]>
jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/viki.py
2 from __future__
import unicode_literals
11 from .common
import InfoExtractor
21 class VikiBaseIE(InfoExtractor
):
22 _VALID_URL_BASE
= r
'https?://(?:www\.)?viki\.(?:com|net|mx|jp|fr)/'
23 _API_QUERY_TEMPLATE
= '/v4/%sapp=%s&t=%s&site=www.viki.com'
24 _API_URL_TEMPLATE
= 'http://api.viki.io%s&sig=%s'
27 _APP_VERSION
= '2.2.5.1428709186'
28 _APP_SECRET
= '-$iJ}@p7!G@SyU/je1bEyWg}upLu-6V6-Lg9VD(]siH,r.,m-r|ulZ,U4LC/SeR)'
31 _NETRC_MACHINE
= 'viki'
36 'geo': 'Sorry, this content is not available in your region.',
37 'upcoming': 'Sorry, this content is not yet available.',
38 # 'paywall': 'paywall',
41 def _prepare_call(self
, path
, timestamp
=None, post_data
=None):
42 path
+= '?' if '?' not in path
else '&'
44 timestamp
= int(time
.time())
45 query
= self
._API
_QUERY
_TEMPLATE
% (path
, self
._APP
, timestamp
)
47 query
+= '&token=%s' % self
._token
49 self
._APP
_SECRET
.encode('ascii'),
50 query
.encode('ascii'),
53 url
= self
._API
_URL
_TEMPLATE
% (query
, sig
)
54 return sanitized_Request(
55 url
, json
.dumps(post_data
).encode('utf-8')) if post_data
else url
57 def _call_api(self
, path
, video_id
, note
, timestamp
=None, post_data
=None):
58 resp
= self
._download
_json
(
59 self
._prepare
_call
(path
, timestamp
, post_data
), video_id
, note
)
61 error
= resp
.get('error')
63 if error
== 'invalid timestamp':
64 resp
= self
._download
_json
(
65 self
._prepare
_call
(path
, int(resp
['current_timestamp']), post_data
),
66 video_id
, '%s (retry)' % note
)
67 error
= resp
.get('error')
69 self
._raise
_error
(resp
['error'])
73 def _raise_error(self
, error
):
75 '%s returned error: %s' % (self
.IE_NAME
, error
),
78 def _check_errors(self
, data
):
79 for reason
, status
in data
.get('blocking', {}).items():
80 if status
and reason
in self
._ERRORS
:
81 message
= self
._ERRORS
[reason
]
83 self
.raise_geo_restricted(msg
=message
)
84 raise ExtractorError('%s said: %s' % (
85 self
.IE_NAME
, message
), expected
=True)
87 def _real_initialize(self
):
91 (username
, password
) = self
._get
_login
_info
()
100 login
= self
._call
_api
(
101 'sessions.json', None,
102 'Logging in as %s' % username
, post_data
=login_form
)
104 self
._token
= login
.get('token')
106 self
.report_warning('Unable to get session token, login has probably failed')
109 def dict_selection(dict_obj
, preferred_key
, allow_fallback
=True):
110 if preferred_key
in dict_obj
:
111 return dict_obj
.get(preferred_key
)
113 if not allow_fallback
:
116 filtered_dict
= list(filter(None, [dict_obj
.get(k
) for k
in dict_obj
.keys()]))
117 return filtered_dict
[0] if filtered_dict
else None
120 class VikiIE(VikiBaseIE
):
122 _VALID_URL
= r
'%s(?:videos|player)/(?P<id>[0-9]+v)' % VikiBaseIE
._VALID
_URL
_BASE
124 'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
128 'title': 'Heirs Episode 14',
130 'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e',
131 'upload_date': '20131121',
134 'skip': 'Blocked in the US',
137 'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
138 'md5': '86c0b5dbd4d83a6611a79987cc7a1989',
142 'title': "'The Avengers: Age of Ultron' Press Conference",
143 'description': 'md5:d70b2f9428f5488321bfe1db10d612ea',
145 'timestamp': 1430380829,
146 'upload_date': '20150430',
147 'uploader': 'Arirang TV',
152 'url': 'http://www.viki.com/videos/1048879v-ankhon-dekhi',
156 'title': 'Ankhon Dekhi',
158 'timestamp': 1408532356,
159 'upload_date': '20140820',
164 'skip': 'Blocked in the US',
167 'url': 'http://www.viki.com/videos/44699v-boys-over-flowers-episode-1',
168 'md5': '5fa476a902e902783ac7a4d615cdbc7a',
172 'title': 'Boys Over Flowers - Episode 1',
173 'description': 'md5:b89cf50038b480b88b5b3c93589a9076',
175 'timestamp': 1270496524,
176 'upload_date': '20100405',
177 'uploader': 'group8',
183 'url': 'http://www.viki.com/videos/50562v-poor-nastya-complete-episode-1',
184 'md5': '63f8600c1da6f01b7640eee7eca4f1da',
188 'title': 'Poor Nastya [COMPLETE] - Episode 1',
191 'timestamp': 1274949505,
192 'upload_date': '20101213',
193 'uploader': 'ad14065n',
194 'uploader_id': 'ad14065n',
199 'url': 'http://www.viki.com/player/44699v',
200 'only_matching': True,
202 # non-English description
203 'url': 'http://www.viki.com/videos/158036v-love-in-magic',
204 'md5': '1713ae35df5a521b31f6dc40730e7c9c',
208 'uploader': 'I Planet Entertainment',
209 'upload_date': '20111122',
210 'timestamp': 1321985454,
211 'description': 'md5:44b1e46619df3a072294645c770cef36',
212 'title': 'Love In Magic',
217 def _real_extract(self
, url
):
218 video_id
= self
._match
_id
(url
)
220 video
= self
._call
_api
(
221 'videos/%s.json' % video_id
, video_id
, 'Downloading video JSON')
223 self
._check
_errors
(video
)
225 title
= self
.dict_selection(video
.get('titles', {}), 'en', allow_fallback
=False)
227 title
= 'Episode %d' % video
.get('number') if video
.get('type') == 'episode' else video
.get('id') or video_id
228 container_titles
= video
.get('container', {}).get('titles', {}
)
229 container_title
= self
.dict_selection(container_titles
, 'en')
230 title
= '%s - %s' % (container_title
, title
)
232 description
= self
.dict_selection(video
.get('descriptions', {}), 'en')
234 duration
= int_or_none(video
.get('duration'))
235 timestamp
= parse_iso8601(video
.get('created_at'))
236 uploader
= video
.get('author')
237 like_count
= int_or_none(video
.get('likes', {}).get('count'))
238 age_limit
= parse_age_limit(video
.get('rating'))
241 for thumbnail_id
, thumbnail
in video
.get('images', {}).items():
244 'url': thumbnail
.get('url'),
248 for subtitle_lang
, _
in video
.get('subtitle_completions', {}).items():
249 subtitles
[subtitle_lang
] = [{
250 'ext': subtitles_format
,
251 'url': self
._prepare
_call
(
252 'videos/%s/subtitles/%s.%s' % (video_id
, subtitle_lang
, subtitles_format
)),
253 } for subtitles_format
in ('srt', 'vtt')]
258 'description': description
,
259 'duration': duration
,
260 'timestamp': timestamp
,
261 'uploader': uploader
,
262 'like_count': like_count
,
263 'age_limit': age_limit
,
264 'thumbnails': thumbnails
,
265 'subtitles': subtitles
,
268 streams
= self
._call
_api
(
269 'videos/%s/streams.json' % video_id
, video_id
,
270 'Downloading video streams JSON')
272 if 'external' in streams
:
274 '_type': 'url_transparent',
275 'url': streams
['external']['url'],
280 for format_id
, stream_dict
in streams
.items():
281 height
= int_or_none(self
._search
_regex
(
282 r
'^(\d+)[pP]$', format_id
, 'height', default
=None))
283 for protocol
, format_dict
in stream_dict
.items():
284 # rtmps URLs does not seem to work
285 if protocol
== 'rtmps':
287 format_url
= format_dict
['url']
288 if format_id
== 'm3u8':
289 m3u8_formats
= self
._extract
_m
3u8_formats
(
290 format_url
, video_id
, 'mp4',
291 entry_protocol
='m3u8_native',
292 m3u8_id
='m3u8-%s' % protocol
, fatal
=False)
293 # Despite CODECS metadata in m3u8 all video-only formats
294 # are actually video+audio
295 for f
in m3u8_formats
:
296 if f
.get('acodec') == 'none' and f
.get('vcodec') != 'none':
298 formats
.extend(m3u8_formats
)
299 elif format_url
.startswith('rtmp'):
301 r
'^(?P<url>rtmp://[^/]+/(?P<app>.+?))/(?P<playpath>mp4:.+)$',
306 'format_id': 'rtmp-%s' % format_id
,
308 'url': mobj
.group('url'),
309 'play_path': mobj
.group('playpath'),
310 'app': mobj
.group('app'),
316 'format_id': '%s-%s' % (format_id
, protocol
),
319 self
._sort
_formats
(formats
)
321 result
['formats'] = formats
325 class VikiChannelIE(VikiBaseIE
):
326 IE_NAME
= 'viki:channel'
327 _VALID_URL
= r
'%s(?:tv|news|movies|artists)/(?P<id>[0-9]+c)' % VikiBaseIE
._VALID
_URL
_BASE
329 'url': 'http://www.viki.com/tv/50c-boys-over-flowers',
332 'title': 'Boys Over Flowers',
333 'description': 'md5:ecd3cff47967fe193cff37c0bec52790',
335 'playlist_mincount': 71,
337 'url': 'http://www.viki.com/tv/1354c-poor-nastya-complete',
340 'title': 'Poor Nastya [COMPLETE]',
341 'description': 'md5:05bf5471385aa8b21c18ad450e350525',
343 'playlist_count': 127,
345 'url': 'http://www.viki.com/news/24569c-showbiz-korea',
346 'only_matching': True,
348 'url': 'http://www.viki.com/movies/22047c-pride-and-prejudice-2005',
349 'only_matching': True,
351 'url': 'http://www.viki.com/artists/2141c-shinee',
352 'only_matching': True,
357 def _real_extract(self
, url
):
358 channel_id
= self
._match
_id
(url
)
360 channel
= self
._call
_api
(
361 'containers/%s.json' % channel_id
, channel_id
,
362 'Downloading channel JSON')
364 self
._check
_errors
(channel
)
366 title
= self
.dict_selection(channel
['titles'], 'en')
368 description
= self
.dict_selection(channel
['descriptions'], 'en')
371 for video_type
in ('episodes', 'clips', 'movies'):
372 for page_num
in itertools
.count(1):
373 page
= self
._call
_api
(
374 'containers/%s/%s.json?per_page=%d&sort=number&direction=asc&with_paging=true&page=%d'
375 % (channel_id
, video_type
, self
._PER
_PAGE
, page_num
), channel_id
,
376 'Downloading %s JSON page #%d' % (video_type
, page_num
))
377 for video
in page
['response']:
378 video_id
= video
['id']
379 entries
.append(self
.url_result(
380 'http://www.viki.com/videos/%s' % video_id
, 'Viki'))
381 if not page
['pagination']['next']:
384 return self
.playlist_result(entries
, channel_id
, title
, description
)