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