Raphaël G. Git Repositories - youtubedl/blob - 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 _get_disco_api_info(self, url, display_id, disco_host, realm):
 101         disco_base = 'https://' + disco_host
 102         token = self._download_json(
 103             '%s/token' % disco_base, display_id, 'Downloading token',
 104             query={
 105                 'realm': realm,
 106             })['data']['attributes']['token']
 107         headers = {
 108             'Referer': url,
 109             'Authorization': 'Bearer ' + token,
 110         }
 111         video = self._download_json(
 112             '%s/content/videos/%s' % (disco_base, display_id), display_id,
 113             headers=headers, query={
 114                 'include': 'show'
 115             })
 116         video_id = video['data']['id']
 117         info = video['data']['attributes']
 118         title = info['name']
 119         formats = []
 120         for format_id, format_dict in self._download_json(
 121                 '%s/playback/videoPlaybackInfo/%s' % (disco_base, video_id),
 122                 display_id, headers=headers)['data']['attributes']['streaming'].items():
 123             if not isinstance(format_dict, dict):
 124                 continue
 125             format_url = format_dict.get('url')
 126             if not format_url:
 127                 continue
 128             ext = determine_ext(format_url)
 129             if format_id == 'dash' or ext == 'mpd':
 130                 formats.extend(self._extract_mpd_formats(
 131                     format_url, display_id, mpd_id='dash', fatal=False))
 132             elif format_id == 'hls' or ext == 'm3u8':
 133                 formats.extend(self._extract_m3u8_formats(
 134                     format_url, display_id, 'mp4',
 135                     entry_protocol='m3u8_native', m3u8_id='hls',
 136                     fatal=False))
 137             else:
 138                 formats.append({
 139                     'url': format_url,
 140                     'format_id': format_id,
 141                 })
 142         self._sort_formats(formats)
 143
 144         series = None
 145         try:
 146             included = video.get('included')
 147             if isinstance(included, list):
 148                 show = next(e for e in included if e.get('type') == 'show')
 149                 series = try_get(
 150                     show, lambda x: x['attributes']['name'], compat_str)
 151         except StopIteration:
 152             pass
 153
 154         return {
 155             'id': video_id,
 156             'display_id': display_id,
 157             'title': title,
 158             'description': info.get('description'),
 159             'duration': float_or_none(
 160                 info.get('videoDuration'), scale=1000),
 161             'timestamp': unified_timestamp(info.get('publishStart')),
 162             'series': series,
 163             'season_number': int_or_none(info.get('seasonNumber')),
 164             'episode_number': int_or_none(info.get('episodeNumber')),
 165             'age_limit': int_or_none(info.get('minimum_age')),
 166             'formats': formats,
 167         }
 168
 169     def _real_extract(self, url):
 170         mobj = re.match(self._VALID_URL, url)
 171         display_id = mobj.group('id')
 172         domain = mobj.group('domain')
 173
 174         self._initialize_geo_bypass({
 175             'countries': [mobj.group('country').upper()],
 176         })
 177
 178         webpage = self._download_webpage(url, display_id)
 179
 180         video_id = self._search_regex(
 181             r'data-video-id=["\'](\d+)', webpage, 'video id', default=None)
 182
 183         if not video_id:
 184             host = mobj.group('host')
 185             return self._get_disco_api_info(
 186                 url, display_id, 'disco-api.' + host, host.replace('.', ''))
 187
 188         info = self._download_json(
 189             'http://%s/api/v2/ajax/videos?video_id=%s' % (domain, video_id),
 190             video_id)['data'][0]
 191
 192         title = info['title']
 193
 194         PROTOCOLS = ('hls', 'hds')
 195         formats = []
 196
 197         def extract_formats(protocol, manifest_url):
 198             if protocol == 'hls':
 199                 m3u8_formats = self._extract_m3u8_formats(
 200                     manifest_url, video_id, ext='mp4',
 201                     entry_protocol='m3u8_native', m3u8_id=protocol, fatal=False)
 202                 # Sometimes final URLs inside m3u8 are unsigned, let's fix this
 203                 # ourselves. Also fragments' URLs are only served signed for
 204                 # Safari user agent.
 205                 query = compat_urlparse.parse_qs(compat_urlparse.urlparse(manifest_url).query)
 206                 for m3u8_format in m3u8_formats:
 207                     m3u8_format.update({
 208                         'url': update_url_query(m3u8_format['url'], query),
 209                         'http_headers': {
 210                             'User-Agent': USER_AGENTS['Safari'],
 211                         },
 212                     })
 213                 formats.extend(m3u8_formats)
 214             elif protocol == 'hds':
 215                 formats.extend(self._extract_f4m_formats(
 216                     manifest_url + '&hdcore=3.8.0&plugin=flowplayer-3.8.0.0',
 217                     video_id, f4m_id=protocol, fatal=False))
 218
 219         domain_tld = domain.split('.')[-1]
 220         if domain_tld in ('se', 'dk', 'no'):
 221             for protocol in PROTOCOLS:
 222                 # Providing dsc-geo allows to bypass geo restriction in some cases
 223                 self._set_cookie(
 224                     'secure.dplay.%s' % domain_tld, 'dsc-geo',
 225                     json.dumps({
 226                         'countryCode': domain_tld.upper(),
 227                         'expiry': (time.time() + 20 * 60) * 1000,
 228                     }))
 229                 stream = self._download_json(
 230                     'https://secure.dplay.%s/secure/api/v2/user/authorization/stream/%s?stream_type=%s'
 231                     % (domain_tld, video_id, protocol), video_id,
 232                     'Downloading %s stream JSON' % protocol, fatal=False)
 233                 if stream and stream.get(protocol):
 234                     extract_formats(protocol, stream[protocol])
 235
 236         # The last resort is to try direct unsigned hls/hds URLs from info dictionary.
 237         # Sometimes this does work even when secure API with dsc-geo has failed (e.g.
 238         # http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/).
 239         if not formats:
 240             for protocol in PROTOCOLS:
 241                 if info.get(protocol):
 242                     extract_formats(protocol, info[protocol])
 243
 244         self._sort_formats(formats)
 245
 246         subtitles = {}
 247         for lang in ('se', 'sv', 'da', 'nl', 'no'):
 248             for format_id in ('web_vtt', 'vtt', 'srt'):
 249                 subtitle_url = info.get('subtitles_%s_%s' % (lang, format_id))
 250                 if subtitle_url:
 251                     subtitles.setdefault(lang, []).append({'url': subtitle_url})
 252
 253         return {
 254             'id': video_id,
 255             'display_id': display_id,
 256             'title': title,
 257             'description': info.get('video_metadata_longDescription'),
 258             'duration': int_or_none(info.get('video_metadata_length'), scale=1000),
 259             'timestamp': int_or_none(info.get('video_publish_date')),
 260             'creator': info.get('video_metadata_homeChannel'),
 261             'series': info.get('video_metadata_show'),
 262             'season_number': int_or_none(info.get('season')),
 263             'episode_number': int_or_none(info.get('episode')),
 264             'age_limit': int_or_none(info.get('minimum_age')),
 265             'formats': formats,
 266             'subtitles': subtitles,
 267         }
 268
 269
 270 class DPlayItIE(InfoExtractor):
 271     _VALID_URL = r'https?://it\.dplay\.com/[^/]+/[^/]+/(?P<id>[^/?#]+)'
 272     _GEO_COUNTRIES = ['IT']
 273     _TEST = {
 274         'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
 275         'md5': '2b808ffb00fc47b884a172ca5d13053c',
 276         'info_dict': {
 277             'id': '6918',
 278             'display_id': 'luigi-di-maio-la-psicosi-di-stanislawskij',
 279             'ext': 'mp4',
 280             'title': 'Biografie imbarazzanti: Luigi Di Maio: la psicosi di Stanislawskij',
 281             'description': 'md5:3c7a4303aef85868f867a26f5cc14813',
 282             'thumbnail': r're:^https?://.*\.jpe?g',
 283             'upload_date': '20160524',
 284             'series': 'Biografie imbarazzanti',
 285             'season_number': 1,
 286             'episode': 'Luigi Di Maio: la psicosi di Stanislawskij',
 287             'episode_number': 1,
 288         },
 289     }
 290
 291     def _real_extract(self, url):
 292         display_id = self._match_id(url)
 293
 294         webpage = self._download_webpage(url, display_id)
 295
 296         title = remove_end(self._og_search_title(webpage), ' | Dplay')
 297
 298         video_id = None
 299
 300         info = self._search_regex(
 301             r'playback_json\s*:\s*JSON\.parse\s*\(\s*("(?:\\.|[^"\\])+?")',
 302             webpage, 'playback JSON', default=None)
 303         if info:
 304             for _ in range(2):
 305                 info = self._parse_json(info, display_id, fatal=False)
 306                 if not info:
 307                     break
 308             else:
 309                 video_id = try_get(info, lambda x: x['data']['id'])
 310
 311         if not info:
 312             info_url = self._search_regex(
 313                 r'url\s*[:=]\s*["\']((?:https?:)?//[^/]+/playback/videoPlaybackInfo/\d+)',
 314                 webpage, 'info url')
 315
 316             video_id = info_url.rpartition('/')[-1]
 317
 318             try:
 319                 info = self._download_json(
 320                     info_url, display_id, headers={
 321                         'Authorization': 'Bearer %s' % self._get_cookies(url).get(
 322                             'dplayit_token').value,
 323                         'Referer': url,
 324                     })
 325             except ExtractorError as e:
 326                 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 403):
 327                     info = self._parse_json(e.cause.read().decode('utf-8'), display_id)
 328                     error = info['errors'][0]
 329                     if error.get('code') == 'access.denied.geoblocked':
 330                         self.raise_geo_restricted(
 331                             msg=error.get('detail'), countries=self._GEO_COUNTRIES)
 332                     raise ExtractorError(info['errors'][0]['detail'], expected=True)
 333                 raise
 334
 335         hls_url = info['data']['attributes']['streaming']['hls']['url']
 336
 337         formats = self._extract_m3u8_formats(
 338             hls_url, display_id, ext='mp4', entry_protocol='m3u8_native',
 339             m3u8_id='hls')
 340
 341         series = self._html_search_regex(
 342             r'(?s)<h1[^>]+class=["\'].*?\bshow_title\b.*?["\'][^>]*>(.+?)</h1>',
 343             webpage, 'series', fatal=False)
 344         episode = self._search_regex(
 345             r'<p[^>]+class=["\'].*?\bdesc_ep\b.*?["\'][^>]*>\s*<br/>\s*<b>([^<]+)',
 346             webpage, 'episode', fatal=False)
 347
 348         mobj = re.search(
 349             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})',
 350             webpage)
 351         if mobj:
 352             season_number = int(mobj.group('season_number'))
 353             episode_number = int(mobj.group('episode_number'))
 354             upload_date = unified_strdate(mobj.group('upload_date'))
 355         else:
 356             season_number = episode_number = upload_date = None
 357
 358         return {
 359             'id': compat_str(video_id or display_id),
 360             'display_id': display_id,
 361             'title': title,
 362             'description': self._og_search_description(webpage),
 363             'thumbnail': self._og_search_thumbnail(webpage),
 364             'series': series,
 365             'season_number': season_number,
 366             'episode': episode,
 367             'episode_number': episode_number,
 368             'upload_date': upload_date,
 369             'formats': formats,
 370         }