]>
Commit | Line | Data |
---|---|---|
1 | # coding: utf-8 | |
2 | from __future__ import unicode_literals | |
3 | ||
4 | import json | |
5 | import time | |
6 | import hmac | |
7 | import hashlib | |
8 | import itertools | |
9 | ||
10 | from .common import InfoExtractor | |
11 | from ..utils import ( | |
12 | ExtractorError, | |
13 | int_or_none, | |
14 | parse_age_limit, | |
15 | parse_iso8601, | |
16 | sanitized_Request, | |
17 | ) | |
18 | ||
19 | ||
20 | class VikiBaseIE(InfoExtractor): | |
21 | _VALID_URL_BASE = r'https?://(?:www\.)?viki\.(?:com|net|mx|jp|fr)/' | |
22 | _API_QUERY_TEMPLATE = '/v4/%sapp=%s&t=%s&site=www.viki.com' | |
23 | _API_URL_TEMPLATE = 'http://api.viki.io%s&sig=%s' | |
24 | ||
25 | _APP = '65535a' | |
26 | _APP_VERSION = '2.2.5.1428709186' | |
27 | _APP_SECRET = '-$iJ}@p7!G@SyU/je1bEyWg}upLu-6V6-Lg9VD(]siH,r.,m-r|ulZ,U4LC/SeR)' | |
28 | ||
29 | _NETRC_MACHINE = 'viki' | |
30 | ||
31 | _token = None | |
32 | ||
33 | _ERRORS = { | |
34 | 'geo': 'Sorry, this content is not available in your region.', | |
35 | 'upcoming': 'Sorry, this content is not yet available.', | |
36 | # 'paywall': 'paywall', | |
37 | } | |
38 | ||
39 | def _prepare_call(self, path, timestamp=None, post_data=None): | |
40 | path += '?' if '?' not in path else '&' | |
41 | if not timestamp: | |
42 | timestamp = int(time.time()) | |
43 | query = self._API_QUERY_TEMPLATE % (path, self._APP, timestamp) | |
44 | if self._token: | |
45 | query += '&token=%s' % self._token | |
46 | sig = hmac.new( | |
47 | self._APP_SECRET.encode('ascii'), | |
48 | query.encode('ascii'), | |
49 | hashlib.sha1 | |
50 | ).hexdigest() | |
51 | url = self._API_URL_TEMPLATE % (query, sig) | |
52 | return sanitized_Request( | |
53 | url, json.dumps(post_data).encode('utf-8')) if post_data else url | |
54 | ||
55 | def _call_api(self, path, video_id, note, timestamp=None, post_data=None): | |
56 | resp = self._download_json( | |
57 | self._prepare_call(path, timestamp, post_data), video_id, note) | |
58 | ||
59 | error = resp.get('error') | |
60 | if error: | |
61 | if error == 'invalid timestamp': | |
62 | resp = self._download_json( | |
63 | self._prepare_call(path, int(resp['current_timestamp']), post_data), | |
64 | video_id, '%s (retry)' % note) | |
65 | error = resp.get('error') | |
66 | if error: | |
67 | self._raise_error(resp['error']) | |
68 | ||
69 | return resp | |
70 | ||
71 | def _raise_error(self, error): | |
72 | raise ExtractorError( | |
73 | '%s returned error: %s' % (self.IE_NAME, error), | |
74 | expected=True) | |
75 | ||
76 | def _check_errors(self, data): | |
77 | for reason, status in data.get('blocking', {}).items(): | |
78 | if status and reason in self._ERRORS: | |
79 | raise ExtractorError('%s said: %s' % ( | |
80 | self.IE_NAME, self._ERRORS[reason]), expected=True) | |
81 | ||
82 | def _real_initialize(self): | |
83 | self._login() | |
84 | ||
85 | def _login(self): | |
86 | (username, password) = self._get_login_info() | |
87 | if username is None: | |
88 | return | |
89 | ||
90 | login_form = { | |
91 | 'login_id': username, | |
92 | 'password': password, | |
93 | } | |
94 | ||
95 | login = self._call_api( | |
96 | 'sessions.json', None, | |
97 | 'Logging in as %s' % username, post_data=login_form) | |
98 | ||
99 | self._token = login.get('token') | |
100 | if not self._token: | |
101 | self.report_warning('Unable to get session token, login has probably failed') | |
102 | ||
103 | @staticmethod | |
104 | def dict_selection(dict_obj, preferred_key): | |
105 | if preferred_key in dict_obj: | |
106 | return dict_obj.get(preferred_key) | |
107 | ||
108 | filtered_dict = list(filter(None, [dict_obj.get(k) for k in dict_obj.keys()])) | |
109 | return filtered_dict[0] if filtered_dict else None | |
110 | ||
111 | ||
112 | class VikiIE(VikiBaseIE): | |
113 | IE_NAME = 'viki' | |
114 | _VALID_URL = r'%s(?:videos|player)/(?P<id>[0-9]+v)' % VikiBaseIE._VALID_URL_BASE | |
115 | _TESTS = [{ | |
116 | 'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14', | |
117 | 'info_dict': { | |
118 | 'id': '1023585v', | |
119 | 'ext': 'mp4', | |
120 | 'title': 'Heirs Episode 14', | |
121 | 'uploader': 'SBS', | |
122 | 'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e', | |
123 | 'upload_date': '20131121', | |
124 | 'age_limit': 13, | |
125 | }, | |
126 | 'skip': 'Blocked in the US', | |
127 | }, { | |
128 | # clip | |
129 | 'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference', | |
130 | 'md5': '86c0b5dbd4d83a6611a79987cc7a1989', | |
131 | 'info_dict': { | |
132 | 'id': '1067139v', | |
133 | 'ext': 'mp4', | |
134 | 'title': "'The Avengers: Age of Ultron' Press Conference", | |
135 | 'description': 'md5:d70b2f9428f5488321bfe1db10d612ea', | |
136 | 'duration': 352, | |
137 | 'timestamp': 1430380829, | |
138 | 'upload_date': '20150430', | |
139 | 'uploader': 'Arirang TV', | |
140 | 'like_count': int, | |
141 | 'age_limit': 0, | |
142 | } | |
143 | }, { | |
144 | 'url': 'http://www.viki.com/videos/1048879v-ankhon-dekhi', | |
145 | 'info_dict': { | |
146 | 'id': '1048879v', | |
147 | 'ext': 'mp4', | |
148 | 'title': 'Ankhon Dekhi', | |
149 | 'duration': 6512, | |
150 | 'timestamp': 1408532356, | |
151 | 'upload_date': '20140820', | |
152 | 'uploader': 'Spuul', | |
153 | 'like_count': int, | |
154 | 'age_limit': 13, | |
155 | }, | |
156 | 'params': { | |
157 | # m3u8 download | |
158 | 'skip_download': True, | |
159 | } | |
160 | }, { | |
161 | # episode | |
162 | 'url': 'http://www.viki.com/videos/44699v-boys-over-flowers-episode-1', | |
163 | 'md5': '190f3ef426005ba3a080a63325955bc3', | |
164 | 'info_dict': { | |
165 | 'id': '44699v', | |
166 | 'ext': 'mp4', | |
167 | 'title': 'Boys Over Flowers - Episode 1', | |
168 | 'description': 'md5:52617e4f729c7d03bfd4bcbbb6e946f2', | |
169 | 'duration': 4155, | |
170 | 'timestamp': 1270496524, | |
171 | 'upload_date': '20100405', | |
172 | 'uploader': 'group8', | |
173 | 'like_count': int, | |
174 | 'age_limit': 13, | |
175 | } | |
176 | }, { | |
177 | # youtube external | |
178 | 'url': 'http://www.viki.com/videos/50562v-poor-nastya-complete-episode-1', | |
179 | 'md5': '216d1afdc0c64d1febc1e9f2bd4b864b', | |
180 | 'info_dict': { | |
181 | 'id': '50562v', | |
182 | 'ext': 'mp4', | |
183 | 'title': 'Poor Nastya [COMPLETE] - Episode 1', | |
184 | 'description': '', | |
185 | 'duration': 607, | |
186 | 'timestamp': 1274949505, | |
187 | 'upload_date': '20101213', | |
188 | 'uploader': 'ad14065n', | |
189 | 'uploader_id': 'ad14065n', | |
190 | 'like_count': int, | |
191 | 'age_limit': 13, | |
192 | } | |
193 | }, { | |
194 | 'url': 'http://www.viki.com/player/44699v', | |
195 | 'only_matching': True, | |
196 | }, { | |
197 | # non-English description | |
198 | 'url': 'http://www.viki.com/videos/158036v-love-in-magic', | |
199 | 'md5': '1713ae35df5a521b31f6dc40730e7c9c', | |
200 | 'info_dict': { | |
201 | 'id': '158036v', | |
202 | 'ext': 'mp4', | |
203 | 'uploader': 'I Planet Entertainment', | |
204 | 'upload_date': '20111122', | |
205 | 'timestamp': 1321985454, | |
206 | 'description': 'md5:44b1e46619df3a072294645c770cef36', | |
207 | 'title': 'Love In Magic', | |
208 | 'age_limit': 13, | |
209 | }, | |
210 | }] | |
211 | ||
212 | def _real_extract(self, url): | |
213 | video_id = self._match_id(url) | |
214 | ||
215 | video = self._call_api( | |
216 | 'videos/%s.json' % video_id, video_id, 'Downloading video JSON') | |
217 | ||
218 | self._check_errors(video) | |
219 | ||
220 | title = self.dict_selection(video.get('titles', {}), 'en') | |
221 | if not title: | |
222 | title = 'Episode %d' % video.get('number') if video.get('type') == 'episode' else video.get('id') or video_id | |
223 | container_titles = video.get('container', {}).get('titles', {}) | |
224 | container_title = self.dict_selection(container_titles, 'en') | |
225 | title = '%s - %s' % (container_title, title) | |
226 | ||
227 | description = self.dict_selection(video.get('descriptions', {}), 'en') | |
228 | ||
229 | duration = int_or_none(video.get('duration')) | |
230 | timestamp = parse_iso8601(video.get('created_at')) | |
231 | uploader = video.get('author') | |
232 | like_count = int_or_none(video.get('likes', {}).get('count')) | |
233 | age_limit = parse_age_limit(video.get('rating')) | |
234 | ||
235 | thumbnails = [] | |
236 | for thumbnail_id, thumbnail in video.get('images', {}).items(): | |
237 | thumbnails.append({ | |
238 | 'id': thumbnail_id, | |
239 | 'url': thumbnail.get('url'), | |
240 | }) | |
241 | ||
242 | subtitles = {} | |
243 | for subtitle_lang, _ in video.get('subtitle_completions', {}).items(): | |
244 | subtitles[subtitle_lang] = [{ | |
245 | 'ext': subtitles_format, | |
246 | 'url': self._prepare_call( | |
247 | 'videos/%s/subtitles/%s.%s' % (video_id, subtitle_lang, subtitles_format)), | |
248 | } for subtitles_format in ('srt', 'vtt')] | |
249 | ||
250 | result = { | |
251 | 'id': video_id, | |
252 | 'title': title, | |
253 | 'description': description, | |
254 | 'duration': duration, | |
255 | 'timestamp': timestamp, | |
256 | 'uploader': uploader, | |
257 | 'like_count': like_count, | |
258 | 'age_limit': age_limit, | |
259 | 'thumbnails': thumbnails, | |
260 | 'subtitles': subtitles, | |
261 | } | |
262 | ||
263 | streams = self._call_api( | |
264 | 'videos/%s/streams.json' % video_id, video_id, | |
265 | 'Downloading video streams JSON') | |
266 | ||
267 | if 'external' in streams: | |
268 | result.update({ | |
269 | '_type': 'url_transparent', | |
270 | 'url': streams['external']['url'], | |
271 | }) | |
272 | return result | |
273 | ||
274 | formats = [] | |
275 | for format_id, stream_dict in streams.items(): | |
276 | height = int_or_none(self._search_regex( | |
277 | r'^(\d+)[pP]$', format_id, 'height', default=None)) | |
278 | for protocol, format_dict in stream_dict.items(): | |
279 | if format_id == 'm3u8': | |
280 | formats.extend(self._extract_m3u8_formats( | |
281 | format_dict['url'], video_id, 'mp4', 'm3u8_native', | |
282 | m3u8_id='m3u8-%s' % protocol, fatal=False)) | |
283 | else: | |
284 | formats.append({ | |
285 | 'url': format_dict['url'], | |
286 | 'format_id': '%s-%s' % (format_id, protocol), | |
287 | 'height': height, | |
288 | }) | |
289 | self._sort_formats(formats) | |
290 | ||
291 | result['formats'] = formats | |
292 | return result | |
293 | ||
294 | ||
295 | class VikiChannelIE(VikiBaseIE): | |
296 | IE_NAME = 'viki:channel' | |
297 | _VALID_URL = r'%s(?:tv|news|movies|artists)/(?P<id>[0-9]+c)' % VikiBaseIE._VALID_URL_BASE | |
298 | _TESTS = [{ | |
299 | 'url': 'http://www.viki.com/tv/50c-boys-over-flowers', | |
300 | 'info_dict': { | |
301 | 'id': '50c', | |
302 | 'title': 'Boys Over Flowers', | |
303 | 'description': 'md5:ecd3cff47967fe193cff37c0bec52790', | |
304 | }, | |
305 | 'playlist_count': 70, | |
306 | }, { | |
307 | 'url': 'http://www.viki.com/tv/1354c-poor-nastya-complete', | |
308 | 'info_dict': { | |
309 | 'id': '1354c', | |
310 | 'title': 'Poor Nastya [COMPLETE]', | |
311 | 'description': 'md5:05bf5471385aa8b21c18ad450e350525', | |
312 | }, | |
313 | 'playlist_count': 127, | |
314 | }, { | |
315 | 'url': 'http://www.viki.com/news/24569c-showbiz-korea', | |
316 | 'only_matching': True, | |
317 | }, { | |
318 | 'url': 'http://www.viki.com/movies/22047c-pride-and-prejudice-2005', | |
319 | 'only_matching': True, | |
320 | }, { | |
321 | 'url': 'http://www.viki.com/artists/2141c-shinee', | |
322 | 'only_matching': True, | |
323 | }] | |
324 | ||
325 | _PER_PAGE = 25 | |
326 | ||
327 | def _real_extract(self, url): | |
328 | channel_id = self._match_id(url) | |
329 | ||
330 | channel = self._call_api( | |
331 | 'containers/%s.json' % channel_id, channel_id, | |
332 | 'Downloading channel JSON') | |
333 | ||
334 | self._check_errors(channel) | |
335 | ||
336 | title = self.dict_selection(channel['titles'], 'en') | |
337 | ||
338 | description = self.dict_selection(channel['descriptions'], 'en') | |
339 | ||
340 | entries = [] | |
341 | for video_type in ('episodes', 'clips', 'movies'): | |
342 | for page_num in itertools.count(1): | |
343 | page = self._call_api( | |
344 | 'containers/%s/%s.json?per_page=%d&sort=number&direction=asc&with_paging=true&page=%d' | |
345 | % (channel_id, video_type, self._PER_PAGE, page_num), channel_id, | |
346 | 'Downloading %s JSON page #%d' % (video_type, page_num)) | |
347 | for video in page['response']: | |
348 | video_id = video['id'] | |
349 | entries.append(self.url_result( | |
350 | 'http://www.viki.com/videos/%s' % video_id, 'Viki')) | |
351 | if not page['pagination']['next']: | |
352 | break | |
353 | ||
354 | return self.playlist_result(entries, channel_id, title, description) |