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     ExtractorError,
  16     int_or_none,
  17     remove_end,
  18     try_get,
  19     unified_strdate,
  20     update_url_query,
  21     USER_AGENTS,
  22 )
  23
  24
  25 class DPlayIE(InfoExtractor):
  26     _VALID_URL = r'https?://(?P<domain>www\.dplay\.(?:dk|se|no))/[^/]+/(?P<id>[^/?#]+)'
  27
  28     _TESTS = [{
  29         # non geo restricted, via secure api, unsigned download hls URL
  30         'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/',
  31         'info_dict': {
  32             'id': '3172',
  33             'display_id': 'season-1-svensken-lar-sig-njuta-av-livet',
  34             'ext': 'mp4',
  35             'title': 'Svensken lär sig njuta av livet',
  36             'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8',
  37             'duration': 2650,
  38             'timestamp': 1365454320,
  39             'upload_date': '20130408',
  40             'creator': 'Kanal 5 (Home)',
  41             'series': 'Nugammalt - 77 händelser som format Sverige',
  42             'season_number': 1,
  43             'episode_number': 1,
  44             'age_limit': 0,
  45         },
  46     }, {
  47         # geo restricted, via secure api, unsigned download hls URL
  48         'url': 'http://www.dplay.dk/mig-og-min-mor/season-6-episode-12/',
  49         'info_dict': {
  50             'id': '70816',
  51             'display_id': 'season-6-episode-12',
  52             'ext': 'mp4',
  53             'title': 'Episode 12',
  54             'description': 'md5:9c86e51a93f8a4401fc9641ef9894c90',
  55             'duration': 2563,
  56             'timestamp': 1429696800,
  57             'upload_date': '20150422',
  58             'creator': 'Kanal 4 (Home)',
  59             'series': 'Mig og min mor',
  60             'season_number': 6,
  61             'episode_number': 12,
  62             'age_limit': 0,
  63         },
  64     }, {
  65         # geo restricted, via direct unsigned hls URL
  66         'url': 'http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/',
  67         'only_matching': True,
  68     }]
  69
  70     def _real_extract(self, url):
  71         mobj = re.match(self._VALID_URL, url)
  72         display_id = mobj.group('id')
  73         domain = mobj.group('domain')
  74
  75         webpage = self._download_webpage(url, display_id)
  76
  77         video_id = self._search_regex(
  78             r'data-video-id=["\'](\d+)', webpage, 'video id')
  79
  80         info = self._download_json(
  81             'http://%s/api/v2/ajax/videos?video_id=%s' % (domain, video_id),
  82             video_id)['data'][0]
  83
  84         title = info['title']
  85
  86         PROTOCOLS = ('hls', 'hds')
  87         formats = []
  88
  89         def extract_formats(protocol, manifest_url):
  90             if protocol == 'hls':
  91                 m3u8_formats = self._extract_m3u8_formats(
  92                     manifest_url, video_id, ext='mp4',
  93                     entry_protocol='m3u8_native', m3u8_id=protocol, fatal=False)
  94                 # Sometimes final URLs inside m3u8 are unsigned, let's fix this
  95                 # ourselves. Also fragments' URLs are only served signed for
  96                 # Safari user agent.
  97                 query = compat_urlparse.parse_qs(compat_urlparse.urlparse(manifest_url).query)
  98                 for m3u8_format in m3u8_formats:
  99                     m3u8_format.update({
 100                         'url': update_url_query(m3u8_format['url'], query),
 101                         'http_headers': {
 102                             'User-Agent': USER_AGENTS['Safari'],
 103                         },
 104                     })
 105                 formats.extend(m3u8_formats)
 106             elif protocol == 'hds':
 107                 formats.extend(self._extract_f4m_formats(
 108                     manifest_url + '&hdcore=3.8.0&plugin=flowplayer-3.8.0.0',
 109                     video_id, f4m_id=protocol, fatal=False))
 110
 111         domain_tld = domain.split('.')[-1]
 112         if domain_tld in ('se', 'dk', 'no'):
 113             for protocol in PROTOCOLS:
 114                 # Providing dsc-geo allows to bypass geo restriction in some cases
 115                 self._set_cookie(
 116                     'secure.dplay.%s' % domain_tld, 'dsc-geo',
 117                     json.dumps({
 118                         'countryCode': domain_tld.upper(),
 119                         'expiry': (time.time() + 20 * 60) * 1000,
 120                     }))
 121                 stream = self._download_json(
 122                     'https://secure.dplay.%s/secure/api/v2/user/authorization/stream/%s?stream_type=%s'
 123                     % (domain_tld, video_id, protocol), video_id,
 124                     'Downloading %s stream JSON' % protocol, fatal=False)
 125                 if stream and stream.get(protocol):
 126                     extract_formats(protocol, stream[protocol])
 127
 128         # The last resort is to try direct unsigned hls/hds URLs from info dictionary.
 129         # Sometimes this does work even when secure API with dsc-geo has failed (e.g.
 130         # http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/).
 131         if not formats:
 132             for protocol in PROTOCOLS:
 133                 if info.get(protocol):
 134                     extract_formats(protocol, info[protocol])
 135
 136         self._sort_formats(formats)
 137
 138         subtitles = {}
 139         for lang in ('se', 'sv', 'da', 'nl', 'no'):
 140             for format_id in ('web_vtt', 'vtt', 'srt'):
 141                 subtitle_url = info.get('subtitles_%s_%s' % (lang, format_id))
 142                 if subtitle_url:
 143                     subtitles.setdefault(lang, []).append({'url': subtitle_url})
 144
 145         return {
 146             'id': video_id,
 147             'display_id': display_id,
 148             'title': title,
 149             'description': info.get('video_metadata_longDescription'),
 150             'duration': int_or_none(info.get('video_metadata_length'), scale=1000),
 151             'timestamp': int_or_none(info.get('video_publish_date')),
 152             'creator': info.get('video_metadata_homeChannel'),
 153             'series': info.get('video_metadata_show'),
 154             'season_number': int_or_none(info.get('season')),
 155             'episode_number': int_or_none(info.get('episode')),
 156             'age_limit': int_or_none(info.get('minimum_age')),
 157             'formats': formats,
 158             'subtitles': subtitles,
 159         }
 160
 161
 162 class DPlayItIE(InfoExtractor):
 163     _VALID_URL = r'https?://it\.dplay\.com/[^/]+/[^/]+/(?P<id>[^/?#]+)'
 164     _GEO_COUNTRIES = ['IT']
 165     _TEST = {
 166         'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
 167         'md5': '2b808ffb00fc47b884a172ca5d13053c',
 168         'info_dict': {
 169             'id': '6918',
 170             'display_id': 'luigi-di-maio-la-psicosi-di-stanislawskij',
 171             'ext': 'mp4',
 172             'title': 'Biografie imbarazzanti: Luigi Di Maio: la psicosi di Stanislawskij',
 173             'description': 'md5:3c7a4303aef85868f867a26f5cc14813',
 174             'thumbnail': r're:^https?://.*\.jpe?g',
 175             'upload_date': '20160524',
 176             'series': 'Biografie imbarazzanti',
 177             'season_number': 1,
 178             'episode': 'Luigi Di Maio: la psicosi di Stanislawskij',
 179             'episode_number': 1,
 180         },
 181     }
 182
 183     def _real_extract(self, url):
 184         display_id = self._match_id(url)
 185
 186         webpage = self._download_webpage(url, display_id)
 187
 188         title = remove_end(self._og_search_title(webpage), ' | Dplay')
 189
 190         video_id = None
 191
 192         info = self._search_regex(
 193             r'playback_json\s*:\s*JSON\.parse\s*\(\s*("(?:\\.|[^"\\])+?")',
 194             webpage, 'playback JSON', default=None)
 195         if info:
 196             for _ in range(2):
 197                 info = self._parse_json(info, display_id, fatal=False)
 198                 if not info:
 199                     break
 200             else:
 201                 video_id = try_get(info, lambda x: x['data']['id'])
 202
 203         if not info:
 204             info_url = self._search_regex(
 205                 r'url\s*[:=]\s*["\']((?:https?:)?//[^/]+/playback/videoPlaybackInfo/\d+)',
 206                 webpage, 'info url')
 207
 208             video_id = info_url.rpartition('/')[-1]
 209
 210             try:
 211                 info = self._download_json(
 212                     info_url, display_id, headers={
 213                         'Authorization': 'Bearer %s' % self._get_cookies(url).get(
 214                             'dplayit_token').value,
 215                         'Referer': url,
 216                     })
 217             except ExtractorError as e:
 218                 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 403):
 219                     info = self._parse_json(e.cause.read().decode('utf-8'), display_id)
 220                     error = info['errors'][0]
 221                     if error.get('code') == 'access.denied.geoblocked':
 222                         self.raise_geo_restricted(
 223                             msg=error.get('detail'), countries=self._GEO_COUNTRIES)
 224                     raise ExtractorError(info['errors'][0]['detail'], expected=True)
 225                 raise
 226
 227         hls_url = info['data']['attributes']['streaming']['hls']['url']
 228
 229         formats = self._extract_m3u8_formats(
 230             hls_url, display_id, ext='mp4', entry_protocol='m3u8_native',
 231             m3u8_id='hls')
 232
 233         series = self._html_search_regex(
 234             r'(?s)<h1[^>]+class=["\'].*?\bshow_title\b.*?["\'][^>]*>(.+?)</h1>',
 235             webpage, 'series', fatal=False)
 236         episode = self._search_regex(
 237             r'<p[^>]+class=["\'].*?\bdesc_ep\b.*?["\'][^>]*>\s*<br/>\s*<b>([^<]+)',
 238             webpage, 'episode', fatal=False)
 239
 240         mobj = re.search(
 241             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})',
 242             webpage)
 243         if mobj:
 244             season_number = int(mobj.group('season_number'))
 245             episode_number = int(mobj.group('episode_number'))
 246             upload_date = unified_strdate(mobj.group('upload_date'))
 247         else:
 248             season_number = episode_number = upload_date = None
 249
 250         return {
 251             'id': compat_str(video_id or display_id),
 252             'display_id': display_id,
 253             'title': title,
 254             'description': self._og_search_description(webpage),
 255             'thumbnail': self._og_search_thumbnail(webpage),
 256             'series': series,
 257             'season_number': season_number,
 258             'episode': episode,
 259             'episode_number': episode_number,
 260             'upload_date': upload_date,
 261             'formats': formats,
 262         }