]>
Commit | Line | Data |
---|---|---|
dcdb292f | 1 | # coding: utf-8 |
214c22c7 | 2 | from __future__ import unicode_literals |
52ad14ae | 3 | |
fb198a8a | 4 | import datetime |
f494ddad | 5 | import functools |
7bdcb4a4 | 6 | import itertools |
abafce59 | 7 | import json |
8 | import re | |
7bdcb4a4 | 9 | import time |
52ad14ae | 10 | |
abafce59 | 11 | from .common import InfoExtractor, SearchInfoExtractor |
1cc79574 | 12 | from ..compat import ( |
bb139491 | 13 | compat_parse_qs, |
29f7c58a | 14 | compat_urllib_parse_urlparse, |
f494ddad | 15 | compat_HTTPError, |
1cc79574 PH |
16 | ) |
17 | from ..utils import ( | |
6110bbbf | 18 | ExtractorError, |
7bdcb4a4 LNO |
19 | OnDemandPagedList, |
20 | bug_reports_message, | |
21 | clean_html, | |
fb198a8a | 22 | float_or_none, |
b868936c | 23 | int_or_none, |
7bdcb4a4 | 24 | join_nonempty, |
1cc79574 | 25 | parse_duration, |
7bdcb4a4 | 26 | parse_filesize, |
bb865f3a | 27 | parse_iso8601, |
ee6a6116 | 28 | remove_start, |
f494ddad | 29 | traverse_obj, |
463e7216 | 30 | try_get, |
f494ddad | 31 | unescapeHTML, |
7bdcb4a4 LNO |
32 | update_url_query, |
33 | url_or_none, | |
6e6bc8da | 34 | urlencode_postdata, |
52ad14ae TT |
35 | ) |
36 | ||
13ebea79 | 37 | |
52ad14ae | 38 | class NiconicoIE(InfoExtractor): |
214c22c7 JMF |
39 | IE_NAME = 'niconico' |
40 | IE_DESC = 'ニコニコ動画' | |
52ad14ae | 41 | |
1c9a1457 | 42 | _TESTS = [{ |
214c22c7 | 43 | 'url': 'http://www.nicovideo.jp/watch/sm22312215', |
7bdcb4a4 | 44 | 'md5': 'd1a75c0823e2f629128c43e1212760f9', |
214c22c7 JMF |
45 | 'info_dict': { |
46 | 'id': 'sm22312215', | |
47 | 'ext': 'mp4', | |
48 | 'title': 'Big Buck Bunny', | |
463e7216 | 49 | 'thumbnail': r're:https?://.*', |
214c22c7 JMF |
50 | 'uploader': 'takuya0301', |
51 | 'uploader_id': '2698420', | |
52 | 'upload_date': '20131123', | |
aaab8c5e | 53 | 'timestamp': int, # timestamp is unstable |
214c22c7 | 54 | 'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org', |
15ce1338 | 55 | 'duration': 33, |
463e7216 YCH |
56 | 'view_count': int, |
57 | 'comment_count': int, | |
52ad14ae | 58 | }, |
8e4988f1 | 59 | 'skip': 'Requires an account', |
1c9a1457 | 60 | }, { |
59d814f7 YCH |
61 | # File downloaded with and without credentials are different, so omit |
62 | # the md5 field | |
1c9a1457 | 63 | 'url': 'http://www.nicovideo.jp/watch/nm14296458', |
1c9a1457 S |
64 | 'info_dict': { |
65 | 'id': 'nm14296458', | |
66 | 'ext': 'swf', | |
67 | 'title': '【鏡音リン】Dance on media【オリジナル】take2!', | |
bb865f3a | 68 | 'description': 'md5:689f066d74610b3b22e0f1739add0f58', |
463e7216 | 69 | 'thumbnail': r're:https?://.*', |
1c9a1457 S |
70 | 'uploader': 'りょうた', |
71 | 'uploader_id': '18822557', | |
72 | 'upload_date': '20110429', | |
bb865f3a | 73 | 'timestamp': 1304065916, |
1c9a1457 S |
74 | 'duration': 209, |
75 | }, | |
8e4988f1 | 76 | 'skip': 'Requires an account', |
bb865f3a YCH |
77 | }, { |
78 | # 'video exists but is marked as "deleted" | |
b2e8e7da | 79 | # md5 is unstable |
bb865f3a | 80 | 'url': 'http://www.nicovideo.jp/watch/sm10000', |
bb865f3a YCH |
81 | 'info_dict': { |
82 | 'id': 'sm10000', | |
83 | 'ext': 'unknown_video', | |
84 | 'description': 'deleted', | |
85 | 'title': 'ドラえもんエターナル第3話「決戦第3新東京市」<前編>', | |
463e7216 | 86 | 'thumbnail': r're:https?://.*', |
b2e8e7da | 87 | 'upload_date': '20071224', |
8e4988f1 | 88 | 'timestamp': int, # timestamp field has different value if logged in |
b2e8e7da | 89 | 'duration': 304, |
463e7216 | 90 | 'view_count': int, |
bb865f3a | 91 | }, |
8e4988f1 | 92 | 'skip': 'Requires an account', |
621ffe7b YCH |
93 | }, { |
94 | 'url': 'http://www.nicovideo.jp/watch/so22543406', | |
95 | 'info_dict': { | |
96 | 'id': '1388129933', | |
97 | 'ext': 'mp4', | |
98 | 'title': '【第1回】RADIOアニメロミックス ラブライブ!~のぞえりRadio Garden~', | |
99 | 'description': 'md5:b27d224bb0ff53d3c8269e9f8b561cf1', | |
463e7216 | 100 | 'thumbnail': r're:https?://.*', |
621ffe7b YCH |
101 | 'timestamp': 1388851200, |
102 | 'upload_date': '20140104', | |
103 | 'uploader': 'アニメロチャンネル', | |
104 | 'uploader_id': '312', | |
8e4988f1 YCH |
105 | }, |
106 | 'skip': 'The viewing period of the video you were searching for has expired.', | |
463e7216 | 107 | }, { |
ee6a6116 | 108 | # video not available via `getflv`; "old" HTML5 video |
463e7216 | 109 | 'url': 'http://www.nicovideo.jp/watch/sm1151009', |
ee6a6116 | 110 | 'md5': '8fa81c364eb619d4085354eab075598a', |
463e7216 YCH |
111 | 'info_dict': { |
112 | 'id': 'sm1151009', | |
ee6a6116 | 113 | 'ext': 'mp4', |
463e7216 YCH |
114 | 'title': 'マスターシステム本体内蔵のスペハリのメインテーマ(PSG版)', |
115 | 'description': 'md5:6ee077e0581ff5019773e2e714cdd0b7', | |
116 | 'thumbnail': r're:https?://.*', | |
117 | 'duration': 184, | |
118 | 'timestamp': 1190868283, | |
119 | 'upload_date': '20070927', | |
120 | 'uploader': 'denden2', | |
121 | 'uploader_id': '1392194', | |
122 | 'view_count': int, | |
123 | 'comment_count': int, | |
124 | }, | |
125 | 'skip': 'Requires an account', | |
ee6a6116 YCH |
126 | }, { |
127 | # "New" HTML5 video | |
aaab8c5e | 128 | # md5 is unstable |
ee6a6116 | 129 | 'url': 'http://www.nicovideo.jp/watch/sm31464864', |
ee6a6116 YCH |
130 | 'info_dict': { |
131 | 'id': 'sm31464864', | |
132 | 'ext': 'mp4', | |
133 | 'title': '新作TVアニメ「戦姫絶唱シンフォギアAXZ」PV 最高画質', | |
134 | 'description': 'md5:e52974af9a96e739196b2c1ca72b5feb', | |
135 | 'timestamp': 1498514060, | |
136 | 'upload_date': '20170626', | |
aaab8c5e | 137 | 'uploader': 'ゲスト', |
ee6a6116 YCH |
138 | 'uploader_id': '40826363', |
139 | 'thumbnail': r're:https?://.*', | |
140 | 'duration': 198, | |
141 | 'view_count': int, | |
142 | 'comment_count': int, | |
143 | }, | |
144 | 'skip': 'Requires an account', | |
aaab8c5e PP |
145 | }, { |
146 | # Video without owner | |
147 | 'url': 'http://www.nicovideo.jp/watch/sm18238488', | |
148 | 'md5': 'd265680a1f92bdcbbd2a507fc9e78a9e', | |
149 | 'info_dict': { | |
150 | 'id': 'sm18238488', | |
151 | 'ext': 'mp4', | |
152 | 'title': '【実写版】ミュータントタートルズ', | |
153 | 'description': 'md5:15df8988e47a86f9e978af2064bf6d8e', | |
154 | 'timestamp': 1341160408, | |
155 | 'upload_date': '20120701', | |
156 | 'uploader': None, | |
157 | 'uploader_id': None, | |
158 | 'thumbnail': r're:https?://.*', | |
159 | 'duration': 5271, | |
160 | 'view_count': int, | |
161 | 'comment_count': int, | |
162 | }, | |
163 | 'skip': 'Requires an account', | |
4a87de72 LS |
164 | }, { |
165 | 'url': 'http://sp.nicovideo.jp/watch/sm28964488?ss_pos=1&cp_in=wt_tg', | |
166 | 'only_matching': True, | |
7bdcb4a4 LNO |
167 | }, { |
168 | 'note': 'a video that is only served as an ENCRYPTED HLS.', | |
169 | 'url': 'https://www.nicovideo.jp/watch/so38016254', | |
170 | 'only_matching': True, | |
1c9a1457 | 171 | }] |
52ad14ae | 172 | |
7bdcb4a4 | 173 | _VALID_URL = r'https?://(?:(?:www\.|secure\.|sp\.)?nicovideo\.jp/watch|nico\.ms)/(?P<id>(?:[a-z]{2})?[0-9]+)' |
52ad14ae | 174 | _NETRC_MACHINE = 'niconico' |
7bdcb4a4 LNO |
175 | _COMMENT_API_ENDPOINTS = ( |
176 | 'https://nvcomment.nicovideo.jp/legacy/api.json', | |
177 | 'https://nmsg.nicovideo.jp/api.json',) | |
2291dbce | 178 | _API_HEADERS = { |
179 | 'X-Frontend-ID': '6', | |
7bdcb4a4 LNO |
180 | 'X-Frontend-Version': '0', |
181 | 'X-Niconico-Language': 'en-us', | |
182 | 'Referer': 'https://www.nicovideo.jp/', | |
183 | 'Origin': 'https://www.nicovideo.jp', | |
2291dbce | 184 | } |
185 | ||
52ad14ae | 186 | def _real_initialize(self): |
23d83ad4 | 187 | self._login() |
52ad14ae TT |
188 | |
189 | def _login(self): | |
68217024 | 190 | username, password = self._get_login_info() |
23d83ad4 NJ |
191 | # No authentication to be performed |
192 | if not username: | |
193 | return True | |
52ad14ae TT |
194 | |
195 | # Log in | |
bb139491 | 196 | login_ok = True |
52ad14ae | 197 | login_form_strs = { |
bb139491 | 198 | 'mail_tel': username, |
214c22c7 | 199 | 'password': password, |
52ad14ae | 200 | } |
7bdcb4a4 LNO |
201 | self._request_webpage( |
202 | 'https://account.nicovideo.jp/login', None, | |
203 | note='Acquiring Login session') | |
bb139491 | 204 | urlh = self._request_webpage( |
7bdcb4a4 | 205 | 'https://account.nicovideo.jp/login/redirector?show_button_twitter=1&site=niconico&show_button_facebook=1', None, |
bb139491 | 206 | note='Logging in', errnote='Unable to log in', |
7bdcb4a4 LNO |
207 | data=urlencode_postdata(login_form_strs), |
208 | headers={ | |
209 | 'Referer': 'https://account.nicovideo.jp/login', | |
210 | 'Content-Type': 'application/x-www-form-urlencoded', | |
211 | }) | |
bb139491 YCH |
212 | if urlh is False: |
213 | login_ok = False | |
214 | else: | |
29f7c58a | 215 | parts = compat_urllib_parse_urlparse(urlh.geturl()) |
bb139491 YCH |
216 | if compat_parse_qs(parts.query).get('message', [None])[0] == 'cant_login': |
217 | login_ok = False | |
218 | if not login_ok: | |
6a39ee13 | 219 | self.report_warning('unable to log in: bad username or password') |
bb139491 | 220 | return login_ok |
52ad14ae | 221 | |
fb198a8a | 222 | def _get_heartbeat_info(self, info_dict): |
fb198a8a | 223 | video_id, video_src_id, audio_src_id = info_dict['url'].split(':')[1].split('/') |
7bdcb4a4 | 224 | dmc_protocol = info_dict['_expected_protocol'] |
ee6a6116 | 225 | |
2291dbce | 226 | api_data = ( |
227 | info_dict.get('_api_data') | |
228 | or self._parse_json( | |
229 | self._html_search_regex( | |
230 | 'data-api-data="([^"]+)"', | |
231 | self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id), | |
232 | 'API data', default='{}'), | |
233 | video_id)) | |
fb198a8a | 234 | |
7978e172 | 235 | session_api_data = try_get(api_data, lambda x: x['media']['delivery']['movie']['session']) |
fb198a8a | 236 | session_api_endpoint = try_get(session_api_data, lambda x: x['urls'][0]) |
237 | ||
2291dbce | 238 | def ping(): |
7bdcb4a4 LNO |
239 | tracking_id = traverse_obj(api_data, ('media', 'delivery', 'trackingId')) |
240 | if tracking_id: | |
241 | tracking_url = update_url_query('https://nvapi.nicovideo.jp/v1/2ab0cbaa/watch', {'t': tracking_id}) | |
242 | watch_request_response = self._download_json( | |
243 | tracking_url, video_id, | |
244 | note='Acquiring permission for downloading video', fatal=False, | |
245 | headers=self._API_HEADERS) | |
246 | if traverse_obj(watch_request_response, ('meta', 'status')) != 200: | |
247 | self.report_warning('Failed to acquire permission for playing video. Video download may fail.') | |
fb198a8a | 248 | |
249 | yesno = lambda x: 'yes' if x else 'no' | |
250 | ||
7bdcb4a4 LNO |
251 | if dmc_protocol == 'http': |
252 | protocol = 'http' | |
253 | protocol_parameters = { | |
254 | 'http_output_download_parameters': { | |
255 | 'use_ssl': yesno(session_api_data['urls'][0]['isSsl']), | |
256 | 'use_well_known_port': yesno(session_api_data['urls'][0]['isWellKnownPort']), | |
257 | } | |
258 | } | |
259 | elif dmc_protocol == 'hls': | |
fb198a8a | 260 | protocol = 'm3u8' |
7bdcb4a4 LNO |
261 | segment_duration = try_get(self._configuration_arg('segment_duration'), lambda x: int(x[0])) or 6000 |
262 | parsed_token = self._parse_json(session_api_data['token'], video_id) | |
263 | encryption = traverse_obj(api_data, ('media', 'delivery', 'encryption')) | |
264 | protocol_parameters = { | |
265 | 'hls_parameters': { | |
266 | 'segment_duration': segment_duration, | |
267 | 'transfer_preset': '', | |
268 | 'use_ssl': yesno(session_api_data['urls'][0]['isSsl']), | |
269 | 'use_well_known_port': yesno(session_api_data['urls'][0]['isWellKnownPort']), | |
fb198a8a | 270 | } |
271 | } | |
7bdcb4a4 LNO |
272 | if 'hls_encryption' in parsed_token and encryption: |
273 | protocol_parameters['hls_parameters']['encryption'] = { | |
274 | parsed_token['hls_encryption']: { | |
275 | 'encrypted_key': encryption['encryptedKey'], | |
276 | 'key_uri': encryption['keyUri'], | |
fb198a8a | 277 | } |
278 | } | |
7bdcb4a4 LNO |
279 | else: |
280 | protocol = 'm3u8_native' | |
281 | else: | |
282 | raise ExtractorError(f'Unsupported DMC protocol: {dmc_protocol}') | |
ee6a6116 YCH |
283 | |
284 | session_response = self._download_json( | |
285 | session_api_endpoint['url'], video_id, | |
286 | query={'_format': 'json'}, | |
287 | headers={'Content-Type': 'application/json'}, | |
fb198a8a | 288 | note='Downloading JSON metadata for %s' % info_dict['format_id'], |
ee6a6116 YCH |
289 | data=json.dumps({ |
290 | 'session': { | |
291 | 'client_info': { | |
7978e172 | 292 | 'player_id': session_api_data.get('playerId'), |
ee6a6116 YCH |
293 | }, |
294 | 'content_auth': { | |
7978e172 | 295 | 'auth_type': try_get(session_api_data, lambda x: x['authTypes'][session_api_data['protocols'][0]]), |
296 | 'content_key_timeout': session_api_data.get('contentKeyTimeout'), | |
ee6a6116 | 297 | 'service_id': 'nicovideo', |
7978e172 | 298 | 'service_user_id': session_api_data.get('serviceUserId') |
ee6a6116 | 299 | }, |
7978e172 | 300 | 'content_id': session_api_data.get('contentId'), |
ee6a6116 YCH |
301 | 'content_src_id_sets': [{ |
302 | 'content_src_ids': [{ | |
303 | 'src_id_to_mux': { | |
fb198a8a | 304 | 'audio_src_ids': [audio_src_id], |
305 | 'video_src_ids': [video_src_id], | |
ee6a6116 YCH |
306 | } |
307 | }] | |
308 | }], | |
309 | 'content_type': 'movie', | |
310 | 'content_uri': '', | |
311 | 'keep_method': { | |
312 | 'heartbeat': { | |
7978e172 | 313 | 'lifetime': session_api_data.get('heartbeatLifetime') |
ee6a6116 YCH |
314 | } |
315 | }, | |
7bdcb4a4 | 316 | 'priority': session_api_data['priority'], |
ee6a6116 YCH |
317 | 'protocol': { |
318 | 'name': 'http', | |
319 | 'parameters': { | |
7bdcb4a4 LNO |
320 | 'http_parameters': { |
321 | 'parameters': protocol_parameters | |
322 | } | |
ee6a6116 YCH |
323 | } |
324 | }, | |
7978e172 | 325 | 'recipe_id': session_api_data.get('recipeId'), |
ee6a6116 YCH |
326 | 'session_operation_auth': { |
327 | 'session_operation_auth_by_signature': { | |
fb198a8a | 328 | 'signature': session_api_data.get('signature'), |
329 | 'token': session_api_data.get('token'), | |
ee6a6116 YCH |
330 | } |
331 | }, | |
332 | 'timing_constraint': 'unlimited' | |
333 | } | |
4d59db5b | 334 | }).encode()) |
ee6a6116 | 335 | |
fb198a8a | 336 | info_dict['url'] = session_response['data']['session']['content_uri'] |
337 | info_dict['protocol'] = protocol | |
338 | ||
339 | # get heartbeat info | |
340 | heartbeat_info_dict = { | |
341 | 'url': session_api_endpoint['url'] + '/' + session_response['data']['session']['id'] + '?_format=json&_method=PUT', | |
342 | 'data': json.dumps(session_response['data']), | |
343 | # interval, convert milliseconds to seconds, then halve to make a buffer. | |
2291dbce | 344 | 'interval': float_or_none(session_api_data.get('heartbeatLifetime'), scale=3000), |
345 | 'ping': ping | |
fb198a8a | 346 | } |
347 | ||
348 | return info_dict, heartbeat_info_dict | |
349 | ||
7bdcb4a4 LNO |
350 | def _extract_format_for_quality(self, video_id, audio_quality, video_quality, dmc_protocol): |
351 | ||
352 | if not audio_quality.get('isAvailable') or not video_quality.get('isAvailable'): | |
353 | return None | |
354 | ||
355 | def extract_video_quality(video_quality): | |
356 | return parse_filesize('%sB' % self._search_regex( | |
357 | r'\| ([0-9]*\.?[0-9]*[MK])', video_quality, 'vbr', default='')) | |
358 | ||
359 | format_id = '-'.join( | |
360 | [remove_start(s['id'], 'archive_') for s in (video_quality, audio_quality)] + [dmc_protocol]) | |
361 | ||
362 | vid_qual_label = traverse_obj(video_quality, ('metadata', 'label')) | |
363 | vid_quality = traverse_obj(video_quality, ('metadata', 'bitrate')) | |
ee6a6116 YCH |
364 | |
365 | return { | |
7bdcb4a4 | 366 | 'url': 'niconico_dmc:%s/%s/%s' % (video_id, video_quality['id'], audio_quality['id']), |
ee6a6116 | 367 | 'format_id': format_id, |
7bdcb4a4 | 368 | 'format_note': join_nonempty('DMC', vid_qual_label, dmc_protocol.upper(), delim=' '), |
ee6a6116 | 369 | 'ext': 'mp4', # Session API are used in HTML5, which always serves mp4 |
7bdcb4a4 LNO |
370 | 'acodec': 'aac', |
371 | 'vcodec': 'h264', | |
372 | 'abr': float_or_none(traverse_obj(audio_quality, ('metadata', 'bitrate')), 1000), | |
373 | 'vbr': float_or_none(vid_quality if vid_quality > 0 else extract_video_quality(vid_qual_label), 1000), | |
374 | 'height': traverse_obj(video_quality, ('metadata', 'resolution', 'height')), | |
375 | 'width': traverse_obj(video_quality, ('metadata', 'resolution', 'width')), | |
376 | 'quality': -2 if 'low' in video_quality['id'] else None, | |
377 | 'protocol': 'niconico_dmc', | |
378 | '_expected_protocol': dmc_protocol, | |
fb198a8a | 379 | 'http_headers': { |
380 | 'Origin': 'https://www.nicovideo.jp', | |
381 | 'Referer': 'https://www.nicovideo.jp/watch/' + video_id, | |
382 | } | |
ee6a6116 YCH |
383 | } |
384 | ||
52ad14ae | 385 | def _real_extract(self, url): |
937daef4 | 386 | video_id = self._match_id(url) |
52ad14ae | 387 | |
7bdcb4a4 LNO |
388 | try: |
389 | webpage, handle = self._download_webpage_handle( | |
390 | 'http://www.nicovideo.jp/watch/' + video_id, video_id) | |
391 | if video_id.startswith('so'): | |
392 | video_id = self._match_id(handle.geturl()) | |
fb198a8a | 393 | |
7bdcb4a4 LNO |
394 | api_data = self._parse_json(self._html_search_regex( |
395 | 'data-api-data="([^"]+)"', webpage, | |
396 | 'API data', default='{}'), video_id) | |
397 | except ExtractorError as e: | |
398 | try: | |
399 | api_data = self._download_json( | |
400 | 'https://www.nicovideo.jp/api/watch/v3/%s?_frontendId=6&_frontendVersion=0&actionTrackId=AAAAAAAAAA_%d' % (video_id, round(time.time() * 1000)), video_id, | |
401 | note='Downloading API JSON', errnote='Unable to fetch data')['data'] | |
402 | except ExtractorError: | |
403 | if not isinstance(e.cause, compat_HTTPError): | |
404 | raise | |
405 | webpage = e.cause.read().decode('utf-8', 'replace') | |
406 | error_msg = self._html_search_regex( | |
407 | r'(?s)<section\s+class="(?:(?:ErrorMessage|WatchExceptionPage-message)\s*)+">(.+?)</section>', | |
408 | webpage, 'error reason', default=None) | |
409 | if not error_msg: | |
410 | raise | |
411 | raise ExtractorError(re.sub(r'\s+', ' ', error_msg), expected=True) | |
7978e172 | 412 | |
7bdcb4a4 | 413 | formats = [] |
7978e172 | 414 | |
7bdcb4a4 LNO |
415 | def get_video_info(*items, get_first=True, **kwargs): |
416 | return traverse_obj(api_data, ('video', *items), get_all=not get_first, **kwargs) | |
7978e172 | 417 | |
7bdcb4a4 LNO |
418 | quality_info = api_data['media']['delivery']['movie'] |
419 | session_api_data = quality_info['session'] | |
420 | for (audio_quality, video_quality, protocol) in itertools.product(quality_info['audios'], quality_info['videos'], session_api_data['protocols']): | |
421 | fmt = self._extract_format_for_quality(video_id, audio_quality, video_quality, protocol) | |
422 | if fmt: | |
423 | formats.append(fmt) | |
fb198a8a | 424 | |
fb198a8a | 425 | self._sort_formats(formats) |
ee6a6116 | 426 | |
52ad14ae | 427 | # Start extracting information |
7bdcb4a4 LNO |
428 | tags = None |
429 | if webpage: | |
430 | # use og:video:tag (not logged in) | |
431 | og_video_tags = re.finditer(r'<meta\s+property="og:video:tag"\s*content="(.*?)">', webpage) | |
432 | tags = list(filter(None, (clean_html(x.group(1)) for x in og_video_tags))) | |
433 | if not tags: | |
434 | # use keywords and split with comma (not logged in) | |
435 | kwds = self._html_search_meta('keywords', webpage, default=None) | |
436 | if kwds: | |
437 | tags = [x for x in kwds.split(',') if x] | |
438 | if not tags: | |
439 | # find in json (logged in) | |
440 | tags = traverse_obj(api_data, ('tag', 'items', ..., 'name')) | |
52ad14ae | 441 | |
b2e8e7da | 442 | return { |
214c22c7 | 443 | 'id': video_id, |
2291dbce | 444 | '_api_data': api_data, |
7bdcb4a4 | 445 | 'title': get_video_info(('originalTitle', 'title')) or self._og_search_title(webpage, default=None), |
ee6a6116 | 446 | 'formats': formats, |
7bdcb4a4 LNO |
447 | 'thumbnail': get_video_info('thumbnail', 'url') or self._html_search_meta( |
448 | ('image', 'og:image'), webpage, 'thumbnail', default=None), | |
449 | 'description': clean_html(get_video_info('description')), | |
450 | 'uploader': traverse_obj(api_data, ('owner', 'nickname')), | |
451 | 'timestamp': parse_iso8601(get_video_info('registeredAt')) or parse_iso8601( | |
452 | self._html_search_meta('video:release_date', webpage, 'date published', default=None)), | |
453 | 'uploader_id': traverse_obj(api_data, ('owner', 'id')), | |
454 | 'channel': traverse_obj(api_data, ('channel', 'name'), ('community', 'name')), | |
455 | 'channel_id': traverse_obj(api_data, ('channel', 'id'), ('community', 'id')), | |
456 | 'view_count': int_or_none(get_video_info('count', 'view')), | |
457 | 'tags': tags, | |
458 | 'genre': traverse_obj(api_data, ('genre', 'label'), ('genre', 'key')), | |
459 | 'comment_count': get_video_info('count', 'comment', expected_type=int), | |
460 | 'duration': ( | |
461 | parse_duration(self._html_search_meta('video:duration', webpage, 'video duration', default=None)) | |
462 | or get_video_info('duration')), | |
463 | 'webpage_url': url_or_none(url) or f'https://www.nicovideo.jp/watch/{video_id}', | |
464 | 'subtitles': self.extract_subtitles(video_id, api_data, session_api_data), | |
52ad14ae | 465 | } |
a9bad429 | 466 | |
7bdcb4a4 LNO |
467 | def _get_subtitles(self, video_id, api_data, session_api_data): |
468 | comment_user_key = traverse_obj(api_data, ('comment', 'keys', 'userKey')) | |
469 | user_id_str = session_api_data.get('serviceUserId') | |
470 | ||
471 | thread_ids = [x for x in traverse_obj(api_data, ('comment', 'threads')) or [] if x['isActive']] | |
472 | raw_danmaku = self._extract_all_comments(video_id, thread_ids, user_id_str, comment_user_key) | |
473 | if not raw_danmaku: | |
474 | self.report_warning(f'Failed to get comments. {bug_reports_message()}') | |
475 | return | |
476 | return { | |
477 | 'comments': [{ | |
478 | 'ext': 'json', | |
479 | 'data': json.dumps(raw_danmaku), | |
480 | }], | |
481 | } | |
482 | ||
483 | def _extract_all_comments(self, video_id, threads, user_id, user_key): | |
484 | auth_data = { | |
485 | 'user_id': user_id, | |
486 | 'userkey': user_key, | |
487 | } if user_id and user_key else {'user_id': ''} | |
488 | ||
489 | # Request Start | |
490 | post_data = [{'ping': {'content': 'rs:0'}}] | |
491 | for i, thread in enumerate(threads): | |
492 | thread_id = thread['id'] | |
493 | thread_fork = thread['fork'] | |
494 | # Post Start (2N) | |
495 | post_data.append({'ping': {'content': f'ps:{i * 2}'}}) | |
496 | post_data.append({'thread': { | |
497 | 'fork': thread_fork, | |
498 | 'language': 0, | |
499 | 'nicoru': 3, | |
500 | 'scores': 1, | |
501 | 'thread': thread_id, | |
502 | 'version': '20090904', | |
503 | 'with_global': 1, | |
504 | **auth_data, | |
505 | }}) | |
506 | # Post Final (2N) | |
507 | post_data.append({'ping': {'content': f'pf:{i * 2}'}}) | |
508 | ||
509 | # Post Start (2N+1) | |
510 | post_data.append({'ping': {'content': f'ps:{i * 2 + 1}'}}) | |
511 | post_data.append({'thread_leaves': { | |
512 | # format is '<bottom of minute range>-<top of minute range>:<comments per minute>,<total last comments' | |
513 | # unfortunately NND limits (deletes?) comment returns this way, so you're only able to grab the last 1000 per language | |
514 | 'content': '0-999999:999999,999999,nicoru:999999', | |
515 | 'fork': thread_fork, | |
516 | 'language': 0, | |
517 | 'nicoru': 3, | |
518 | 'scores': 1, | |
519 | 'thread': thread_id, | |
520 | **auth_data, | |
521 | }}) | |
522 | # Post Final (2N+1) | |
523 | post_data.append({'ping': {'content': f'pf:{i * 2 + 1}'}}) | |
524 | # Request Final | |
525 | post_data.append({'ping': {'content': 'rf:0'}}) | |
526 | ||
527 | for api_url in self._COMMENT_API_ENDPOINTS: | |
528 | comments = self._download_json( | |
529 | api_url, video_id, data=json.dumps(post_data).encode(), fatal=False, | |
530 | headers={ | |
531 | 'Referer': 'https://www.nicovideo.jp/watch/%s' % video_id, | |
532 | 'Origin': 'https://www.nicovideo.jp', | |
533 | 'Content-Type': 'text/plain;charset=UTF-8', | |
534 | }, | |
535 | note='Downloading comments', errnote=f'Failed to access endpoint {api_url}') | |
536 | if comments: | |
537 | return comments | |
538 | ||
a9bad429 | 539 | |
f494ddad LNO |
540 | class NiconicoPlaylistBaseIE(InfoExtractor): |
541 | _PAGE_SIZE = 100 | |
542 | ||
543 | _API_HEADERS = { | |
544 | 'X-Frontend-ID': '6', | |
545 | 'X-Frontend-Version': '0', | |
546 | 'X-Niconico-Language': 'en-us' | |
547 | } | |
548 | ||
549 | def _call_api(self, list_id, resource, query): | |
550 | "Implement this in child class" | |
551 | pass | |
552 | ||
553 | @staticmethod | |
554 | def _parse_owner(item): | |
555 | return { | |
556 | 'uploader': traverse_obj(item, ('owner', 'name')), | |
557 | 'uploader_id': traverse_obj(item, ('owner', 'id')), | |
558 | } | |
559 | ||
560 | def _fetch_page(self, list_id, page): | |
561 | page += 1 | |
562 | resp = self._call_api(list_id, 'page %d' % page, { | |
563 | 'page': page, | |
564 | 'pageSize': self._PAGE_SIZE, | |
565 | }) | |
566 | # this is needed to support both mylist and user | |
567 | for video in traverse_obj(resp, ('items', ..., ('video', None))) or []: | |
568 | video_id = video.get('id') | |
569 | if not video_id: | |
570 | # skip {"video": {"id": "blablabla", ...}} | |
571 | continue | |
572 | count = video.get('count') or {} | |
573 | get_count = lambda x: int_or_none(count.get(x)) | |
574 | yield { | |
575 | '_type': 'url', | |
576 | 'id': video_id, | |
577 | 'title': video.get('title'), | |
578 | 'url': f'https://www.nicovideo.jp/watch/{video_id}', | |
579 | 'description': video.get('shortDescription'), | |
580 | 'duration': int_or_none(video.get('duration')), | |
581 | 'view_count': get_count('view'), | |
582 | 'comment_count': get_count('comment'), | |
583 | 'thumbnail': traverse_obj(video, ('thumbnail', ('nHdUrl', 'largeUrl', 'listingUrl', 'url'))), | |
584 | 'ie_key': NiconicoIE.ie_key(), | |
585 | **self._parse_owner(video), | |
586 | } | |
587 | ||
588 | def _entries(self, list_id): | |
589 | return OnDemandPagedList(functools.partial(self._fetch_page, list_id), self._PAGE_SIZE) | |
590 | ||
591 | ||
592 | class NiconicoPlaylistIE(NiconicoPlaylistBaseIE): | |
593 | IE_NAME = 'niconico:playlist' | |
594 | _VALID_URL = r'https?://(?:(?:www\.|sp\.)?nicovideo\.jp|nico\.ms)/(?:user/\d+/)?(?:my/)?mylist/(?:#/)?(?P<id>\d+)' | |
a9bad429 | 595 | |
29f7c58a | 596 | _TESTS = [{ |
a9bad429 JMF |
597 | 'url': 'http://www.nicovideo.jp/mylist/27411728', |
598 | 'info_dict': { | |
599 | 'id': '27411728', | |
600 | 'title': 'AKB48のオールナイトニッポン', | |
29f7c58a | 601 | 'description': 'md5:d89694c5ded4b6c693dea2db6e41aa08', |
602 | 'uploader': 'のっく', | |
603 | 'uploader_id': '805442', | |
a9bad429 | 604 | }, |
f494ddad | 605 | 'playlist_mincount': 291, |
29f7c58a | 606 | }, { |
607 | 'url': 'https://www.nicovideo.jp/user/805442/mylist/27411728', | |
608 | 'only_matching': True, | |
f494ddad LNO |
609 | }, { |
610 | 'url': 'https://www.nicovideo.jp/my/mylist/#/68048635', | |
611 | 'only_matching': True, | |
29f7c58a | 612 | }] |
a9bad429 | 613 | |
f494ddad LNO |
614 | def _call_api(self, list_id, resource, query): |
615 | return self._download_json( | |
616 | f'https://nvapi.nicovideo.jp/v2/mylists/{list_id}', list_id, | |
617 | f'Downloading {resource}', query=query, | |
618 | headers=self._API_HEADERS)['data']['mylist'] | |
2291dbce | 619 | |
a9bad429 JMF |
620 | def _real_extract(self, url): |
621 | list_id = self._match_id(url) | |
f494ddad LNO |
622 | mylist = self._call_api(list_id, 'list', { |
623 | 'pageSize': 1, | |
624 | }) | |
625 | return self.playlist_result( | |
626 | self._entries(list_id), list_id, | |
627 | mylist.get('name'), mylist.get('description'), **self._parse_owner(mylist)) | |
fb198a8a | 628 | |
fb198a8a | 629 | |
f494ddad LNO |
630 | class NiconicoSeriesIE(InfoExtractor): |
631 | IE_NAME = 'niconico:series' | |
632 | _VALID_URL = r'https?://(?:(?:www\.|sp\.)?nicovideo\.jp|nico\.ms)/series/(?P<id>\d+)' | |
633 | ||
634 | _TESTS = [{ | |
635 | 'url': 'https://www.nicovideo.jp/series/110226', | |
636 | 'info_dict': { | |
637 | 'id': '110226', | |
638 | 'title': 'ご立派ァ!のシリーズ', | |
639 | }, | |
640 | 'playlist_mincount': 10, # as of 2021/03/17 | |
641 | }, { | |
642 | 'url': 'https://www.nicovideo.jp/series/12312/', | |
643 | 'info_dict': { | |
644 | 'id': '12312', | |
645 | 'title': 'バトルスピリッツ お勧めカード紹介(調整中)', | |
646 | }, | |
647 | 'playlist_mincount': 97, # as of 2021/03/17 | |
648 | }, { | |
649 | 'url': 'https://nico.ms/series/203559', | |
650 | 'only_matching': True, | |
651 | }] | |
652 | ||
653 | def _real_extract(self, url): | |
654 | list_id = self._match_id(url) | |
655 | webpage = self._download_webpage(f'https://www.nicovideo.jp/series/{list_id}', list_id) | |
656 | ||
657 | title = self._search_regex( | |
658 | (r'<title>「(.+)(全', | |
659 | r'<div class="TwitterShareButton"\s+data-text="(.+)\s+https:'), | |
660 | webpage, 'title', fatal=False) | |
661 | if title: | |
662 | title = unescapeHTML(title) | |
663 | playlist = [ | |
664 | self.url_result(f'https://www.nicovideo.jp/watch/{v_id}', video_id=v_id) | |
665 | for v_id in re.findall(r'href="/watch/([a-z0-9]+)" data-href="/watch/\1', webpage)] | |
666 | return self.playlist_result(playlist, list_id, title) | |
667 | ||
668 | ||
669 | class NiconicoHistoryIE(NiconicoPlaylistBaseIE): | |
670 | IE_NAME = 'niconico:history' | |
671 | IE_DESC = 'NicoNico user history. Requires cookies.' | |
672 | _VALID_URL = r'https?://(?:www\.|sp\.)?nicovideo\.jp/my/history' | |
673 | ||
674 | _TESTS = [{ | |
675 | 'note': 'PC page, with /video', | |
676 | 'url': 'https://www.nicovideo.jp/my/history/video', | |
677 | 'only_matching': True, | |
678 | }, { | |
679 | 'note': 'PC page, without /video', | |
680 | 'url': 'https://www.nicovideo.jp/my/history', | |
681 | 'only_matching': True, | |
682 | }, { | |
683 | 'note': 'mobile page, with /video', | |
684 | 'url': 'https://sp.nicovideo.jp/my/history/video', | |
685 | 'only_matching': True, | |
686 | }, { | |
687 | 'note': 'mobile page, without /video', | |
688 | 'url': 'https://sp.nicovideo.jp/my/history', | |
689 | 'only_matching': True, | |
690 | }] | |
691 | ||
692 | def _call_api(self, list_id, resource, query): | |
693 | return self._download_json( | |
694 | 'https://nvapi.nicovideo.jp/v1/users/me/watch/history', 'history', | |
695 | f'Downloading {resource}', query=query, | |
696 | headers=self._API_HEADERS)['data'] | |
697 | ||
698 | def _real_extract(self, url): | |
699 | list_id = 'history' | |
700 | try: | |
701 | mylist = self._call_api(list_id, 'list', { | |
702 | 'pageSize': 1, | |
703 | }) | |
704 | except ExtractorError as e: | |
705 | if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401: | |
706 | self.raise_login_required('You have to be logged in to get your watch history') | |
707 | raise | |
708 | return self.playlist_result(self._entries(list_id), list_id, **self._parse_owner(mylist)) | |
c1d3a4a8 | 709 | |
710 | ||
63ccf4ff | 711 | class NicovideoSearchBaseIE(InfoExtractor): |
9a5b0125 LNO |
712 | _SEARCH_TYPE = 'search' |
713 | ||
63ccf4ff | 714 | def _entries(self, url, item_id, query=None, note='Downloading page %(page)s'): |
715 | query = query or {} | |
716 | pages = [query['page']] if 'page' in query else itertools.count(1) | |
717 | for page_num in pages: | |
718 | query['page'] = str(page_num) | |
719 | webpage = self._download_webpage(url, item_id, query=query, note=note % {'page': page_num}) | |
720 | results = re.findall(r'(?<=data-video-id=)["\']?(?P<videoid>.*?)(?=["\'])', webpage) | |
721 | for item in results: | |
722 | yield self.url_result(f'http://www.nicovideo.jp/watch/{item}', 'Niconico', item) | |
723 | if not results: | |
724 | break | |
725 | ||
146cc411 | 726 | def _search_results(self, query): |
727 | return self._entries( | |
9a5b0125 | 728 | self._proto_relative_url(f'//www.nicovideo.jp/{self._SEARCH_TYPE}/{query}'), query) |
146cc411 | 729 | |
63ccf4ff | 730 | |
731 | class NicovideoSearchIE(NicovideoSearchBaseIE, SearchInfoExtractor): | |
732 | IE_DESC = 'Nico video search' | |
733 | IE_NAME = 'nicovideo:search' | |
734 | _SEARCH_KEY = 'nicosearch' | |
abafce59 | 735 | |
63ccf4ff | 736 | |
737 | class NicovideoSearchURLIE(NicovideoSearchBaseIE): | |
738 | IE_NAME = f'{NicovideoSearchIE.IE_NAME}_url' | |
abafce59 | 739 | IE_DESC = 'Nico video search URLs' |
740 | _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/search/(?P<id>[^?#&]+)?' | |
741 | _TESTS = [{ | |
742 | 'url': 'http://www.nicovideo.jp/search/sm9', | |
743 | 'info_dict': { | |
744 | 'id': 'sm9', | |
745 | 'title': 'sm9' | |
746 | }, | |
747 | 'playlist_mincount': 40, | |
748 | }, { | |
749 | 'url': 'https://www.nicovideo.jp/search/sm9?sort=h&order=d&end=2020-12-31&start=2020-01-01', | |
750 | 'info_dict': { | |
751 | 'id': 'sm9', | |
752 | 'title': 'sm9' | |
753 | }, | |
754 | 'playlist_count': 31, | |
755 | }] | |
756 | ||
abafce59 | 757 | def _real_extract(self, url): |
758 | query = self._match_id(url) | |
759 | return self.playlist_result(self._entries(url, query), query, query) | |
760 | ||
761 | ||
63ccf4ff | 762 | class NicovideoSearchDateIE(NicovideoSearchBaseIE, SearchInfoExtractor): |
f304da8a | 763 | IE_DESC = 'Nico video search, newest first' |
63ccf4ff | 764 | IE_NAME = f'{NicovideoSearchIE.IE_NAME}:date' |
abafce59 | 765 | _SEARCH_KEY = 'nicosearchdate' |
766 | _TESTS = [{ | |
767 | 'url': 'nicosearchdateall:a', | |
768 | 'info_dict': { | |
769 | 'id': 'a', | |
770 | 'title': 'a' | |
771 | }, | |
772 | 'playlist_mincount': 1610, | |
773 | }] | |
774 | ||
775 | _START_DATE = datetime.date(2007, 1, 1) | |
776 | _RESULTS_PER_PAGE = 32 | |
777 | _MAX_PAGES = 50 | |
778 | ||
779 | def _entries(self, url, item_id, start_date=None, end_date=None): | |
780 | start_date, end_date = start_date or self._START_DATE, end_date or datetime.datetime.now().date() | |
781 | ||
782 | # If the last page has a full page of videos, we need to break down the query interval further | |
783 | last_page_len = len(list(self._get_entries_for_date( | |
784 | url, item_id, start_date, end_date, self._MAX_PAGES, | |
785 | note=f'Checking number of videos from {start_date} to {end_date}'))) | |
786 | if (last_page_len == self._RESULTS_PER_PAGE and start_date != end_date): | |
787 | midpoint = start_date + ((end_date - start_date) // 2) | |
788 | yield from self._entries(url, item_id, midpoint, end_date) | |
789 | yield from self._entries(url, item_id, start_date, midpoint) | |
790 | else: | |
791 | self.to_screen(f'{item_id}: Downloading results from {start_date} to {end_date}') | |
792 | yield from self._get_entries_for_date( | |
793 | url, item_id, start_date, end_date, note=' Downloading page %(page)s') | |
794 | ||
795 | def _get_entries_for_date(self, url, item_id, start_date, end_date=None, page_num=None, note=None): | |
796 | query = { | |
797 | 'start': str(start_date), | |
798 | 'end': str(end_date or start_date), | |
799 | 'sort': 'f', | |
800 | 'order': 'd', | |
801 | } | |
802 | if page_num: | |
803 | query['page'] = str(page_num) | |
804 | ||
63ccf4ff | 805 | yield from super()._entries(url, item_id, query=query, note=note) |
abafce59 | 806 | |
807 | ||
9a5b0125 LNO |
808 | class NicovideoTagURLIE(NicovideoSearchBaseIE): |
809 | IE_NAME = 'niconico:tag' | |
810 | IE_DESC = 'NicoNico video tag URLs' | |
811 | _SEARCH_TYPE = 'tag' | |
812 | _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/tag/(?P<id>[^?#&]+)?' | |
813 | _TESTS = [{ | |
814 | 'url': 'https://www.nicovideo.jp/tag/ドキュメンタリー淫夢', | |
815 | 'info_dict': { | |
816 | 'id': 'ドキュメンタリー淫夢', | |
817 | 'title': 'ドキュメンタリー淫夢' | |
818 | }, | |
819 | 'playlist_mincount': 400, | |
820 | }] | |
821 | ||
822 | def _real_extract(self, url): | |
823 | query = self._match_id(url) | |
824 | return self.playlist_result(self._entries(url, query), query, query) | |
825 | ||
826 | ||
c1d3a4a8 | 827 | class NiconicoUserIE(InfoExtractor): |
828 | _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/user/(?P<id>\d+)/?(?:$|[#?])' | |
829 | _TEST = { | |
830 | 'url': 'https://www.nicovideo.jp/user/419948', | |
831 | 'info_dict': { | |
832 | 'id': '419948', | |
833 | }, | |
834 | 'playlist_mincount': 101, | |
835 | } | |
836 | _API_URL = "https://nvapi.nicovideo.jp/v1/users/%s/videos?sortKey=registeredAt&sortOrder=desc&pageSize=%s&page=%s" | |
2291dbce | 837 | _PAGE_SIZE = 100 |
838 | ||
839 | _API_HEADERS = { | |
c1d3a4a8 | 840 | 'X-Frontend-ID': '6', |
2291dbce | 841 | 'X-Frontend-Version': '0' |
c1d3a4a8 | 842 | } |
c1d3a4a8 | 843 | |
abafce59 | 844 | def _entries(self, list_id): |
c1d3a4a8 | 845 | total_count = 1 |
846 | count = page_num = 0 | |
847 | while count < total_count: | |
848 | json_parsed = self._download_json( | |
849 | self._API_URL % (list_id, self._PAGE_SIZE, page_num + 1), list_id, | |
2291dbce | 850 | headers=self._API_HEADERS, |
c1d3a4a8 | 851 | note='Downloading JSON metadata%s' % (' page %d' % page_num if page_num else '')) |
852 | if not page_num: | |
853 | total_count = int_or_none(json_parsed['data'].get('totalCount')) | |
854 | for entry in json_parsed["data"]["items"]: | |
855 | count += 1 | |
856 | yield self.url_result('https://www.nicovideo.jp/watch/%s' % entry['id']) | |
857 | page_num += 1 | |
858 | ||
859 | def _real_extract(self, url): | |
860 | list_id = self._match_id(url) | |
861 | return self.playlist_result(self._entries(list_id), list_id, ie=NiconicoIE.ie_key()) |