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     urljoin,
  25     USER_AGENTS,
  26 )
  27
  28
  29 class DPlayIE(InfoExtractor):
  30     _VALID_URL = r'https?://(?P<domain>www\.(?P<host>dplay\.(?P<country>dk|se|no)))/(?:video(?:er|s)/)?(?P<id>[^/]+/[^/?#]+)'
  31
  32     _TESTS = [{
  33         # non geo restricted, via secure api, unsigned download hls URL
  34         'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/',
  35         'info_dict': {
  36             'id': '3172',
  37             'display_id': 'nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet',
  38             'ext': 'mp4',
  39             'title': 'Svensken lär sig njuta av livet',
  40             'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8',
  41             'duration': 2650,
  42             'timestamp': 1365454320,
  43             'upload_date': '20130408',
  44             'creator': 'Kanal 5 (Home)',
  45             'series': 'Nugammalt - 77 händelser som format Sverige',
  46             'season_number': 1,
  47             'episode_number': 1,
  48             'age_limit': 0,
  49         },
  50     }, {
  51         # geo restricted, via secure api, unsigned download hls URL
  52         'url': 'http://www.dplay.dk/mig-og-min-mor/season-6-episode-12/',
  53         'info_dict': {
  54             'id': '70816',
  55             'display_id': 'mig-og-min-mor/season-6-episode-12',
  56             'ext': 'mp4',
  57             'title': 'Episode 12',
  58             'description': 'md5:9c86e51a93f8a4401fc9641ef9894c90',
  59             'duration': 2563,
  60             'timestamp': 1429696800,
  61             'upload_date': '20150422',
  62             'creator': 'Kanal 4 (Home)',
  63             'series': 'Mig og min mor',
  64             'season_number': 6,
  65             'episode_number': 12,
  66             'age_limit': 0,
  67         },
  68     }, {
  69         # geo restricted, via direct unsigned hls URL
  70         'url': 'http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/',
  71         'only_matching': True,
  72     }, {
  73         # disco-api
  74         'url': 'https://www.dplay.no/videoer/i-kongens-klr/sesong-1-episode-7',
  75         'info_dict': {
  76             'id': '40206',
  77             'display_id': 'i-kongens-klr/sesong-1-episode-7',
  78             'ext': 'mp4',
  79             'title': 'Episode 7',
  80             'description': 'md5:e3e1411b2b9aebeea36a6ec5d50c60cf',
  81             'duration': 2611.16,
  82             'timestamp': 1516726800,
  83             'upload_date': '20180123',
  84             'series': 'I kongens klær',
  85             'season_number': 1,
  86             'episode_number': 7,
  87         },
  88         'params': {
  89             'format': 'bestvideo',
  90             'skip_download': True,
  91         },
  92     }, {
  93
  94         'url': 'https://www.dplay.dk/videoer/singleliv/season-5-episode-3',
  95         'only_matching': True,
  96     }, {
  97         'url': 'https://www.dplay.se/videos/sofias-anglar/sofias-anglar-1001',
  98         'only_matching': True,
  99     }]
 100
 101     def _get_disco_api_info(self, url, display_id, disco_host, realm):
 102         disco_base = 'https://' + disco_host
 103         token = self._download_json(
 104             '%s/token' % disco_base, display_id, 'Downloading token',
 105             query={
 106                 'realm': realm,
 107             })['data']['attributes']['token']
 108         headers = {
 109             'Referer': url,
 110             'Authorization': 'Bearer ' + token,
 111         }
 112         video = self._download_json(
 113             '%s/content/videos/%s' % (disco_base, display_id), display_id,
 114             headers=headers, query={
 115                 'include': 'show'
 116             })
 117         video_id = video['data']['id']
 118         info = video['data']['attributes']
 119         title = info['name']
 120         formats = []
 121         for format_id, format_dict in self._download_json(
 122                 '%s/playback/videoPlaybackInfo/%s' % (disco_base, video_id),
 123                 display_id, headers=headers)['data']['attributes']['streaming'].items():
 124             if not isinstance(format_dict, dict):
 125                 continue
 126             format_url = format_dict.get('url')
 127             if not format_url:
 128                 continue
 129             ext = determine_ext(format_url)
 130             if format_id == 'dash' or ext == 'mpd':
 131                 formats.extend(self._extract_mpd_formats(
 132                     format_url, display_id, mpd_id='dash', fatal=False))
 133             elif format_id == 'hls' or ext == 'm3u8':
 134                 formats.extend(self._extract_m3u8_formats(
 135                     format_url, display_id, 'mp4',
 136                     entry_protocol='m3u8_native', m3u8_id='hls',
 137                     fatal=False))
 138             else:
 139                 formats.append({
 140                     'url': format_url,
 141                     'format_id': format_id,
 142                 })
 143         self._sort_formats(formats)
 144
 145         series = None
 146         try:
 147             included = video.get('included')
 148             if isinstance(included, list):
 149                 show = next(e for e in included if e.get('type') == 'show')
 150                 series = try_get(
 151                     show, lambda x: x['attributes']['name'], compat_str)
 152         except StopIteration:
 153             pass
 154
 155         return {
 156             'id': video_id,
 157             'display_id': display_id,
 158             'title': title,
 159             'description': info.get('description'),
 160             'duration': float_or_none(
 161                 info.get('videoDuration'), scale=1000),
 162             'timestamp': unified_timestamp(info.get('publishStart')),
 163             'series': series,
 164             'season_number': int_or_none(info.get('seasonNumber')),
 165             'episode_number': int_or_none(info.get('episodeNumber')),
 166             'age_limit': int_or_none(info.get('minimum_age')),
 167             'formats': formats,
 168         }
 169
 170     def _real_extract(self, url):
 171         mobj = re.match(self._VALID_URL, url)
 172         display_id = mobj.group('id')
 173         domain = mobj.group('domain')
 174
 175         self._initialize_geo_bypass({
 176             'countries': [mobj.group('country').upper()],
 177         })
 178
 179         webpage = self._download_webpage(url, display_id)
 180
 181         video_id = self._search_regex(
 182             r'data-video-id=["\'](\d+)', webpage, 'video id', default=None)
 183
 184         if not video_id:
 185             host = mobj.group('host')
 186             return self._get_disco_api_info(
 187                 url, display_id, 'disco-api.' + host, host.replace('.', ''))
 188
 189         info = self._download_json(
 190             'http://%s/api/v2/ajax/videos?video_id=%s' % (domain, video_id),
 191             video_id)['data'][0]
 192
 193         title = info['title']
 194
 195         PROTOCOLS = ('hls', 'hds')
 196         formats = []
 197
 198         def extract_formats(protocol, manifest_url):
 199             if protocol == 'hls':
 200                 m3u8_formats = self._extract_m3u8_formats(
 201                     manifest_url, video_id, ext='mp4',
 202                     entry_protocol='m3u8_native', m3u8_id=protocol, fatal=False)
 203                 # Sometimes final URLs inside m3u8 are unsigned, let's fix this
 204                 # ourselves. Also fragments' URLs are only served signed for
 205                 # Safari user agent.
 206                 query = compat_urlparse.parse_qs(compat_urlparse.urlparse(manifest_url).query)
 207                 for m3u8_format in m3u8_formats:
 208                     m3u8_format.update({
 209                         'url': update_url_query(m3u8_format['url'], query),
 210                         'http_headers': {
 211                             'User-Agent': USER_AGENTS['Safari'],
 212                         },
 213                     })
 214                 formats.extend(m3u8_formats)
 215             elif protocol == 'hds':
 216                 formats.extend(self._extract_f4m_formats(
 217                     manifest_url + '&hdcore=3.8.0&plugin=flowplayer-3.8.0.0',
 218                     video_id, f4m_id=protocol, fatal=False))
 219
 220         domain_tld = domain.split('.')[-1]
 221         if domain_tld in ('se', 'dk', 'no'):
 222             for protocol in PROTOCOLS:
 223                 # Providing dsc-geo allows to bypass geo restriction in some cases
 224                 self._set_cookie(
 225                     'secure.dplay.%s' % domain_tld, 'dsc-geo',
 226                     json.dumps({
 227                         'countryCode': domain_tld.upper(),
 228                         'expiry': (time.time() + 20 * 60) * 1000,
 229                     }))
 230                 stream = self._download_json(
 231                     'https://secure.dplay.%s/secure/api/v2/user/authorization/stream/%s?stream_type=%s'
 232                     % (domain_tld, video_id, protocol), video_id,
 233                     'Downloading %s stream JSON' % protocol, fatal=False)
 234                 if stream and stream.get(protocol):
 235                     extract_formats(protocol, stream[protocol])
 236
 237         # The last resort is to try direct unsigned hls/hds URLs from info dictionary.
 238         # Sometimes this does work even when secure API with dsc-geo has failed (e.g.
 239         # http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/).
 240         if not formats:
 241             for protocol in PROTOCOLS:
 242                 if info.get(protocol):
 243                     extract_formats(protocol, info[protocol])
 244
 245         self._sort_formats(formats)
 246
 247         subtitles = {}
 248         for lang in ('se', 'sv', 'da', 'nl', 'no'):
 249             for format_id in ('web_vtt', 'vtt', 'srt'):
 250                 subtitle_url = info.get('subtitles_%s_%s' % (lang, format_id))
 251                 if subtitle_url:
 252                     subtitles.setdefault(lang, []).append({'url': subtitle_url})
 253
 254         return {
 255             'id': video_id,
 256             'display_id': display_id,
 257             'title': title,
 258             'description': info.get('video_metadata_longDescription'),
 259             'duration': int_or_none(info.get('video_metadata_length'), scale=1000),
 260             'timestamp': int_or_none(info.get('video_publish_date')),
 261             'creator': info.get('video_metadata_homeChannel'),
 262             'series': info.get('video_metadata_show'),
 263             'season_number': int_or_none(info.get('season')),
 264             'episode_number': int_or_none(info.get('episode')),
 265             'age_limit': int_or_none(info.get('minimum_age')),
 266             'formats': formats,
 267             'subtitles': subtitles,
 268         }
 269
 270
 271 class DPlayItIE(InfoExtractor):
 272     _VALID_URL = r'https?://it\.dplay\.com/[^/]+/[^/]+/(?P<id>[^/?#]+)'
 273     _GEO_COUNTRIES = ['IT']
 274     _TEST = {
 275         'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
 276         'md5': '2b808ffb00fc47b884a172ca5d13053c',
 277         'info_dict': {
 278             'id': '6918',
 279             'display_id': 'luigi-di-maio-la-psicosi-di-stanislawskij',
 280             'ext': 'mp4',
 281             'title': 'Biografie imbarazzanti: Luigi Di Maio: la psicosi di Stanislawskij',
 282             'description': 'md5:3c7a4303aef85868f867a26f5cc14813',
 283             'thumbnail': r're:^https?://.*\.jpe?g',
 284             'upload_date': '20160524',
 285             'series': 'Biografie imbarazzanti',
 286             'season_number': 1,
 287             'episode': 'Luigi Di Maio: la psicosi di Stanislawskij',
 288             'episode_number': 1,
 289         },
 290     }
 291
 292     def _real_extract(self, url):
 293         display_id = self._match_id(url)
 294
 295         webpage = self._download_webpage(url, display_id)
 296
 297         title = remove_end(self._og_search_title(webpage), ' | Dplay')
 298
 299         video_id = None
 300
 301         info = self._search_regex(
 302             r'playback_json\s*:\s*JSON\.parse\s*\(\s*("(?:\\.|[^"\\])+?")',
 303             webpage, 'playback JSON', default=None)
 304         if info:
 305             for _ in range(2):
 306                 info = self._parse_json(info, display_id, fatal=False)
 307                 if not info:
 308                     break
 309             else:
 310                 video_id = try_get(info, lambda x: x['data']['id'])
 311
 312         if not info:
 313             info_url = self._search_regex(
 314                 (r'playback_json_url\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1',
 315                  r'url\s*[:=]\s*["\'](?P<url>(?:https?:)?//[^/]+/playback/videoPlaybackInfo/\d+)'),
 316                 webpage, 'info url', group='url')
 317
 318             info_url = urljoin(url, info_url)
 319             video_id = info_url.rpartition('/')[-1]
 320
 321             try:
 322                 info = self._download_json(
 323                     info_url, display_id, headers={
 324                         'Authorization': 'Bearer %s' % self._get_cookies(url).get(
 325                             'dplayit_token').value,
 326                         'Referer': url,
 327                     })
 328                 if isinstance(info, compat_str):
 329                     info = self._parse_json(info, display_id)
 330             except ExtractorError as e:
 331                 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 403):
 332                     info = self._parse_json(e.cause.read().decode('utf-8'), display_id)
 333                     error = info['errors'][0]
 334                     if error.get('code') == 'access.denied.geoblocked':
 335                         self.raise_geo_restricted(
 336                             msg=error.get('detail'), countries=self._GEO_COUNTRIES)
 337                     raise ExtractorError(info['errors'][0]['detail'], expected=True)
 338                 raise
 339
 340         hls_url = info['data']['attributes']['streaming']['hls']['url']
 341
 342         formats = self._extract_m3u8_formats(
 343             hls_url, display_id, ext='mp4', entry_protocol='m3u8_native',
 344             m3u8_id='hls')
 345         self._sort_formats(formats)
 346
 347         series = self._html_search_regex(
 348             r'(?s)<h1[^>]+class=["\'].*?\bshow_title\b.*?["\'][^>]*>(.+?)</h1>',
 349             webpage, 'series', fatal=False)
 350         episode = self._search_regex(
 351             r'<p[^>]+class=["\'].*?\bdesc_ep\b.*?["\'][^>]*>\s*<br/>\s*<b>([^<]+)',
 352             webpage, 'episode', fatal=False)
 353
 354         mobj = re.search(
 355             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})',
 356             webpage)
 357         if mobj:
 358             season_number = int(mobj.group('season_number'))
 359             episode_number = int(mobj.group('episode_number'))
 360             upload_date = unified_strdate(mobj.group('upload_date'))
 361         else:
 362             season_number = episode_number = upload_date = None
 363
 364         return {
 365             'id': compat_str(video_id or display_id),
 366             'display_id': display_id,
 367             'title': title,
 368             'description': self._og_search_description(webpage),
 369             'thumbnail': self._og_search_thumbnail(webpage),
 370             'series': series,
 371             'season_number': season_number,
 372             'episode': episode,
 373             'episode_number': episode_number,
 374             'upload_date': upload_date,
 375             'formats': formats,
 376         }