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