream.get('_server')
 115                 for stream_url in stream_urls:
 116                     if not url_or_none(stream_url):
 117                         continue
 118                     ext = determine_ext(stream_url)
 119                     if quality != 'auto' and ext in ('f4m', 'm3u8'):
 120                         continue
 121                     if ext == 'f4m':
 122                         formats.extend(self._extract_f4m_formats(
 123                             update_url_query(stream_url, {
 124                                 'hdcore': '3.1.1',
 125                                 'plugin': 'aasp-3.1.1.69.124'
 126                             }),
 127                             video_id, f4m_id='hds', fatal=False))
 128                     elif ext == 'm3u8':
 129                         formats.extend(self._extract_m3u8_formats(
 130                             stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
 131                     else:
 132                         if server and server.startswith('rtmp'):
 133                             f = {
 134                                 'url': server,
 135                                 'play_path': stream_url,
 136                                 'format_id': 'a%s-rtmp-%s' % (num, quality),
 137                             }
 138                         else:
 139                             f = {
 140                                 'url': stream_url,
 141                                 'format_id': 'a%s-%s-%s' % (num, ext, quality)
 142                             }
 143                         m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
 144                         if m:
 145                             f.update({
 146                                 'width': int(m.group('width')),
 147                                 'height': int(m.group('height')),
 148                             })
 149                         if type_ == 'audio':
 150                             f['vcodec'] = 'none'
 151                         formats.append(f)
 152         return formats
 153
 154     def _real_extract(self, url):
 155         # determine video id from url
 156         m = re.match(self._VALID_URL, url)
 157
 158         document_id = None
 159
 160         numid = re.search(r'documentId=([0-9]+)', url)
 161         if numid:
 162             document_id = video_id = numid.group(1)
 163         else:
 164             video_id = m.group('video_id')
 165
 166         webpage = self._download_webpage(url, video_id)
 167
 168         ERRORS = (
 169             ('>Leider liegt eine Störung vor.', 'Video %s is unavailable'),
 170             ('>Der gewünschte Beitrag ist nicht mehr verfügbar.<',
 171              'Video %s is no longer available'),
 172         )
 173
 174         for pattern, message in ERRORS:
 175             if pattern in webpage:
 176                 raise ExtractorError(message % video_id, expected=True)
 177
 178         if re.search(r'[\?&]rss($|[=&])', url):
 179             doc = compat_etree_fromstring(webpage.encode('utf-8'))
 180             if doc.tag == 'rss':
 181                 return GenericIE()._extract_rss(url, video_id, doc)
 182
 183         title = self._html_search_regex(
 184             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
 185              r'<meta name="dcterms\.title" content="(.*?)"/>',
 186              r'<h4 class="headline">(.*?)</h4>',
 187              r'<title[^>]*>(.*?)</title>'],
 188             webpage, 'title')
 189         description = self._html_search_meta(
 190             'dcterms.abstract', webpage, 'description', default=None)
 191         if description is None:
 192             description = self._html_search_meta(
 193                 'description', webpage, 'meta description', default=None)
 194         if description is None:
 195             description = self._html_search_regex(
 196                 r'<p\s+class="teasertext">(.+?)</p>',
 197                 webpage, 'teaser text', default=None)
 198
 199         # Thumbnail is sometimes not present.
 200         # It is in the mobile version, but that seems to use a different URL
 201         # structure altogether.
 202         thumbnail = self._og_search_thumbnail(webpage, default=None)
 203
 204         media_streams = re.findall(r'''(?x)
 205             mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
 206             "([^"]+)"''', webpage)
 207
 208         if media_streams:
 209             QUALITIES = qualities(['lo', 'hi', 'hq'])
 210             formats = []
 211             for furl in set(media_streams):
 212                 if furl.endswith('.f4m'):
 213                     fid = 'f4m'
 214                 else:
 215                     fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
 216                     fid = fid_m.group(1) if fid_m else None
 217                 formats.append({
 218                     'quality': QUALITIES(fid),
 219                     'format_id': fid,
 220                     'url': furl,
 221                 })
 222             self._sort_formats(formats)
 223             info = {
 224                 'formats': formats,
 225             }
 226         else:  # request JSON file
 227             if not document_id:
 228                 video_id = self._search_regex(
 229                     r'/play/(?:config|media)/(\d+)', webpage, 'media id')
 230             info = self._extract_media_info(
 231                 'http://www.ardmediathek.de/play/media/%s' % video_id,
 232                 webpage, video_id)
 233
 234         info.update({
 235             'id': video_id,
 236             'title': self._live_title(title) if info.get('is_live') else title,
 237             'description': description,
 238             'thumbnail': thumbnail,
 239         })
 240
 241         return info
 242
 243
 244 class ARDIE(InfoExtractor):
 245     _VALID_URL = r'(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
 246     _TESTS = [{
 247         # available till 14.02.2019
 248         'url': 'http://www.daserste.de/information/talk/maischberger/videos/das-groko-drama-zerlegen-sich-die-volksparteien-video-102.html',
 249         'md5': '8e4ec85f31be7c7fc08a26cdbc5a1f49',
 250         'info_dict': {
 251             'display_id': 'das-groko-drama-zerlegen-sich-die-volksparteien-video',
 252             'id': '102',
 253             'ext': 'mp4',
 254             'duration': 4435.0,
 255             'title': 'Das GroKo-Drama: Zerlegen sich die Volksparteien?',
 256             'upload_date': '20180214',
 257             'thumbnail': r're:^https?://.*\.jpg$',
 258         },
 259     }, {
 260         'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
 261         'only_matching': True,
 262     }]
 263
 264     def _real_extract(self, url):
 265         mobj = re.match(self._VALID_URL, url)
 266         display_id = mobj.group('display_id')
 267
 268         player_url = mobj.group('mainurl') + '~playerXml.xml'
 269         doc = self._download_xml(player_url, display_id)
 270         video_node = doc.find('./video')
 271         upload_date = unified_strdate(xpath_text(
 272             video_node, './broadcastDate'))
 273         thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
 274
 275         formats = []
 276         for a in video_node.findall('.//asset'):
 277             f = {
 278                 'format_id': a.attrib['type'],
 279                 'width': int_or_none(a.find('./frameWidth').text),
 280                 'height': int_or_none(a.find('./frameHeight').text),
 281                 'vbr': int_or_none(a.find('./bitrateVideo').text),
 282                 'abr': int_or_none(a.find('./bitrateAudio').text),
 283                 'vcodec': a.find('./codecVideo').text,
 284                 'tbr': int_or_none(a.find('./totalBitrate').text),
 285             }
 286             if a.find('./serverPrefix').text:
 287                 f['url'] = a.find('./serverPrefix').text
 288                 f['playpath'] = a.find('./fileName').text
 289             else:
 290                 f['url'] = a.find('./fileName').text
 291             formats.append(f)
 292         self._sort_formats(formats)
 293
 294         return {
 295             'id': mobj.group('id'),
 296             'formats': formats,
 297             'display_id': display_id,
 298             'title': video_node.find('./title').text,
 299             'duration': parse_duration(video_node.find('./duration').text),
 300             'upload_date': upload_date,
 301             'thumbnail': thumbnail,
 302         }
 303
 304
 305 class ARDBetaMediathekIE(InfoExtractor):
 306     _VALID_URL = r'https://(?:beta|www)\.ardmediathek\.de/[^/]+/(?:player|live)/(?P<video_id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^/?#]+))?'
 307     _TESTS = [{
 308         'url': 'https://beta.ardmediathek.de/ard/player/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE/die-robuste-roswita',
 309         'md5': '2d02d996156ea3c397cfc5036b5d7f8f',
 310         'info_dict': {
 311             'display_id': 'die-robuste-roswita',
 312             'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE',
 313             'title': 'Tatort: Die robuste Roswita',
 314             'description': r're:^Der Mord.*trüber ist als die Ilm.',
 315             'duration': 5316,
 316             'thumbnail': 'https://img.ardmediathek.de/standard/00/55/43/59/34/-1774185891/16x9/960?mandant=ard',
 317             'upload_date': '20180826',
 318             'ext': 'mp4',
 319         },
 320     }, {
 321         'url': 'https://www.ardmediathek.de/ard/player/Y3JpZDovL3N3ci5kZS9hZXgvbzEwNzE5MTU/',
 322         'only_matching': True,
 323     }, {
 324         'url': 'https://www.ardmediathek.de/swr/live/Y3JpZDovL3N3ci5kZS8xMzQ4MTA0Mg',
 325         'only_matching': True,
 326     }]
 327
 328     def _real_extract(self, url):
 329         mobj = re.match(self._VALID_URL, url)
 330         video_id = mobj.group('video_id')
 331         display_id = mobj.group('display_id') or video_id
 332
 333         webpage = self._download_webpage(url, display_id)
 334         data_json = self._search_regex(r'window\.__APOLLO_STATE__\s*=\s*(\{.*);\n', webpage, 'json')
 335         data = self._parse_json(data_json, display_id)
 336
 337         res = {
 338             'id': video_id,
 339             'display_id': display_id,
 340         }
 341         formats = []
 342         subtitles = {}
 343         geoblocked = False
 344         for widget in data.values():
 345             if widget.get('_geoblocked') is True:
 346                 geoblocked = True
 347             if '_duration' in widget:
 348                 res['duration'] = int_or_none(widget['_duration'])
 349             if 'clipTitle' in widget:
 350                 res['title'] = widget['clipTitle']
 351             if '_previewImage' in widget:
 352                 res['thumbnail'] = widget['_previewImage']
 353             if 'broadcastedOn' in widget:
 354                 res['timestamp'] = unified_timestamp(widget['broadcastedOn'])
 355             if 'synopsis' in widget:
 356                 res['description'] = widget['synopsis']
 357             subtitle_url = url_or_none(widget.get('_subtitleUrl'))
 358             if subtitle_url:
 359                 subtitles.setdefault('de', []).append({
 360                     'ext': 'ttml',
 361                     'url': subtitle_url,
 362                 })
 363             if '_quality' in widget:
 364                 format_url = url_or_none(try_get(
 365                     widget, lambda x: x['_stream']['json'][0]))
 366                 if not format_url:
 367                     continue
 368                 ext = determine_ext(format_url)
 369                 if ext == 'f4m':
 370                     formats.extend(self._extract_f4m_formats(
 371                         format_url + '?hdcore=3.11.0',
 372                         video_id, f4m_id='hds', fatal=False))
 373                 elif ext == 'm3u8':
 374                     formats.extend(self._extract_m3u8_formats(
 375                         format_url, video_id, 'mp4', m3u8_id='hls',
 376                         fatal=False))
 377                 else:
 378                     # HTTP formats are not available when geoblocked is True,
 379                     # other formats are fine though
 380                     if geoblocked:
 381                         continue
 382                     quality = str_or_none(widget.get('_quality'))
 383                     formats.append({
 384                         'format_id': ('http-' + quality) if quality else 'http',
 385                         'url': format_url,
 386                         'preference': 10,  # Plain HTTP, that's nice
 387                     })
 388
 389         if not formats and geoblocked:
 390             self.raise_geo_restricted(
 391                 msg='This video is not available due to geoblocking',
 392                 countries=['DE'])
 393
 394         self._sort_formats(formats)
 395         res.update({
 396             'subtitles': subtitles,
 397             'formats': formats,
 398         })
 399
 400         return res