yt_dlp/extractor/orf.py

   1 import functools
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..networking import HEADRequest
   6 from ..utils import (
   7     clean_html,
   8     determine_ext,
   9     float_or_none,
  10     InAdvancePagedList,
  11     int_or_none,
  12     join_nonempty,
  13     orderedSet,
  14     remove_end,
  15     make_archive_id,
  16     smuggle_url,
  17     strip_jsonp,
  18     try_call,
  19     unescapeHTML,
  20     unified_strdate,
  21     unsmuggle_url,
  22     url_or_none,
  23 )
  24
  25
  26 class ORFTVthekIE(InfoExtractor):
  27     IE_NAME = 'orf:tvthek'
  28     IE_DESC = 'ORF TVthek'
  29     _VALID_URL = r'(?P<url>https?://tvthek\.orf\.at/(?:(?:[^/]+/){2}){1,2}(?P<id>\d+))(/[^/]+/(?P<vid>\d+))?(?:$|[?#])'
  30
  31     _TESTS = [{
  32         'url': 'https://tvthek.orf.at/profile/ZIB-2/1211/ZIB-2/14121079',
  33         'info_dict': {
  34             'id': '14121079',
  35         },
  36         'playlist_count': 11,
  37         'params': {'noplaylist': True}
  38     }, {
  39         'url': 'https://tvthek.orf.at/profile/ZIB-2/1211/ZIB-2/14121079/Umfrage-Welches-Tier-ist-Sebastian-Kurz/15083150',
  40         'info_dict': {
  41             'id': '14121079',
  42         },
  43         'playlist_count': 1,
  44         'params': {'playlist_items': '5'}
  45     }, {
  46         'url': 'https://tvthek.orf.at/profile/ZIB-2/1211/ZIB-2/14121079/Umfrage-Welches-Tier-ist-Sebastian-Kurz/15083150',
  47         'info_dict': {
  48             'id': '14121079',
  49             'playlist_count': 1
  50         },
  51         'playlist': [{
  52             'info_dict': {
  53                 'id': '15083150',
  54                 'ext': 'mp4',
  55                 'description': 'md5:7be1c485425f5f255a5e4e4815e77d04',
  56                 'thumbnail': 'https://api-tvthek.orf.at/uploads/media/segments/0130/59/824271ea35cd8931a0fb08ab316a5b0a1562342c.jpeg',
  57                 'title': 'Umfrage: Welches Tier ist Sebastian Kurz?',
  58             }
  59         }],
  60         'playlist_count': 1,
  61         'params': {'noplaylist': True, 'skip_download': 'm3u8'}
  62     }, {
  63         'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389',
  64         'playlist': [{
  65             'md5': '2942210346ed779588f428a92db88712',
  66             'info_dict': {
  67                 'id': '8896777',
  68                 'ext': 'mp4',
  69                 'title': 'Aufgetischt: Mit der Steirischen Tafelrunde',
  70                 'description': 'md5:c1272f0245537812d4e36419c207b67d',
  71                 'duration': 2668,
  72                 'upload_date': '20141208',
  73             },
  74         }],
  75         'skip': 'Blocked outside of Austria / Germany',
  76     }, {
  77         'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256',
  78         'info_dict': {
  79             'id': '7982259',
  80             'ext': 'mp4',
  81             'title': 'Best of Ingrid Thurnher',
  82             'upload_date': '20140527',
  83             'description': 'Viele Jahre war Ingrid Thurnher das "Gesicht" der ZIB 2. Vor ihrem Wechsel zur ZIB 2 im Jahr 1995 moderierte sie unter anderem "Land und Leute", "Österreich-Bild" und "Niederösterreich heute".',
  84         },
  85         'params': {
  86             'skip_download': True,  # rtsp downloads
  87         },
  88         'skip': 'Blocked outside of Austria / Germany',
  89     }, {
  90         'url': 'http://tvthek.orf.at/topic/Fluechtlingskrise/10463081/Heimat-Fremde-Heimat/13879132/Senioren-betreuen-Migrantenkinder/13879141',
  91         'only_matching': True,
  92     }, {
  93         'url': 'http://tvthek.orf.at/profile/Universum/35429',
  94         'only_matching': True,
  95     }]
  96
  97     def _pagefunc(self, url, data_jsb, n, *, image=None):
  98         sd = data_jsb[n]
  99         video_id, title = str(sd['id']), sd['title']
 100         formats = []
 101         for fd in sd['sources']:
 102             src = url_or_none(fd.get('src'))
 103             if not src:
 104                 continue
 105             format_id = join_nonempty('delivery', 'quality', 'quality_string', from_dict=fd)
 106             ext = determine_ext(src)
 107             if ext == 'm3u8':
 108                 m3u8_formats = self._extract_m3u8_formats(
 109                     src, video_id, 'mp4', m3u8_id=format_id, fatal=False, note=f'Downloading {format_id} m3u8 manifest')
 110                 if any('/geoprotection' in f['url'] for f in m3u8_formats):
 111                     self.raise_geo_restricted()
 112                 formats.extend(m3u8_formats)
 113             elif ext == 'f4m':
 114                 formats.extend(self._extract_f4m_formats(
 115                     src, video_id, f4m_id=format_id, fatal=False))
 116             elif ext == 'mpd':
 117                 formats.extend(self._extract_mpd_formats(
 118                     src, video_id, mpd_id=format_id, fatal=False, note=f'Downloading {format_id} mpd manifest'))
 119             else:
 120                 formats.append({
 121                     'format_id': format_id,
 122                     'url': src,
 123                     'protocol': fd.get('protocol'),
 124                 })
 125
 126         # Check for geoblocking.
 127         # There is a property is_geoprotection, but that's always false
 128         geo_str = sd.get('geoprotection_string')
 129         http_url = next(
 130             (f['url'] for f in formats if re.match(r'^https?://.*\.mp4$', f['url'])),
 131             None) if geo_str else None
 132         if http_url:
 133             self._request_webpage(
 134                 HEADRequest(http_url), video_id, fatal=False, note='Testing for geoblocking',
 135                 errnote=f'This video seems to be blocked outside of {geo_str}. You may want to try the streaming-* formats')
 136
 137         subtitles = {}
 138         for sub in sd.get('subtitles', []):
 139             sub_src = sub.get('src')
 140             if not sub_src:
 141                 continue
 142             subtitles.setdefault(sub.get('lang', 'de-AT'), []).append({
 143                 'url': sub_src,
 144             })
 145
 146         upload_date = unified_strdate(sd.get('created_date'))
 147
 148         thumbnails = []
 149         preview = sd.get('preview_image_url')
 150         if preview:
 151             thumbnails.append({
 152                 'id': 'preview',
 153                 'url': preview,
 154                 'preference': 0,
 155             })
 156         image = sd.get('image_full_url') or image
 157         if image:
 158             thumbnails.append({
 159                 'id': 'full',
 160                 'url': image,
 161                 'preference': 1,
 162             })
 163
 164         yield {
 165             'id': video_id,
 166             'title': title,
 167             'webpage_url': smuggle_url(f'{url}/part/{video_id}', {'force_noplaylist': True}),
 168             'formats': formats,
 169             'subtitles': subtitles,
 170             'description': sd.get('description'),
 171             'duration': int_or_none(sd.get('duration_in_seconds')),
 172             'upload_date': upload_date,
 173             'thumbnails': thumbnails,
 174         }
 175
 176     def _real_extract(self, url):
 177         url, smuggled_data = unsmuggle_url(url)
 178         playlist_id, video_id, base_url = self._match_valid_url(url).group('id', 'vid', 'url')
 179         webpage = self._download_webpage(url, playlist_id)
 180
 181         data_jsb = self._parse_json(
 182             self._search_regex(
 183                 r'<div[^>]+class=(["\']).*?VideoPlaylist.*?\1[^>]+data-jsb=(["\'])(?P<json>.+?)\2',
 184                 webpage, 'playlist', group='json'),
 185             playlist_id, transform_source=unescapeHTML)['playlist']['videos']
 186
 187         if not self._yes_playlist(playlist_id, video_id, smuggled_data):
 188             data_jsb = [sd for sd in data_jsb if str(sd.get('id')) == video_id]
 189
 190         playlist_count = len(data_jsb)
 191         image = self._og_search_thumbnail(webpage) if playlist_count == 1 else None
 192
 193         page_func = functools.partial(self._pagefunc, base_url, data_jsb, image=image)
 194         return {
 195             '_type': 'playlist',
 196             'entries': InAdvancePagedList(page_func, playlist_count, 1),
 197             'id': playlist_id,
 198         }
 199
 200
 201 class ORFRadioIE(InfoExtractor):
 202     IE_NAME = 'orf:radio'
 203
 204     STATION_INFO = {
 205         'fm4': ('fm4', 'fm4', 'orffm4'),
 206         'noe': ('noe', 'oe2n', 'orfnoe'),
 207         'wien': ('wie', 'oe2w', 'orfwie'),
 208         'burgenland': ('bgl', 'oe2b', 'orfbgl'),
 209         'ooe': ('ooe', 'oe2o', 'orfooe'),
 210         'steiermark': ('stm', 'oe2st', 'orfstm'),
 211         'kaernten': ('ktn', 'oe2k', 'orfktn'),
 212         'salzburg': ('sbg', 'oe2s', 'orfsbg'),
 213         'tirol': ('tir', 'oe2t', 'orftir'),
 214         'vorarlberg': ('vbg', 'oe2v', 'orfvbg'),
 215         'oe3': ('oe3', 'oe3', 'orfoe3'),
 216         'oe1': ('oe1', 'oe1', 'orfoe1'),
 217     }
 218     _STATION_RE = '|'.join(map(re.escape, STATION_INFO.keys()))
 219
 220     _VALID_URL = rf'''(?x)
 221         https?://(?:
 222             (?P<station>{_STATION_RE})\.orf\.at/player|
 223             radiothek\.orf\.at/(?P<station2>{_STATION_RE})
 224         )/(?P<date>[0-9]+)/(?P<show>\w+)'''
 225
 226     _TESTS = [{
 227         'url': 'https://radiothek.orf.at/ooe/20220801/OGMO',
 228         'info_dict': {
 229             'id': 'OGMO',
 230             'title': 'Guten Morgen OÖ',
 231             'description': 'md5:a3f6083399ef92b8cbe2d421b180835a',
 232         },
 233         'playlist': [{
 234             'md5': 'f33147d954a326e338ea52572c2810e8',
 235             'info_dict': {
 236                 'id': '2022-08-01_0459_tl_66_7DaysMon1_319062',
 237                 'ext': 'mp3',
 238                 'title': 'Guten Morgen OÖ',
 239                 'upload_date': '20220801',
 240                 'duration': 18000,
 241                 'timestamp': 1659322789,
 242                 'description': 'md5:a3f6083399ef92b8cbe2d421b180835a',
 243             }
 244         }]
 245     }, {
 246         'url': 'https://ooe.orf.at/player/20220801/OGMO',
 247         'info_dict': {
 248             'id': 'OGMO',
 249             'title': 'Guten Morgen OÖ',
 250             'description': 'md5:a3f6083399ef92b8cbe2d421b180835a',
 251         },
 252         'playlist': [{
 253             'md5': 'f33147d954a326e338ea52572c2810e8',
 254             'info_dict': {
 255                 'id': '2022-08-01_0459_tl_66_7DaysMon1_319062',
 256                 'ext': 'mp3',
 257                 'title': 'Guten Morgen OÖ',
 258                 'upload_date': '20220801',
 259                 'duration': 18000,
 260                 'timestamp': 1659322789,
 261                 'description': 'md5:a3f6083399ef92b8cbe2d421b180835a',
 262             }
 263         }]
 264     }, {
 265         'url': 'http://fm4.orf.at/player/20170107/4CC',
 266         'only_matching': True,
 267     }, {
 268         'url': 'https://noe.orf.at/player/20200423/NGM',
 269         'only_matching': True,
 270     }, {
 271         'url': 'https://wien.orf.at/player/20200423/WGUM',
 272         'only_matching': True,
 273     }, {
 274         'url': 'https://burgenland.orf.at/player/20200423/BGM',
 275         'only_matching': True,
 276     }, {
 277         'url': 'https://steiermark.orf.at/player/20200423/STGMS',
 278         'only_matching': True,
 279     }, {
 280         'url': 'https://kaernten.orf.at/player/20200423/KGUMO',
 281         'only_matching': True,
 282     }, {
 283         'url': 'https://salzburg.orf.at/player/20200423/SGUM',
 284         'only_matching': True,
 285     }, {
 286         'url': 'https://tirol.orf.at/player/20200423/TGUMO',
 287         'only_matching': True,
 288     }, {
 289         'url': 'https://vorarlberg.orf.at/player/20200423/VGUM',
 290         'only_matching': True,
 291     }, {
 292         'url': 'https://oe3.orf.at/player/20200424/3WEK',
 293         'only_matching': True,
 294     }, {
 295         'url': 'http://oe1.orf.at/player/20170108/456544',
 296         'md5': '34d8a6e67ea888293741c86a099b745b',
 297         'info_dict': {
 298             'id': '2017-01-08_0759_tl_51_7DaysSun6_256141',
 299             'ext': 'mp3',
 300             'title': 'Morgenjournal',
 301             'duration': 609,
 302             'timestamp': 1483858796,
 303             'upload_date': '20170108',
 304         },
 305         'skip': 'Shows from ORF radios are only available for 7 days.'
 306     }]
 307
 308     def _entries(self, data, station):
 309         _, loop_station, old_ie = self.STATION_INFO[station]
 310         for info in data['streams']:
 311             item_id = info.get('loopStreamId')
 312             if not item_id:
 313                 continue
 314             video_id = item_id.replace('.mp3', '')
 315             yield {
 316                 'id': video_id,
 317                 'ext': 'mp3',
 318                 'url': f'https://loopstream01.apa.at/?channel={loop_station}&id={item_id}',
 319                 '_old_archive_ids': [make_archive_id(old_ie, video_id)],
 320                 'title': data.get('title'),
 321                 'description': clean_html(data.get('subtitle')),
 322                 'duration': try_call(lambda: (info['end'] - info['start']) / 1000),
 323                 'timestamp': int_or_none(info.get('start'), scale=1000),
 324                 'series': data.get('programTitle'),
 325             }
 326
 327     def _real_extract(self, url):
 328         station, station2, show_date, show_id = self._match_valid_url(url).group('station', 'station2', 'date', 'show')
 329         api_station, _, _ = self.STATION_INFO[station or station2]
 330         data = self._download_json(
 331             f'http://audioapi.orf.at/{api_station}/api/json/current/broadcast/{show_id}/{show_date}', show_id)
 332
 333         return self.playlist_result(
 334             self._entries(data, station or station2), show_id, data.get('title'), clean_html(data.get('subtitle')))
 335
 336
 337 class ORFIPTVIE(InfoExtractor):
 338     IE_NAME = 'orf:iptv'
 339     IE_DESC = 'iptv.ORF.at'
 340     _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
 341
 342     _TEST = {
 343         'url': 'http://iptv.orf.at/stories/2275236/',
 344         'md5': 'c8b22af4718a4b4af58342529453e3e5',
 345         'info_dict': {
 346             'id': '350612',
 347             'ext': 'flv',
 348             'title': 'Weitere Evakuierungen um Vulkan Calbuco',
 349             'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
 350             'duration': 68.197,
 351             'thumbnail': r're:^https?://.*\.jpg$',
 352             'upload_date': '20150425',
 353         },
 354     }
 355
 356     def _real_extract(self, url):
 357         story_id = self._match_id(url)
 358
 359         webpage = self._download_webpage(
 360             'http://iptv.orf.at/stories/%s' % story_id, story_id)
 361
 362         video_id = self._search_regex(
 363             r'data-video(?:id)?="(\d+)"', webpage, 'video id')
 364
 365         data = self._download_json(
 366             'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 367             video_id)[0]
 368
 369         duration = float_or_none(data['duration'], 1000)
 370
 371         video = data['sources']['default']
 372         load_balancer_url = video['loadBalancerUrl']
 373         abr = int_or_none(video.get('audioBitrate'))
 374         vbr = int_or_none(video.get('bitrate'))
 375         fps = int_or_none(video.get('videoFps'))
 376         width = int_or_none(video.get('videoWidth'))
 377         height = int_or_none(video.get('videoHeight'))
 378         thumbnail = video.get('preview')
 379
 380         rendition = self._download_json(
 381             load_balancer_url, video_id, transform_source=strip_jsonp)
 382
 383         f = {
 384             'abr': abr,
 385             'vbr': vbr,
 386             'fps': fps,
 387             'width': width,
 388             'height': height,
 389         }
 390
 391         formats = []
 392         for format_id, format_url in rendition['redirect'].items():
 393             if format_id == 'rtmp':
 394                 ff = f.copy()
 395                 ff.update({
 396                     'url': format_url,
 397                     'format_id': format_id,
 398                 })
 399                 formats.append(ff)
 400             elif determine_ext(format_url) == 'f4m':
 401                 formats.extend(self._extract_f4m_formats(
 402                     format_url, video_id, f4m_id=format_id))
 403             elif determine_ext(format_url) == 'm3u8':
 404                 formats.extend(self._extract_m3u8_formats(
 405                     format_url, video_id, 'mp4', m3u8_id=format_id))
 406             else:
 407                 continue
 408
 409         title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
 410         description = self._og_search_description(webpage)
 411         upload_date = unified_strdate(self._html_search_meta(
 412             'dc.date', webpage, 'upload date'))
 413
 414         return {
 415             'id': video_id,
 416             'title': title,
 417             'description': description,
 418             'duration': duration,
 419             'thumbnail': thumbnail,
 420             'upload_date': upload_date,
 421             'formats': formats,
 422         }
 423
 424
 425 class ORFFM4StoryIE(InfoExtractor):
 426     IE_NAME = 'orf:fm4:story'
 427     IE_DESC = 'fm4.orf.at stories'
 428     _VALID_URL = r'https?://fm4\.orf\.at/stories/(?P<id>\d+)'
 429
 430     _TEST = {
 431         'url': 'http://fm4.orf.at/stories/2865738/',
 432         'playlist': [{
 433             'md5': 'e1c2c706c45c7b34cf478bbf409907ca',
 434             'info_dict': {
 435                 'id': '547792',
 436                 'ext': 'flv',
 437                 'title': 'Manu Delago und Inner Tongue live',
 438                 'description': 'Manu Delago und Inner Tongue haben bei der FM4 Soundpark Session live alles gegeben. Hier gibt es Fotos und die gesamte Session als Video.',
 439                 'duration': 1748.52,
 440                 'thumbnail': r're:^https?://.*\.jpg$',
 441                 'upload_date': '20170913',
 442             },
 443         }, {
 444             'md5': 'c6dd2179731f86f4f55a7b49899d515f',
 445             'info_dict': {
 446                 'id': '547798',
 447                 'ext': 'flv',
 448                 'title': 'Manu Delago und Inner Tongue live (2)',
 449                 'duration': 1504.08,
 450                 'thumbnail': r're:^https?://.*\.jpg$',
 451                 'upload_date': '20170913',
 452                 'description': 'Manu Delago und Inner Tongue haben bei der FM4 Soundpark Session live alles gegeben. Hier gibt es Fotos und die gesamte Session als Video.',
 453             },
 454         }],
 455     }
 456
 457     def _real_extract(self, url):
 458         story_id = self._match_id(url)
 459         webpage = self._download_webpage(url, story_id)
 460
 461         entries = []
 462         all_ids = orderedSet(re.findall(r'data-video(?:id)?="(\d+)"', webpage))
 463         for idx, video_id in enumerate(all_ids):
 464             data = self._download_json(
 465                 'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 466                 video_id)[0]
 467
 468             duration = float_or_none(data['duration'], 1000)
 469
 470             video = data['sources']['q8c']
 471             load_balancer_url = video['loadBalancerUrl']
 472             abr = int_or_none(video.get('audioBitrate'))
 473             vbr = int_or_none(video.get('bitrate'))
 474             fps = int_or_none(video.get('videoFps'))
 475             width = int_or_none(video.get('videoWidth'))
 476             height = int_or_none(video.get('videoHeight'))
 477             thumbnail = video.get('preview')
 478
 479             rendition = self._download_json(
 480                 load_balancer_url, video_id, transform_source=strip_jsonp)
 481
 482             f = {
 483                 'abr': abr,
 484                 'vbr': vbr,
 485                 'fps': fps,
 486                 'width': width,
 487                 'height': height,
 488             }
 489
 490             formats = []
 491             for format_id, format_url in rendition['redirect'].items():
 492                 if format_id == 'rtmp':
 493                     ff = f.copy()
 494                     ff.update({
 495                         'url': format_url,
 496                         'format_id': format_id,
 497                     })
 498                     formats.append(ff)
 499                 elif determine_ext(format_url) == 'f4m':
 500                     formats.extend(self._extract_f4m_formats(
 501                         format_url, video_id, f4m_id=format_id))
 502                 elif determine_ext(format_url) == 'm3u8':
 503                     formats.extend(self._extract_m3u8_formats(
 504                         format_url, video_id, 'mp4', m3u8_id=format_id))
 505                 else:
 506                     continue
 507
 508             title = remove_end(self._og_search_title(webpage), ' - fm4.ORF.at')
 509             if idx >= 1:
 510                 # Titles are duplicates, make them unique
 511                 title += ' (' + str(idx + 1) + ')'
 512             description = self._og_search_description(webpage)
 513             upload_date = unified_strdate(self._html_search_meta(
 514                 'dc.date', webpage, 'upload date'))
 515
 516             entries.append({
 517                 'id': video_id,
 518                 'title': title,
 519                 'description': description,
 520                 'duration': duration,
 521                 'thumbnail': thumbnail,
 522                 'upload_date': upload_date,
 523                 'formats': formats,
 524             })
 525
 526         return self.playlist_result(entries)