youtube_dl/extractor/dplay.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6 import time
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_HTTPError,
  11     compat_str,
  12     compat_urlparse,
  13 )
  14 from ..utils import (
  15     determine_ext,
  16     ExtractorError,
  17     float_or_none,
  18     int_or_none,
  19     remove_end,
  20     try_get,
  21     unified_strdate,
  22     unified_timestamp,
  23     update_url_query,
  24     USER_AGENTS,
  25 )
  26
  27
  28 class DPlayIE(InfoExtractor):
  29     _VALID_URL = r'https?://(?P<domain>www\.(?P<host>dplay\.(?P<country>dk|se|no)))/(?:video(?:er|s)/)?(?P<id>[^/]+/[^/?#]+)'
  30
  31     _TESTS = [{
  32         # non geo restricted, via secure api, unsigned download hls URL
  33         'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/',
  34         'info_dict': {
  35             'id': '3172',
  36             'display_id': 'nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet',
  37             'ext': 'mp4',
  38             'title': 'Svensken lär sig njuta av livet',
  39             'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8',
  40             'duration': 2650,
  41             'timestamp': 1365454320,
  42             'upload_date': '20130408',
  43             'creator': 'Kanal 5 (Home)',
  44             'series': 'Nugammalt - 77 händelser som format Sverige',
  45             'season_number': 1,
  46             'episode_number': 1,
  47             'age_limit': 0,
  48         },
  49     }, {
  50         # geo restricted, via secure api, unsigned download hls URL
  51         'url': 'http://www.dplay.dk/mig-og-min-mor/season-6-episode-12/',
  52         'info_dict': {
  53             'id': '70816',
  54             'display_id': 'mig-og-min-mor/season-6-episode-12',
  55             'ext': 'mp4',
  56             'title': 'Episode 12',
  57             'description': 'md5:9c86e51a93f8a4401fc9641ef9894c90',
  58             'duration': 2563,
  59             'timestamp': 1429696800,
  60             'upload_date': '20150422',
  61             'creator': 'Kanal 4 (Home)',
  62             'series': 'Mig og min mor',
  63             'season_number': 6,
  64             'episode_number': 12,
  65             'age_limit': 0,
  66         },
  67     }, {
  68         # geo restricted, via direct unsigned hls URL
  69         'url': 'http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/',
  70         'only_matching': True,
  71     }, {
  72         # disco-api
  73         'url': 'https://www.dplay.no/videoer/i-kongens-klr/sesong-1-episode-7',
  74         'info_dict': {
  75             'id': '40206',
  76             'display_id': 'i-kongens-klr/sesong-1-episode-7',
  77             'ext': 'mp4',
  78             'title': 'Episode 7',
  79             'description': 'md5:e3e1411b2b9aebeea36a6ec5d50c60cf',
  80             'duration': 2611.16,
  81             'timestamp': 1516726800,
  82             'upload_date': '20180123',
  83             'series': 'I kongens klær',
  84             'season_number': 1,
  85             'episode_number': 7,
  86         },
  87         'params': {
  88             'format': 'bestvideo',
  89             'skip_download': True,
  90         },
  91     }, {
  92
  93         'url': 'https://www.dplay.dk/videoer/singleliv/season-5-episode-3',
  94         'only_matching': True,
  95     }, {
  96         'url': 'https://www.dplay.se/videos/sofias-anglar/sofias-anglar-1001',
  97         'only_matching': True,
  98     }]
  99
 100     def _real_extract(self, url):
 101         mobj = re.match(self._VALID_URL, url)
 102         display_id = mobj.group('id')
 103         domain = mobj.group('domain')
 104
 105         self._initialize_geo_bypass({
 106             'countries': [mobj.group('country').upper()],
 107         })
 108
 109         webpage = self._download_webpage(url, display_id)
 110
 111         video_id = self._search_regex(
 112             r'data-video-id=["\'](\d+)', webpage, 'video id', default=None)
 113
 114         if not video_id:
 115             host = mobj.group('host')
 116             disco_base = 'https://disco-api.%s' % host
 117             self._download_json(
 118                 '%s/token' % disco_base, display_id, 'Downloading token',
 119                 query={
 120                     'realm': host.replace('.', ''),
 121                 })
 122             video = self._download_json(
 123                 '%s/content/videos/%s' % (disco_base, display_id), display_id,
 124                 headers={
 125                     'Referer': url,
 126                     'x-disco-client': 'WEB:UNKNOWN:dplay-client:0.0.1',
 127                 }, query={
 128                     'include': 'show'
 129                 })
 130             video_id = video['data']['id']
 131             info = video['data']['attributes']
 132             title = info['name']
 133             formats = []
 134             for format_id, format_dict in self._download_json(
 135                     '%s/playback/videoPlaybackInfo/%s' % (disco_base, video_id),
 136                     display_id)['data']['attributes']['streaming'].items():
 137                 if not isinstance(format_dict, dict):
 138                     continue
 139                 format_url = format_dict.get('url')
 140                 if not format_url:
 141                     continue
 142                 ext = determine_ext(format_url)
 143                 if format_id == 'dash' or ext == 'mpd':
 144                     formats.extend(self._extract_mpd_formats(
 145                         format_url, display_id, mpd_id='dash', fatal=False))
 146                 elif format_id == 'hls' or ext == 'm3u8':
 147                     formats.extend(self._extract_m3u8_formats(
 148                         format_url, display_id, 'mp4',
 149                         entry_protocol='m3u8_native', m3u8_id='hls',
 150                         fatal=False))
 151                 else:
 152                     formats.append({
 153                         'url': format_url,
 154                         'format_id': format_id,
 155                     })
 156             self._sort_formats(formats)
 157
 158             series = None
 159             try:
 160                 included = video.get('included')
 161                 if isinstance(included, list):
 162                     show = next(e for e in included if e.get('type') == 'show')
 163                     series = try_get(
 164                         show, lambda x: x['attributes']['name'], compat_str)
 165             except StopIteration:
 166                 pass
 167
 168             return {
 169                 'id': video_id,
 170                 'display_id': display_id,
 171                 'title': title,
 172                 'description': info.get('description'),
 173                 'duration': float_or_none(
 174                     info.get('videoDuration'), scale=1000),
 175                 'timestamp': unified_timestamp(info.get('publishStart')),
 176                 'series': series,
 177                 'season_number': int_or_none(info.get('seasonNumber')),
 178                 'episode_number': int_or_none(info.get('episodeNumber')),
 179                 'age_limit': int_or_none(info.get('minimum_age')),
 180                 'formats': formats,
 181             }
 182
 183         info = self._download_json(
 184             'http://%s/api/v2/ajax/videos?video_id=%s' % (domain, video_id),
 185             video_id)['data'][0]
 186
 187         title = info['title']
 188
 189         PROTOCOLS = ('hls', 'hds')
 190         formats = []
 191
 192         def extract_formats(protocol, manifest_url):
 193             if protocol == 'hls':
 194                 m3u8_formats = self._extract_m3u8_formats(
 195                     manifest_url, video_id, ext='mp4',
 196                     entry_protocol='m3u8_native', m3u8_id=protocol, fatal=False)
 197                 # Sometimes final URLs inside m3u8 are unsigned, let's fix this
 198                 # ourselves. Also fragments' URLs are only served signed for
 199                 # Safari user agent.
 200                 query = compat_urlparse.parse_qs(compat_urlparse.urlparse(manifest_url).query)
 201                 for m3u8_format in m3u8_formats:
 202                     m3u8_format.update({
 203                         'url': update_url_query(m3u8_format['url'], query),
 204                         'http_headers': {
 205                             'User-Agent': USER_AGENTS['Safari'],
 206                         },
 207                     })
 208                 formats.extend(m3u8_formats)
 209             elif protocol == 'hds':
 210                 formats.extend(self._extract_f4m_formats(
 211                     manifest_url + '&hdcore=3.8.0&plugin=flowplayer-3.8.0.0',
 212                     video_id, f4m_id=protocol, fatal=False))
 213
 214         domain_tld = domain.split('.')[-1]
 215         if domain_tld in ('se', 'dk', 'no'):
 216             for protocol in PROTOCOLS:
 217                 # Providing dsc-geo allows to bypass geo restriction in some cases
 218                 self._set_cookie(
 219                     'secure.dplay.%s' % domain_tld, 'dsc-geo',
 220                     json.dumps({
 221                         'countryCode': domain_tld.upper(),
 222                         'expiry': (time.time() + 20 * 60) * 1000,
 223                     }))
 224                 stream = self._download_json(
 225                     'https://secure.dplay.%s/secure/api/v2/user/authorization/stream/%s?stream_type=%s'
 226                     % (domain_tld, video_id, protocol), video_id,
 227                     'Downloading %s stream JSON' % protocol, fatal=False)
 228                 if stream and stream.get(protocol):
 229                     extract_formats(protocol, stream[protocol])
 230
 231         # The last resort is to try direct unsigned hls/hds URLs from info dictionary.
 232         # Sometimes this does work even when secure API with dsc-geo has failed (e.g.
 233         # http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/).
 234         if not formats:
 235             for protocol in PROTOCOLS:
 236                 if info.get(protocol):
 237                     extract_formats(protocol, info[protocol])
 238
 239         self._sort_formats(formats)
 240
 241         subtitles = {}
 242         for lang in ('se', 'sv', 'da', 'nl', 'no'):
 243             for format_id in ('web_vtt', 'vtt', 'srt'):
 244                 subtitle_url = info.get('subtitles_%s_%s' % (lang, format_id))
 245                 if subtitle_url:
 246                     subtitles.setdefault(lang, []).append({'url': subtitle_url})
 247
 248         return {
 249             'id': video_id,
 250             'display_id': display_id,
 251             'title': title,
 252             'description': info.get('video_metadata_longDescription'),
 253             'duration': int_or_none(info.get('video_metadata_length'), scale=1000),
 254             'timestamp': int_or_none(info.get('video_publish_date')),
 255             'creator': info.get('video_metadata_homeChannel'),
 256             'series': info.get('video_metadata_show'),
 257             'season_number': int_or_none(info.get('season')),
 258             'episode_number': int_or_none(info.get('episode')),
 259             'age_limit': int_or_none(info.get('minimum_age')),
 260             'formats': formats,
 261             'subtitles': subtitles,
 262         }
 263
 264
 265 class DPlayItIE(InfoExtractor):
 266     _VALID_URL = r'https?://it\.dplay\.com/[^/]+/[^/]+/(?P<id>[^/?#]+)'
 267     _GEO_COUNTRIES = ['IT']
 268     _TEST = {
 269         'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
 270         'md5': '2b808ffb00fc47b884a172ca5d13053c',
 271         'info_dict': {
 272             'id': '6918',
 273             'display_id': 'luigi-di-maio-la-psicosi-di-stanislawskij',
 274             'ext': 'mp4',
 275             'title': 'Biografie imbarazzanti: Luigi Di Maio: la psicosi di Stanislawskij',
 276             'description': 'md5:3c7a4303aef85868f867a26f5cc14813',
 277             'thumbnail': r're:^https?://.*\.jpe?g',
 278             'upload_date': '20160524',
 279             'series': 'Biografie imbarazzanti',
 280             'season_number': 1,
 281             'episode': 'Luigi Di Maio: la psicosi di Stanislawskij',
 282             'episode_number': 1,
 283         },
 284     }
 285
 286     def _real_extract(self, url):
 287         display_id = self._match_id(url)
 288
 289         webpage = self._download_webpage(url, display_id)
 290
 291         title = remove_end(self._og_search_title(webpage), ' | Dplay')
 292
 293         video_id = None
 294
 295         info = self._search_regex(
 296             r'playback_json\s*:\s*JSON\.parse\s*\(\s*("(?:\\.|[^"\\])+?")',
 297             webpage, 'playback JSON', default=None)
 298         if info:
 299             for _ in range(2):
 300                 info = self._parse_json(info, display_id, fatal=False)
 301                 if not info:
 302                     break
 303             else:
 304                 video_id = try_get(info, lambda x: x['data']['id'])
 305
 306         if not info:
 307             info_url = self._search_regex(
 308                 r'url\s*[:=]\s*["\']((?:https?:)?//[^/]+/playback/videoPlaybackInfo/\d+)',
 309                 webpage, 'info url')
 310
 311             video_id = info_url.rpartition('/')[-1]
 312
 313             try:
 314                 info = self._download_json(
 315                     info_url, display_id, headers={
 316                         'Authorization': 'Bearer %s' % self._get_cookies(url).get(
 317                             'dplayit_token').value,
 318                         'Referer': url,
 319                     })
 320             except ExtractorError as e:
 321                 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 403):
 322                     info = self._parse_json(e.cause.read().decode('utf-8'), display_id)
 323                     error = info['errors'][0]
 324                     if error.get('code') == 'access.denied.geoblocked':
 325                         self.raise_geo_restricted(
 326                             msg=error.get('detail'), countries=self._GEO_COUNTRIES)
 327                     raise ExtractorError(info['errors'][0]['detail'], expected=True)
 328                 raise
 329
 330         hls_url = info['data']['attributes']['streaming']['hls']['url']
 331
 332         formats = self._extract_m3u8_formats(
 333             hls_url, display_id, ext='mp4', entry_protocol='m3u8_native',
 334             m3u8_id='hls')
 335
 336         series = self._html_search_regex(
 337             r'(?s)<h1[^>]+class=["\'].*?\bshow_title\b.*?["\'][^>]*>(.+?)</h1>',
 338             webpage, 'series', fatal=False)
 339         episode = self._search_regex(
 340             r'<p[^>]+class=["\'].*?\bdesc_ep\b.*?["\'][^>]*>\s*<br/>\s*<b>([^<]+)',
 341             webpage, 'episode', fatal=False)
 342
 343         mobj = re.search(
 344             r'(?s)<span[^>]+class=["\']dates["\'][^>]*>.+?\bS\.(?P<season_number>\d+)\s+E\.(?P<episode_number>\d+)\s*-\s*(?P<upload_date>\d{2}/\d{2}/\d{4})',
 345             webpage)
 346         if mobj:
 347             season_number = int(mobj.group('season_number'))
 348             episode_number = int(mobj.group('episode_number'))
 349             upload_date = unified_strdate(mobj.group('upload_date'))
 350         else:
 351             season_number = episode_number = upload_date = None
 352
 353         return {
 354             'id': compat_str(video_id or display_id),
 355             'display_id': display_id,
 356             'title': title,
 357             'description': self._og_search_description(webpage),
 358             'thumbnail': self._og_search_thumbnail(webpage),
 359             'series': series,
 360             'season_number': season_number,
 361             'episode': episode,
 362             'episode_number': episode_number,
 363             'upload_date': upload_date,
 364             'formats': formats,
 365         }