_ Git - youtube-dl/blob - youtube_dl/extractor/orf.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 compat_str
   8 from ..utils import (
   9     clean_html,
  10     determine_ext,
  11     float_or_none,
  12     HEADRequest,
  13     int_or_none,
  14     orderedSet,
  15     remove_end,
  16     str_or_none,
  17     strip_jsonp,
  18     unescapeHTML,
  19     unified_strdate,
  20     url_or_none,
  21 )
  22
  23
  24 class ORFTVthekIE(InfoExtractor):
  25     IE_NAME = 'orf:tvthek'
  26     IE_DESC = 'ORF TVthek'
  27     _VALID_URL = r'https?://tvthek\.orf\.at/(?:[^/]+/)+(?P<id>\d+)'
  28
  29     _TESTS = [{
  30         'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389',
  31         'playlist': [{
  32             'md5': '2942210346ed779588f428a92db88712',
  33             'info_dict': {
  34                 'id': '8896777',
  35                 'ext': 'mp4',
  36                 'title': 'Aufgetischt: Mit der Steirischen Tafelrunde',
  37                 'description': 'md5:c1272f0245537812d4e36419c207b67d',
  38                 'duration': 2668,
  39                 'upload_date': '20141208',
  40             },
  41         }],
  42         'skip': 'Blocked outside of Austria / Germany',
  43     }, {
  44         'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256',
  45         'info_dict': {
  46             'id': '7982259',
  47             'ext': 'mp4',
  48             'title': 'Best of Ingrid Thurnher',
  49             'upload_date': '20140527',
  50             '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".',
  51         },
  52         'params': {
  53             'skip_download': True,  # rtsp downloads
  54         },
  55         'skip': 'Blocked outside of Austria / Germany',
  56     }, {
  57         'url': 'http://tvthek.orf.at/topic/Fluechtlingskrise/10463081/Heimat-Fremde-Heimat/13879132/Senioren-betreuen-Migrantenkinder/13879141',
  58         'only_matching': True,
  59     }, {
  60         'url': 'http://tvthek.orf.at/profile/Universum/35429',
  61         'only_matching': True,
  62     }]
  63
  64     def _real_extract(self, url):
  65         playlist_id = self._match_id(url)
  66         webpage = self._download_webpage(url, playlist_id)
  67
  68         data_jsb = self._parse_json(
  69             self._search_regex(
  70                 r'<div[^>]+class=(["\']).*?VideoPlaylist.*?\1[^>]+data-jsb=(["\'])(?P<json>.+?)\2',
  71                 webpage, 'playlist', group='json'),
  72             playlist_id, transform_source=unescapeHTML)['playlist']['videos']
  73
  74         entries = []
  75         for sd in data_jsb:
  76             video_id, title = sd.get('id'), sd.get('title')
  77             if not video_id or not title:
  78                 continue
  79             video_id = compat_str(video_id)
  80             formats = []
  81             for fd in sd['sources']:
  82                 src = url_or_none(fd.get('src'))
  83                 if not src:
  84                     continue
  85                 format_id_list = []
  86                 for key in ('delivery', 'quality', 'quality_string'):
  87                     value = fd.get(key)
  88                     if value:
  89                         format_id_list.append(value)
  90                 format_id = '-'.join(format_id_list)
  91                 ext = determine_ext(src)
  92                 if ext == 'm3u8':
  93                     formats.extend(self._extract_m3u8_formats(
  94                         src, video_id, 'mp4', m3u8_id=format_id, fatal=False))
  95                 elif ext == 'f4m':
  96                     formats.extend(self._extract_f4m_formats(
  97                         src, video_id, f4m_id=format_id, fatal=False))
  98                 else:
  99                     formats.append({
 100                         'format_id': format_id,
 101                         'url': src,
 102                         'protocol': fd.get('protocol'),
 103                     })
 104
 105             # Check for geoblocking.
 106             # There is a property is_geoprotection, but that's always false
 107             geo_str = sd.get('geoprotection_string')
 108             if geo_str:
 109                 try:
 110                     http_url = next(
 111                         f['url']
 112                         for f in formats
 113                         if re.match(r'^https?://.*\.mp4$', f['url']))
 114                 except StopIteration:
 115                     pass
 116                 else:
 117                     req = HEADRequest(http_url)
 118                     self._request_webpage(
 119                         req, video_id,
 120                         note='Testing for geoblocking',
 121                         errnote=((
 122                             'This video seems to be blocked outside of %s. '
 123                             'You may want to try the streaming-* formats.')
 124                             % geo_str),
 125                         fatal=False)
 126
 127             self._check_formats(formats, video_id)
 128             self._sort_formats(formats)
 129
 130             subtitles = {}
 131             for sub in sd.get('subtitles', []):
 132                 sub_src = sub.get('src')
 133                 if not sub_src:
 134                     continue
 135                 subtitles.setdefault(sub.get('lang', 'de-AT'), []).append({
 136                     'url': sub_src,
 137                 })
 138
 139             upload_date = unified_strdate(sd.get('created_date'))
 140             entries.append({
 141                 '_type': 'video',
 142                 'id': video_id,
 143                 'title': title,
 144                 'formats': formats,
 145                 'subtitles': subtitles,
 146                 'description': sd.get('description'),
 147                 'duration': int_or_none(sd.get('duration_in_seconds')),
 148                 'upload_date': upload_date,
 149                 'thumbnail': sd.get('image_full_url'),
 150             })
 151
 152         return {
 153             '_type': 'playlist',
 154             'entries': entries,
 155             'id': playlist_id,
 156         }
 157
 158
 159 class ORFRadioIE(InfoExtractor):
 160     def _real_extract(self, url):
 161         mobj = re.match(self._VALID_URL, url)
 162         station = mobj.group('station')
 163         show_date = mobj.group('date')
 164         show_id = mobj.group('show')
 165
 166         data = self._download_json(
 167             'http://audioapi.orf.at/%s/api/json/current/broadcast/%s/%s'
 168             % (station, show_id, show_date), show_id)
 169
 170         entries = []
 171         for info in data['streams']:
 172             loop_stream_id = str_or_none(info.get('loopStreamId'))
 173             if not loop_stream_id:
 174                 continue
 175             title = str_or_none(data.get('title'))
 176             if not title:
 177                 continue
 178             start = int_or_none(info.get('start'), scale=1000)
 179             end = int_or_none(info.get('end'), scale=1000)
 180             duration = end - start if end and start else None
 181             entries.append({
 182                 'id': loop_stream_id.replace('.mp3', ''),
 183                 'url': 'http://loopstream01.apa.at/?channel=%s&id=%s' % (station, loop_stream_id),
 184                 'title': title,
 185                 'description': clean_html(data.get('subtitle')),
 186                 'duration': duration,
 187                 'timestamp': start,
 188                 'ext': 'mp3',
 189                 'series': data.get('programTitle'),
 190             })
 191
 192         return {
 193             '_type': 'playlist',
 194             'id': show_id,
 195             'title': data.get('title'),
 196             'description': clean_html(data.get('subtitle')),
 197             'entries': entries,
 198         }
 199
 200
 201 class ORFFM4IE(ORFRadioIE):
 202     IE_NAME = 'orf:fm4'
 203     IE_DESC = 'radio FM4'
 204     _VALID_URL = r'https?://(?P<station>fm4)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>4\w+)'
 205
 206     _TEST = {
 207         'url': 'http://fm4.orf.at/player/20170107/4CC',
 208         'md5': '2b0be47375432a7ef104453432a19212',
 209         'info_dict': {
 210             'id': '2017-01-07_2100_tl_54_7DaysSat18_31295',
 211             'ext': 'mp3',
 212             'title': 'Solid Steel Radioshow',
 213             'description': 'Die Mixshow von Coldcut und Ninja Tune.',
 214             'duration': 3599,
 215             'timestamp': 1483819257,
 216             'upload_date': '20170107',
 217         },
 218         'skip': 'Shows from ORF radios are only available for 7 days.',
 219         'only_matching': True,
 220     }
 221
 222
 223 class ORFOE1IE(ORFRadioIE):
 224     IE_NAME = 'orf:oe1'
 225     IE_DESC = 'Radio Österreich 1'
 226     _VALID_URL = r'https?://(?P<station>oe1)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
 227
 228     _TEST = {
 229         'url': 'http://oe1.orf.at/player/20170108/456544',
 230         'md5': '34d8a6e67ea888293741c86a099b745b',
 231         'info_dict': {
 232             'id': '2017-01-08_0759_tl_51_7DaysSun6_256141',
 233             'ext': 'mp3',
 234             'title': 'Morgenjournal',
 235             'duration': 609,
 236             'timestamp': 1483858796,
 237             'upload_date': '20170108',
 238         },
 239         'skip': 'Shows from ORF radios are only available for 7 days.'
 240     }
 241
 242
 243 class ORFIPTVIE(InfoExtractor):
 244     IE_NAME = 'orf:iptv'
 245     IE_DESC = 'iptv.ORF.at'
 246     _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
 247
 248     _TEST = {
 249         'url': 'http://iptv.orf.at/stories/2275236/',
 250         'md5': 'c8b22af4718a4b4af58342529453e3e5',
 251         'info_dict': {
 252             'id': '350612',
 253             'ext': 'flv',
 254             'title': 'Weitere Evakuierungen um Vulkan Calbuco',
 255             'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
 256             'duration': 68.197,
 257             'thumbnail': r're:^https?://.*\.jpg$',
 258             'upload_date': '20150425',
 259         },
 260     }
 261
 262     def _real_extract(self, url):
 263         story_id = self._match_id(url)
 264
 265         webpage = self._download_webpage(
 266             'http://iptv.orf.at/stories/%s' % story_id, story_id)
 267
 268         video_id = self._search_regex(
 269             r'data-video(?:id)?="(\d+)"', webpage, 'video id')
 270
 271         data = self._download_json(
 272             'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 273             video_id)[0]
 274
 275         duration = float_or_none(data['duration'], 1000)
 276
 277         video = data['sources']['default']
 278         load_balancer_url = video['loadBalancerUrl']
 279         abr = int_or_none(video.get('audioBitrate'))
 280         vbr = int_or_none(video.get('bitrate'))
 281         fps = int_or_none(video.get('videoFps'))
 282         width = int_or_none(video.get('videoWidth'))
 283         height = int_or_none(video.get('videoHeight'))
 284         thumbnail = video.get('preview')
 285
 286         rendition = self._download_json(
 287             load_balancer_url, video_id, transform_source=strip_jsonp)
 288
 289         f = {
 290             'abr': abr,
 291             'vbr': vbr,
 292             'fps': fps,
 293             'width': width,
 294             'height': height,
 295         }
 296
 297         formats = []
 298         for format_id, format_url in rendition['redirect'].items():
 299             if format_id == 'rtmp':
 300                 ff = f.copy()
 301                 ff.update({
 302                     'url': format_url,
 303                     'format_id': format_id,
 304                 })
 305                 formats.append(ff)
 306             elif determine_ext(format_url) == 'f4m':
 307                 formats.extend(self._extract_f4m_formats(
 308                     format_url, video_id, f4m_id=format_id))
 309             elif determine_ext(format_url) == 'm3u8':
 310                 formats.extend(self._extract_m3u8_formats(
 311                     format_url, video_id, 'mp4', m3u8_id=format_id))
 312             else:
 313                 continue
 314         self._sort_formats(formats)
 315
 316         title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
 317         description = self._og_search_description(webpage)
 318         upload_date = unified_strdate(self._html_search_meta(
 319             'dc.date', webpage, 'upload date'))
 320
 321         return {
 322             'id': video_id,
 323             'title': title,
 324             'description': description,
 325             'duration': duration,
 326             'thumbnail': thumbnail,
 327             'upload_date': upload_date,
 328             'formats': formats,
 329         }
 330
 331
 332 class ORFFM4StoryIE(InfoExtractor):
 333     IE_NAME = 'orf:fm4:story'
 334     IE_DESC = 'fm4.orf.at stories'
 335     _VALID_URL = r'https?://fm4\.orf\.at/stories/(?P<id>\d+)'
 336
 337     _TEST = {
 338         'url': 'http://fm4.orf.at/stories/2865738/',
 339         'playlist': [{
 340             'md5': 'e1c2c706c45c7b34cf478bbf409907ca',
 341             'info_dict': {
 342                 'id': '547792',
 343                 'ext': 'flv',
 344                 'title': 'Manu Delago und Inner Tongue live',
 345                 '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.',
 346                 'duration': 1748.52,
 347                 'thumbnail': r're:^https?://.*\.jpg$',
 348                 'upload_date': '20170913',
 349             },
 350         }, {
 351             'md5': 'c6dd2179731f86f4f55a7b49899d515f',
 352             'info_dict': {
 353                 'id': '547798',
 354                 'ext': 'flv',
 355                 'title': 'Manu Delago und Inner Tongue live (2)',
 356                 'duration': 1504.08,
 357                 'thumbnail': r're:^https?://.*\.jpg$',
 358                 'upload_date': '20170913',
 359                 '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.',
 360             },
 361         }],
 362     }
 363
 364     def _real_extract(self, url):
 365         story_id = self._match_id(url)
 366         webpage = self._download_webpage(url, story_id)
 367
 368         entries = []
 369         all_ids = orderedSet(re.findall(r'data-video(?:id)?="(\d+)"', webpage))
 370         for idx, video_id in enumerate(all_ids):
 371             data = self._download_json(
 372                 'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 373                 video_id)[0]
 374
 375             duration = float_or_none(data['duration'], 1000)
 376
 377             video = data['sources']['q8c']
 378             load_balancer_url = video['loadBalancerUrl']
 379             abr = int_or_none(video.get('audioBitrate'))
 380             vbr = int_or_none(video.get('bitrate'))
 381             fps = int_or_none(video.get('videoFps'))
 382             width = int_or_none(video.get('videoWidth'))
 383             height = int_or_none(video.get('videoHeight'))
 384             thumbnail = video.get('preview')
 385
 386             rendition = self._download_json(
 387                 load_balancer_url, video_id, transform_source=strip_jsonp)
 388
 389             f = {
 390                 'abr': abr,
 391                 'vbr': vbr,
 392                 'fps': fps,
 393                 'width': width,
 394                 'height': height,
 395             }
 396
 397             formats = []
 398             for format_id, format_url in rendition['redirect'].items():
 399                 if format_id == 'rtmp':
 400                     ff = f.copy()
 401                     ff.update({
 402                         'url': format_url,
 403                         'format_id': format_id,
 404                     })
 405                     formats.append(ff)
 406                 elif determine_ext(format_url) == 'f4m':
 407                     formats.extend(self._extract_f4m_formats(
 408                         format_url, video_id, f4m_id=format_id))
 409                 elif determine_ext(format_url) == 'm3u8':
 410                     formats.extend(self._extract_m3u8_formats(
 411                         format_url, video_id, 'mp4', m3u8_id=format_id))
 412                 else:
 413                     continue
 414             self._sort_formats(formats)
 415
 416             title = remove_end(self._og_search_title(webpage), ' - fm4.ORF.at')
 417             if idx >= 1:
 418                 # Titles are duplicates, make them unique
 419                 title += ' (' + str(idx + 1) + ')'
 420             description = self._og_search_description(webpage)
 421             upload_date = unified_strdate(self._html_search_meta(
 422                 'dc.date', webpage, 'upload date'))
 423
 424             entries.append({
 425                 'id': video_id,
 426                 'title': title,
 427                 'description': description,
 428                 'duration': duration,
 429                 'thumbnail': thumbnail,
 430                 'upload_date': upload_date,
 431                 'formats': formats,
 432             })
 433
 434         return self.playlist_result(entries)