_ 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                     m3u8_formats = self._extract_m3u8_formats(
  94                         src, video_id, 'mp4', m3u8_id=format_id, fatal=False)
  95                     if any('/geoprotection' in f['url'] for f in m3u8_formats):
  96                         self.raise_geo_restricted()
  97                     formats.extend(m3u8_formats)
  98                 elif ext == 'f4m':
  99                     formats.extend(self._extract_f4m_formats(
 100                         src, video_id, f4m_id=format_id, fatal=False))
 101                 else:
 102                     formats.append({
 103                         'format_id': format_id,
 104                         'url': src,
 105                         'protocol': fd.get('protocol'),
 106                     })
 107
 108             # Check for geoblocking.
 109             # There is a property is_geoprotection, but that's always false
 110             geo_str = sd.get('geoprotection_string')
 111             if geo_str:
 112                 try:
 113                     http_url = next(
 114                         f['url']
 115                         for f in formats
 116                         if re.match(r'^https?://.*\.mp4$', f['url']))
 117                 except StopIteration:
 118                     pass
 119                 else:
 120                     req = HEADRequest(http_url)
 121                     self._request_webpage(
 122                         req, video_id,
 123                         note='Testing for geoblocking',
 124                         errnote=((
 125                             'This video seems to be blocked outside of %s. '
 126                             'You may want to try the streaming-* formats.')
 127                             % geo_str),
 128                         fatal=False)
 129
 130             self._check_formats(formats, video_id)
 131             self._sort_formats(formats)
 132
 133             subtitles = {}
 134             for sub in sd.get('subtitles', []):
 135                 sub_src = sub.get('src')
 136                 if not sub_src:
 137                     continue
 138                 subtitles.setdefault(sub.get('lang', 'de-AT'), []).append({
 139                     'url': sub_src,
 140                 })
 141
 142             upload_date = unified_strdate(sd.get('created_date'))
 143             entries.append({
 144                 '_type': 'video',
 145                 'id': video_id,
 146                 'title': title,
 147                 'formats': formats,
 148                 'subtitles': subtitles,
 149                 'description': sd.get('description'),
 150                 'duration': int_or_none(sd.get('duration_in_seconds')),
 151                 'upload_date': upload_date,
 152                 'thumbnail': sd.get('image_full_url'),
 153             })
 154
 155         return {
 156             '_type': 'playlist',
 157             'entries': entries,
 158             'id': playlist_id,
 159         }
 160
 161
 162 class ORFRadioIE(InfoExtractor):
 163     def _real_extract(self, url):
 164         mobj = re.match(self._VALID_URL, url)
 165         station = mobj.group('station')
 166         show_date = mobj.group('date')
 167         show_id = mobj.group('show')
 168
 169         data = self._download_json(
 170             'http://audioapi.orf.at/%s/api/json/current/broadcast/%s/%s'
 171             % (station, show_id, show_date), show_id)
 172
 173         entries = []
 174         for info in data['streams']:
 175             loop_stream_id = str_or_none(info.get('loopStreamId'))
 176             if not loop_stream_id:
 177                 continue
 178             title = str_or_none(data.get('title'))
 179             if not title:
 180                 continue
 181             start = int_or_none(info.get('start'), scale=1000)
 182             end = int_or_none(info.get('end'), scale=1000)
 183             duration = end - start if end and start else None
 184             entries.append({
 185                 'id': loop_stream_id.replace('.mp3', ''),
 186                 'url': 'http://loopstream01.apa.at/?channel=%s&id=%s' % (station, loop_stream_id),
 187                 'title': title,
 188                 'description': clean_html(data.get('subtitle')),
 189                 'duration': duration,
 190                 'timestamp': start,
 191                 'ext': 'mp3',
 192                 'series': data.get('programTitle'),
 193             })
 194
 195         return {
 196             '_type': 'playlist',
 197             'id': show_id,
 198             'title': data.get('title'),
 199             'description': clean_html(data.get('subtitle')),
 200             'entries': entries,
 201         }
 202
 203
 204 class ORFFM4IE(ORFRadioIE):
 205     IE_NAME = 'orf:fm4'
 206     IE_DESC = 'radio FM4'
 207     _VALID_URL = r'https?://(?P<station>fm4)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>4\w+)'
 208
 209     _TEST = {
 210         'url': 'http://fm4.orf.at/player/20170107/4CC',
 211         'md5': '2b0be47375432a7ef104453432a19212',
 212         'info_dict': {
 213             'id': '2017-01-07_2100_tl_54_7DaysSat18_31295',
 214             'ext': 'mp3',
 215             'title': 'Solid Steel Radioshow',
 216             'description': 'Die Mixshow von Coldcut und Ninja Tune.',
 217             'duration': 3599,
 218             'timestamp': 1483819257,
 219             'upload_date': '20170107',
 220         },
 221         'skip': 'Shows from ORF radios are only available for 7 days.',
 222         'only_matching': True,
 223     }
 224
 225
 226 class ORFOE1IE(ORFRadioIE):
 227     IE_NAME = 'orf:oe1'
 228     IE_DESC = 'Radio Österreich 1'
 229     _VALID_URL = r'https?://(?P<station>oe1)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
 230
 231     _TEST = {
 232         'url': 'http://oe1.orf.at/player/20170108/456544',
 233         'md5': '34d8a6e67ea888293741c86a099b745b',
 234         'info_dict': {
 235             'id': '2017-01-08_0759_tl_51_7DaysSun6_256141',
 236             'ext': 'mp3',
 237             'title': 'Morgenjournal',
 238             'duration': 609,
 239             'timestamp': 1483858796,
 240             'upload_date': '20170108',
 241         },
 242         'skip': 'Shows from ORF radios are only available for 7 days.'
 243     }
 244
 245
 246 class ORFIPTVIE(InfoExtractor):
 247     IE_NAME = 'orf:iptv'
 248     IE_DESC = 'iptv.ORF.at'
 249     _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
 250
 251     _TEST = {
 252         'url': 'http://iptv.orf.at/stories/2275236/',
 253         'md5': 'c8b22af4718a4b4af58342529453e3e5',
 254         'info_dict': {
 255             'id': '350612',
 256             'ext': 'flv',
 257             'title': 'Weitere Evakuierungen um Vulkan Calbuco',
 258             'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
 259             'duration': 68.197,
 260             'thumbnail': r're:^https?://.*\.jpg$',
 261             'upload_date': '20150425',
 262         },
 263     }
 264
 265     def _real_extract(self, url):
 266         story_id = self._match_id(url)
 267
 268         webpage = self._download_webpage(
 269             'http://iptv.orf.at/stories/%s' % story_id, story_id)
 270
 271         video_id = self._search_regex(
 272             r'data-video(?:id)?="(\d+)"', webpage, 'video id')
 273
 274         data = self._download_json(
 275             'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 276             video_id)[0]
 277
 278         duration = float_or_none(data['duration'], 1000)
 279
 280         video = data['sources']['default']
 281         load_balancer_url = video['loadBalancerUrl']
 282         abr = int_or_none(video.get('audioBitrate'))
 283         vbr = int_or_none(video.get('bitrate'))
 284         fps = int_or_none(video.get('videoFps'))
 285         width = int_or_none(video.get('videoWidth'))
 286         height = int_or_none(video.get('videoHeight'))
 287         thumbnail = video.get('preview')
 288
 289         rendition = self._download_json(
 290             load_balancer_url, video_id, transform_source=strip_jsonp)
 291
 292         f = {
 293             'abr': abr,
 294             'vbr': vbr,
 295             'fps': fps,
 296             'width': width,
 297             'height': height,
 298         }
 299
 300         formats = []
 301         for format_id, format_url in rendition['redirect'].items():
 302             if format_id == 'rtmp':
 303                 ff = f.copy()
 304                 ff.update({
 305                     'url': format_url,
 306                     'format_id': format_id,
 307                 })
 308                 formats.append(ff)
 309             elif determine_ext(format_url) == 'f4m':
 310                 formats.extend(self._extract_f4m_formats(
 311                     format_url, video_id, f4m_id=format_id))
 312             elif determine_ext(format_url) == 'm3u8':
 313                 formats.extend(self._extract_m3u8_formats(
 314                     format_url, video_id, 'mp4', m3u8_id=format_id))
 315             else:
 316                 continue
 317         self._sort_formats(formats)
 318
 319         title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
 320         description = self._og_search_description(webpage)
 321         upload_date = unified_strdate(self._html_search_meta(
 322             'dc.date', webpage, 'upload date'))
 323
 324         return {
 325             'id': video_id,
 326             'title': title,
 327             'description': description,
 328             'duration': duration,
 329             'thumbnail': thumbnail,
 330             'upload_date': upload_date,
 331             'formats': formats,
 332         }
 333
 334
 335 class ORFFM4StoryIE(InfoExtractor):
 336     IE_NAME = 'orf:fm4:story'
 337     IE_DESC = 'fm4.orf.at stories'
 338     _VALID_URL = r'https?://fm4\.orf\.at/stories/(?P<id>\d+)'
 339
 340     _TEST = {
 341         'url': 'http://fm4.orf.at/stories/2865738/',
 342         'playlist': [{
 343             'md5': 'e1c2c706c45c7b34cf478bbf409907ca',
 344             'info_dict': {
 345                 'id': '547792',
 346                 'ext': 'flv',
 347                 'title': 'Manu Delago und Inner Tongue live',
 348                 '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.',
 349                 'duration': 1748.52,
 350                 'thumbnail': r're:^https?://.*\.jpg$',
 351                 'upload_date': '20170913',
 352             },
 353         }, {
 354             'md5': 'c6dd2179731f86f4f55a7b49899d515f',
 355             'info_dict': {
 356                 'id': '547798',
 357                 'ext': 'flv',
 358                 'title': 'Manu Delago und Inner Tongue live (2)',
 359                 'duration': 1504.08,
 360                 'thumbnail': r're:^https?://.*\.jpg$',
 361                 'upload_date': '20170913',
 362                 '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.',
 363             },
 364         }],
 365     }
 366
 367     def _real_extract(self, url):
 368         story_id = self._match_id(url)
 369         webpage = self._download_webpage(url, story_id)
 370
 371         entries = []
 372         all_ids = orderedSet(re.findall(r'data-video(?:id)?="(\d+)"', webpage))
 373         for idx, video_id in enumerate(all_ids):
 374             data = self._download_json(
 375                 'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
 376                 video_id)[0]
 377
 378             duration = float_or_none(data['duration'], 1000)
 379
 380             video = data['sources']['q8c']
 381             load_balancer_url = video['loadBalancerUrl']
 382             abr = int_or_none(video.get('audioBitrate'))
 383             vbr = int_or_none(video.get('bitrate'))
 384             fps = int_or_none(video.get('videoFps'))
 385             width = int_or_none(video.get('videoWidth'))
 386             height = int_or_none(video.get('videoHeight'))
 387             thumbnail = video.get('preview')
 388
 389             rendition = self._download_json(
 390                 load_balancer_url, video_id, transform_source=strip_jsonp)
 391
 392             f = {
 393                 'abr': abr,
 394                 'vbr': vbr,
 395                 'fps': fps,
 396                 'width': width,
 397                 'height': height,
 398             }
 399
 400             formats = []
 401             for format_id, format_url in rendition['redirect'].items():
 402                 if format_id == 'rtmp':
 403                     ff = f.copy()
 404                     ff.update({
 405                         'url': format_url,
 406                         'format_id': format_id,
 407                     })
 408                     formats.append(ff)
 409                 elif determine_ext(format_url) == 'f4m':
 410                     formats.extend(self._extract_f4m_formats(
 411                         format_url, video_id, f4m_id=format_id))
 412                 elif determine_ext(format_url) == 'm3u8':
 413                     formats.extend(self._extract_m3u8_formats(
 414                         format_url, video_id, 'mp4', m3u8_id=format_id))
 415                 else:
 416                     continue
 417             self._sort_formats(formats)
 418
 419             title = remove_end(self._og_search_title(webpage), ' - fm4.ORF.at')
 420             if idx >= 1:
 421                 # Titles are duplicates, make them unique
 422                 title += ' (' + str(idx + 1) + ')'
 423             description = self._og_search_description(webpage)
 424             upload_date = unified_strdate(self._html_search_meta(
 425                 'dc.date', webpage, 'upload date'))
 426
 427             entries.append({
 428                 'id': video_id,
 429                 'title': title,
 430                 'description': description,
 431                 'duration': duration,
 432                 'thumbnail': thumbnail,
 433                 'upload_date': upload_date,
 434                 'formats': formats,
 435             })
 436
 437         return self.playlist_result(entries)