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