2 from __future__ import unicode_literals
6 from .common import InfoExtractor
17 class EaglePlatformIE(InfoExtractor):
20 eagleplatform:(?P<custom_host>[^/]+):|
21 https?://(?P<host>.+?\.media\.eagleplatform\.com)/index/player\?.*\brecord_id=
26 # http://lenta.ru/news/2015/03/06/navalny/
27 'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',
28 # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
32 'title': 'Навальный вышел на свободу',
33 'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
34 'thumbnail': r're:^https?://.*\.jpg$',
40 # http://muz-tv.ru/play/7129/
41 # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true
42 'url': 'eagleplatform:media.clipyou.ru:12820',
43 'md5': '358597369cf8ba56675c1df15e7af624',
47 'title': "'O Sole Mio",
48 'thumbnail': r're:^https?://.*\.jpg$',
52 'skip': 'Georestricted',
56 def _extract_url(webpage):
57 # Regular iframe embedding
59 r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1',
62 return mobj.group('url')
65 src=(?P<qjs>["\'])(?:https?:)?//(?P<host>(?:(?!(?P=qjs)).)+\.media\.eagleplatform\.com)/player/player\.js(?P=qjs)
68 # "Basic usage" embedding (see http://dultonmedia.github.io/eplayer/)
73 class=(?P<qclass>["\'])eagleplayer(?P=qclass)[^>]+
74 data-id=["\'](?P<id>\d+)
75 ''' % PLAYER_JS_RE, webpage)
77 return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
78 # Generalization of "Javascript code usage", "Combined usage" and
79 # "Usage without attaching to DOM" embeddings (see
80 # http://dultonmedia.github.io/eplayer/)
90 \bid\s*:\s*["\']?(?P<id>\d+)
96 ''' % PLAYER_JS_RE, webpage)
98 return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
101 def _handle_error(response):
102 status = int_or_none(response.get('status', 200))
104 raise ExtractorError(' '.join(response['errors']), expected=True)
106 def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', *args, **kwargs):
108 response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
109 except ExtractorError as ee:
110 if isinstance(ee.cause, compat_HTTPError):
111 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
112 self._handle_error(response)
116 def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'):
117 return self._download_json(url_or_request, video_id, note)['data'][0]
119 def _real_extract(self, url):
120 mobj = re.match(self._VALID_URL, url)
121 host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
123 player_data = self._download_json(
124 'http://%s/api/player_data?id=%s' % (host, video_id), video_id)
126 media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
128 title = media['title']
129 description = media.get('description')
130 thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:')
131 duration = int_or_none(media.get('duration'))
132 view_count = int_or_none(media.get('views'))
134 age_restriction = media.get('age_restriction')
137 age_limit = 0 if age_restriction == 'allow_all' else 18
139 secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:')
143 m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
144 m3u8_formats = self._extract_m3u8_formats(
145 m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
146 m3u8_id='hls', fatal=False)
147 formats.extend(m3u8_formats)
149 m3u8_formats_dict = {}
150 for f in m3u8_formats:
151 if f.get('height') is not None:
152 m3u8_formats_dict[f['height']] = f
154 mp4_data = self._download_json(
155 # Secure mp4 URL is constructed according to Player.prototype.mp4 from
156 # http://lentaru.media.eagleplatform.com/player/player.js
157 re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4s', secure_m3u8),
158 video_id, 'Downloading mp4 JSON', fatal=False)
160 for format_id, format_url in mp4_data.get('data', {}).items():
161 if not isinstance(format_url, compat_str):
163 height = int_or_none(format_id)
164 if height is not None and m3u8_formats_dict.get(height):
165 f = m3u8_formats_dict[height].copy()
167 'format_id': f['format_id'].replace('hls', 'http'),
172 'format_id': 'http-%s' % format_id,
173 'height': int_or_none(format_id),
175 f['url'] = format_url
178 self._sort_formats(formats)
183 'description': description,
184 'thumbnail': thumbnail,
185 'duration': duration,
186 'view_count': view_count,
187 'age_limit': age_limit,