]>
Commit | Line | Data |
---|---|---|
fb198a8a | 1 | import datetime |
f494ddad | 2 | import functools |
7bdcb4a4 | 3 | import itertools |
abafce59 | 4 | import json |
5 | import re | |
7bdcb4a4 | 6 | import time |
52ad14ae | 7 | |
abafce59 | 8 | from .common import InfoExtractor, SearchInfoExtractor |
1cc79574 | 9 | from ..compat import ( |
f494ddad | 10 | compat_HTTPError, |
1cc79574 PH |
11 | ) |
12 | from ..utils import ( | |
6110bbbf | 13 | ExtractorError, |
7bdcb4a4 LNO |
14 | OnDemandPagedList, |
15 | bug_reports_message, | |
16 | clean_html, | |
fb198a8a | 17 | float_or_none, |
b868936c | 18 | int_or_none, |
7bdcb4a4 | 19 | join_nonempty, |
1cc79574 | 20 | parse_duration, |
7bdcb4a4 | 21 | parse_filesize, |
bb865f3a | 22 | parse_iso8601, |
5d45484c LNO |
23 | parse_resolution, |
24 | qualities, | |
ee6a6116 | 25 | remove_start, |
5d45484c | 26 | str_or_none, |
f494ddad | 27 | traverse_obj, |
463e7216 | 28 | try_get, |
f494ddad | 29 | unescapeHTML, |
7bdcb4a4 LNO |
30 | update_url_query, |
31 | url_or_none, | |
6e6bc8da | 32 | urlencode_postdata, |
83bfb5e2 | 33 | urljoin, |
52ad14ae TT |
34 | ) |
35 | ||
13ebea79 | 36 | |
52ad14ae | 37 | class NiconicoIE(InfoExtractor): |
214c22c7 JMF |
38 | IE_NAME = 'niconico' |
39 | IE_DESC = 'ニコニコ動画' | |
52ad14ae | 40 | |
1c9a1457 | 41 | _TESTS = [{ |
214c22c7 | 42 | 'url': 'http://www.nicovideo.jp/watch/sm22312215', |
7bdcb4a4 | 43 | 'md5': 'd1a75c0823e2f629128c43e1212760f9', |
214c22c7 JMF |
44 | 'info_dict': { |
45 | 'id': 'sm22312215', | |
46 | 'ext': 'mp4', | |
47 | 'title': 'Big Buck Bunny', | |
463e7216 | 48 | 'thumbnail': r're:https?://.*', |
214c22c7 JMF |
49 | 'uploader': 'takuya0301', |
50 | 'uploader_id': '2698420', | |
51 | 'upload_date': '20131123', | |
aaab8c5e | 52 | 'timestamp': int, # timestamp is unstable |
214c22c7 | 53 | 'description': '(c) copyright 2008, Blender Foundation / www.bigbuckbunny.org', |
15ce1338 | 54 | 'duration': 33, |
463e7216 YCH |
55 | 'view_count': int, |
56 | 'comment_count': int, | |
52ad14ae | 57 | }, |
8e4988f1 | 58 | 'skip': 'Requires an account', |
1c9a1457 | 59 | }, { |
59d814f7 YCH |
60 | # File downloaded with and without credentials are different, so omit |
61 | # the md5 field | |
1c9a1457 | 62 | 'url': 'http://www.nicovideo.jp/watch/nm14296458', |
1c9a1457 S |
63 | 'info_dict': { |
64 | 'id': 'nm14296458', | |
65 | 'ext': 'swf', | |
66 | 'title': '【鏡音リン】Dance on media【オリジナル】take2!', | |
bb865f3a | 67 | 'description': 'md5:689f066d74610b3b22e0f1739add0f58', |
463e7216 | 68 | 'thumbnail': r're:https?://.*', |
1c9a1457 S |
69 | 'uploader': 'りょうた', |
70 | 'uploader_id': '18822557', | |
71 | 'upload_date': '20110429', | |
bb865f3a | 72 | 'timestamp': 1304065916, |
1c9a1457 S |
73 | 'duration': 209, |
74 | }, | |
8e4988f1 | 75 | 'skip': 'Requires an account', |
bb865f3a YCH |
76 | }, { |
77 | # 'video exists but is marked as "deleted" | |
b2e8e7da | 78 | # md5 is unstable |
bb865f3a | 79 | 'url': 'http://www.nicovideo.jp/watch/sm10000', |
bb865f3a YCH |
80 | 'info_dict': { |
81 | 'id': 'sm10000', | |
82 | 'ext': 'unknown_video', | |
83 | 'description': 'deleted', | |
84 | '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 | 85 | 'thumbnail': r're:https?://.*', |
b2e8e7da | 86 | 'upload_date': '20071224', |
8e4988f1 | 87 | 'timestamp': int, # timestamp field has different value if logged in |
b2e8e7da | 88 | 'duration': 304, |
463e7216 | 89 | 'view_count': int, |
bb865f3a | 90 | }, |
8e4988f1 | 91 | 'skip': 'Requires an account', |
621ffe7b YCH |
92 | }, { |
93 | 'url': 'http://www.nicovideo.jp/watch/so22543406', | |
94 | 'info_dict': { | |
95 | 'id': '1388129933', | |
96 | 'ext': 'mp4', | |
97 | 'title': '【第1回】RADIOアニメロミックス ラブライブ!~のぞえりRadio Garden~', | |
98 | 'description': 'md5:b27d224bb0ff53d3c8269e9f8b561cf1', | |
463e7216 | 99 | 'thumbnail': r're:https?://.*', |
621ffe7b YCH |
100 | 'timestamp': 1388851200, |
101 | 'upload_date': '20140104', | |
102 | 'uploader': 'アニメロチャンネル', | |
103 | 'uploader_id': '312', | |
8e4988f1 YCH |
104 | }, |
105 | 'skip': 'The viewing period of the video you were searching for has expired.', | |
463e7216 | 106 | }, { |
ee6a6116 | 107 | # video not available via `getflv`; "old" HTML5 video |
463e7216 | 108 | 'url': 'http://www.nicovideo.jp/watch/sm1151009', |
ee6a6116 | 109 | 'md5': '8fa81c364eb619d4085354eab075598a', |
463e7216 YCH |
110 | 'info_dict': { |
111 | 'id': 'sm1151009', | |
ee6a6116 | 112 | 'ext': 'mp4', |
463e7216 YCH |
113 | '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)', |
114 | 'description': 'md5:6ee077e0581ff5019773e2e714cdd0b7', | |
115 | 'thumbnail': r're:https?://.*', | |
116 | 'duration': 184, | |
117 | 'timestamp': 1190868283, | |
118 | 'upload_date': '20070927', | |
119 | 'uploader': 'denden2', | |
120 | 'uploader_id': '1392194', | |
121 | 'view_count': int, | |
122 | 'comment_count': int, | |
123 | }, | |
124 | 'skip': 'Requires an account', | |
ee6a6116 YCH |
125 | }, { |
126 | # "New" HTML5 video | |
aaab8c5e | 127 | # md5 is unstable |
ee6a6116 | 128 | 'url': 'http://www.nicovideo.jp/watch/sm31464864', |
ee6a6116 YCH |
129 | 'info_dict': { |
130 | 'id': 'sm31464864', | |
131 | 'ext': 'mp4', | |
132 | 'title': '新作TVアニメ「戦姫絶唱シンフォギアAXZ」PV 最高画質', | |
133 | 'description': 'md5:e52974af9a96e739196b2c1ca72b5feb', | |
134 | 'timestamp': 1498514060, | |
135 | 'upload_date': '20170626', | |
aaab8c5e | 136 | 'uploader': 'ゲスト', |
ee6a6116 YCH |
137 | 'uploader_id': '40826363', |
138 | 'thumbnail': r're:https?://.*', | |
139 | 'duration': 198, | |
140 | 'view_count': int, | |
141 | 'comment_count': int, | |
142 | }, | |
143 | 'skip': 'Requires an account', | |
aaab8c5e PP |
144 | }, { |
145 | # Video without owner | |
146 | 'url': 'http://www.nicovideo.jp/watch/sm18238488', | |
147 | 'md5': 'd265680a1f92bdcbbd2a507fc9e78a9e', | |
148 | 'info_dict': { | |
149 | 'id': 'sm18238488', | |
150 | 'ext': 'mp4', | |
151 | 'title': '【実写版】ミュータントタートルズ', | |
152 | 'description': 'md5:15df8988e47a86f9e978af2064bf6d8e', | |
153 | 'timestamp': 1341160408, | |
154 | 'upload_date': '20120701', | |
155 | 'uploader': None, | |
156 | 'uploader_id': None, | |
157 | 'thumbnail': r're:https?://.*', | |
158 | 'duration': 5271, | |
159 | 'view_count': int, | |
160 | 'comment_count': int, | |
161 | }, | |
162 | 'skip': 'Requires an account', | |
4a87de72 LS |
163 | }, { |
164 | 'url': 'http://sp.nicovideo.jp/watch/sm28964488?ss_pos=1&cp_in=wt_tg', | |
165 | 'only_matching': True, | |
7bdcb4a4 LNO |
166 | }, { |
167 | 'note': 'a video that is only served as an ENCRYPTED HLS.', | |
168 | 'url': 'https://www.nicovideo.jp/watch/so38016254', | |
169 | 'only_matching': True, | |
1c9a1457 | 170 | }] |
52ad14ae | 171 | |
7bdcb4a4 | 172 | _VALID_URL = r'https?://(?:(?:www\.|secure\.|sp\.)?nicovideo\.jp/watch|nico\.ms)/(?P<id>(?:[a-z]{2})?[0-9]+)' |
52ad14ae | 173 | _NETRC_MACHINE = 'niconico' |
7bdcb4a4 LNO |
174 | _COMMENT_API_ENDPOINTS = ( |
175 | 'https://nvcomment.nicovideo.jp/legacy/api.json', | |
176 | 'https://nmsg.nicovideo.jp/api.json',) | |
2291dbce | 177 | _API_HEADERS = { |
178 | 'X-Frontend-ID': '6', | |
7bdcb4a4 LNO |
179 | 'X-Frontend-Version': '0', |
180 | 'X-Niconico-Language': 'en-us', | |
181 | 'Referer': 'https://www.nicovideo.jp/', | |
182 | 'Origin': 'https://www.nicovideo.jp', | |
2291dbce | 183 | } |
184 | ||
52efa4b3 | 185 | def _perform_login(self, username, password): |
bb139491 | 186 | login_ok = True |
52ad14ae | 187 | login_form_strs = { |
bb139491 | 188 | 'mail_tel': username, |
214c22c7 | 189 | 'password': password, |
52ad14ae | 190 | } |
7bdcb4a4 LNO |
191 | self._request_webpage( |
192 | 'https://account.nicovideo.jp/login', None, | |
193 | note='Acquiring Login session') | |
83bfb5e2 | 194 | page = self._download_webpage( |
7bdcb4a4 | 195 | 'https://account.nicovideo.jp/login/redirector?show_button_twitter=1&site=niconico&show_button_facebook=1', None, |
bb139491 | 196 | note='Logging in', errnote='Unable to log in', |
7bdcb4a4 LNO |
197 | data=urlencode_postdata(login_form_strs), |
198 | headers={ | |
199 | 'Referer': 'https://account.nicovideo.jp/login', | |
200 | 'Content-Type': 'application/x-www-form-urlencoded', | |
201 | }) | |
83bfb5e2 | 202 | if 'oneTimePw' in page: |
203 | post_url = self._search_regex( | |
204 | r'<form[^>]+action=(["\'])(?P<url>.+?)\1', page, 'post url', group='url') | |
205 | page = self._download_webpage( | |
206 | urljoin('https://account.nicovideo.jp', post_url), None, | |
207 | note='Performing MFA', errnote='Unable to complete MFA', | |
208 | data=urlencode_postdata({ | |
209 | 'otp': self._get_tfa_info('6 digits code') | |
210 | }), headers={ | |
211 | 'Content-Type': 'application/x-www-form-urlencoded', | |
212 | }) | |
213 | if 'oneTimePw' in page or 'formError' in page: | |
214 | err_msg = self._html_search_regex( | |
215 | r'formError["\']+>(.*?)</div>', page, 'form_error', | |
216 | default='There\'s an error but the message can\'t be parsed.', | |
217 | flags=re.DOTALL) | |
218 | self.report_warning(f'Unable to log in: MFA challenge failed, "{err_msg}"') | |
219 | return False | |
220 | login_ok = 'class="notice error"' not in page | |
bb139491 | 221 | if not login_ok: |
83bfb5e2 | 222 | self.report_warning('Unable to log in: bad username or password') |
bb139491 | 223 | return login_ok |
52ad14ae | 224 | |
fb198a8a | 225 | def _get_heartbeat_info(self, info_dict): |
fb198a8a | 226 | video_id, video_src_id, audio_src_id = info_dict['url'].split(':')[1].split('/') |
2e25ce3a | 227 | dmc_protocol = info_dict['expected_protocol'] |
ee6a6116 | 228 | |
2291dbce | 229 | api_data = ( |
230 | info_dict.get('_api_data') | |
231 | or self._parse_json( | |
232 | self._html_search_regex( | |
233 | 'data-api-data="([^"]+)"', | |
234 | self._download_webpage('http://www.nicovideo.jp/watch/' + video_id, video_id), | |
235 | 'API data', default='{}'), | |
236 | video_id)) | |
fb198a8a | 237 | |
7978e172 | 238 | session_api_data = try_get(api_data, lambda x: x['media']['delivery']['movie']['session']) |
fb198a8a | 239 | session_api_endpoint = try_get(session_api_data, lambda x: x['urls'][0]) |
240 | ||
2291dbce | 241 | def ping(): |
7bdcb4a4 LNO |
242 | tracking_id = traverse_obj(api_data, ('media', 'delivery', 'trackingId')) |
243 | if tracking_id: | |
244 | tracking_url = update_url_query('https://nvapi.nicovideo.jp/v1/2ab0cbaa/watch', {'t': tracking_id}) | |
245 | watch_request_response = self._download_json( | |
246 | tracking_url, video_id, | |
247 | note='Acquiring permission for downloading video', fatal=False, | |
248 | headers=self._API_HEADERS) | |
249 | if traverse_obj(watch_request_response, ('meta', 'status')) != 200: | |
250 | self.report_warning('Failed to acquire permission for playing video. Video download may fail.') | |
fb198a8a | 251 | |
252 | yesno = lambda x: 'yes' if x else 'no' | |
253 | ||
7bdcb4a4 LNO |
254 | if dmc_protocol == 'http': |
255 | protocol = 'http' | |
256 | protocol_parameters = { | |
257 | 'http_output_download_parameters': { | |
258 | 'use_ssl': yesno(session_api_data['urls'][0]['isSsl']), | |
259 | 'use_well_known_port': yesno(session_api_data['urls'][0]['isWellKnownPort']), | |
260 | } | |
261 | } | |
262 | elif dmc_protocol == 'hls': | |
fb198a8a | 263 | protocol = 'm3u8' |
7bdcb4a4 LNO |
264 | segment_duration = try_get(self._configuration_arg('segment_duration'), lambda x: int(x[0])) or 6000 |
265 | parsed_token = self._parse_json(session_api_data['token'], video_id) | |
266 | encryption = traverse_obj(api_data, ('media', 'delivery', 'encryption')) | |
267 | protocol_parameters = { | |
268 | 'hls_parameters': { | |
269 | 'segment_duration': segment_duration, | |
270 | 'transfer_preset': '', | |
271 | 'use_ssl': yesno(session_api_data['urls'][0]['isSsl']), | |
272 | 'use_well_known_port': yesno(session_api_data['urls'][0]['isWellKnownPort']), | |
fb198a8a | 273 | } |
274 | } | |
7bdcb4a4 LNO |
275 | if 'hls_encryption' in parsed_token and encryption: |
276 | protocol_parameters['hls_parameters']['encryption'] = { | |
277 | parsed_token['hls_encryption']: { | |
278 | 'encrypted_key': encryption['encryptedKey'], | |
279 | 'key_uri': encryption['keyUri'], | |
fb198a8a | 280 | } |
281 | } | |
7bdcb4a4 LNO |
282 | else: |
283 | protocol = 'm3u8_native' | |
284 | else: | |
285 | raise ExtractorError(f'Unsupported DMC protocol: {dmc_protocol}') | |
ee6a6116 YCH |
286 | |
287 | session_response = self._download_json( | |
288 | session_api_endpoint['url'], video_id, | |
289 | query={'_format': 'json'}, | |
290 | headers={'Content-Type': 'application/json'}, | |
fb198a8a | 291 | note='Downloading JSON metadata for %s' % info_dict['format_id'], |
ee6a6116 YCH |
292 | data=json.dumps({ |
293 | 'session': { | |
294 | 'client_info': { | |
7978e172 | 295 | 'player_id': session_api_data.get('playerId'), |
ee6a6116 YCH |
296 | }, |
297 | 'content_auth': { | |
7978e172 | 298 | 'auth_type': try_get(session_api_data, lambda x: x['authTypes'][session_api_data['protocols'][0]]), |
299 | 'content_key_timeout': session_api_data.get('contentKeyTimeout'), | |
ee6a6116 | 300 | 'service_id': 'nicovideo', |
7978e172 | 301 | 'service_user_id': session_api_data.get('serviceUserId') |
ee6a6116 | 302 | }, |
7978e172 | 303 | 'content_id': session_api_data.get('contentId'), |
ee6a6116 YCH |
304 | 'content_src_id_sets': [{ |
305 | 'content_src_ids': [{ | |
306 | 'src_id_to_mux': { | |
fb198a8a | 307 | 'audio_src_ids': [audio_src_id], |
308 | 'video_src_ids': [video_src_id], | |
ee6a6116 YCH |
309 | } |
310 | }] | |
311 | }], | |
312 | 'content_type': 'movie', | |
313 | 'content_uri': '', | |
314 | 'keep_method': { | |
315 | 'heartbeat': { | |
7978e172 | 316 | 'lifetime': session_api_data.get('heartbeatLifetime') |
ee6a6116 YCH |
317 | } |
318 | }, | |
7bdcb4a4 | 319 | 'priority': session_api_data['priority'], |
ee6a6116 YCH |
320 | 'protocol': { |
321 | 'name': 'http', | |
322 | 'parameters': { | |
7bdcb4a4 LNO |
323 | 'http_parameters': { |
324 | 'parameters': protocol_parameters | |
325 | } | |
ee6a6116 YCH |
326 | } |
327 | }, | |
7978e172 | 328 | 'recipe_id': session_api_data.get('recipeId'), |
ee6a6116 YCH |
329 | 'session_operation_auth': { |
330 | 'session_operation_auth_by_signature': { | |
fb198a8a | 331 | 'signature': session_api_data.get('signature'), |
332 | 'token': session_api_data.get('token'), | |
ee6a6116 YCH |
333 | } |
334 | }, | |
335 | 'timing_constraint': 'unlimited' | |
336 | } | |
4d59db5b | 337 | }).encode()) |
ee6a6116 | 338 | |
fb198a8a | 339 | info_dict['url'] = session_response['data']['session']['content_uri'] |
340 | info_dict['protocol'] = protocol | |
341 | ||
342 | # get heartbeat info | |
343 | heartbeat_info_dict = { | |
344 | 'url': session_api_endpoint['url'] + '/' + session_response['data']['session']['id'] + '?_format=json&_method=PUT', | |
345 | 'data': json.dumps(session_response['data']), | |
346 | # interval, convert milliseconds to seconds, then halve to make a buffer. | |
2291dbce | 347 | 'interval': float_or_none(session_api_data.get('heartbeatLifetime'), scale=3000), |
348 | 'ping': ping | |
fb198a8a | 349 | } |
350 | ||
351 | return info_dict, heartbeat_info_dict | |
352 | ||
7bdcb4a4 LNO |
353 | def _extract_format_for_quality(self, video_id, audio_quality, video_quality, dmc_protocol): |
354 | ||
355 | if not audio_quality.get('isAvailable') or not video_quality.get('isAvailable'): | |
356 | return None | |
357 | ||
358 | def extract_video_quality(video_quality): | |
359 | return parse_filesize('%sB' % self._search_regex( | |
360 | r'\| ([0-9]*\.?[0-9]*[MK])', video_quality, 'vbr', default='')) | |
361 | ||
362 | format_id = '-'.join( | |
363 | [remove_start(s['id'], 'archive_') for s in (video_quality, audio_quality)] + [dmc_protocol]) | |
364 | ||
365 | vid_qual_label = traverse_obj(video_quality, ('metadata', 'label')) | |
366 | vid_quality = traverse_obj(video_quality, ('metadata', 'bitrate')) | |
ee6a6116 YCH |
367 | |
368 | return { | |
7bdcb4a4 | 369 | 'url': 'niconico_dmc:%s/%s/%s' % (video_id, video_quality['id'], audio_quality['id']), |
ee6a6116 | 370 | 'format_id': format_id, |
7bdcb4a4 | 371 | 'format_note': join_nonempty('DMC', vid_qual_label, dmc_protocol.upper(), delim=' '), |
ee6a6116 | 372 | 'ext': 'mp4', # Session API are used in HTML5, which always serves mp4 |
7bdcb4a4 LNO |
373 | 'acodec': 'aac', |
374 | 'vcodec': 'h264', | |
375 | 'abr': float_or_none(traverse_obj(audio_quality, ('metadata', 'bitrate')), 1000), | |
376 | 'vbr': float_or_none(vid_quality if vid_quality > 0 else extract_video_quality(vid_qual_label), 1000), | |
377 | 'height': traverse_obj(video_quality, ('metadata', 'resolution', 'height')), | |
378 | 'width': traverse_obj(video_quality, ('metadata', 'resolution', 'width')), | |
379 | 'quality': -2 if 'low' in video_quality['id'] else None, | |
380 | 'protocol': 'niconico_dmc', | |
2e25ce3a | 381 | 'expected_protocol': dmc_protocol, # XXX: This is not a documented field |
fb198a8a | 382 | 'http_headers': { |
383 | 'Origin': 'https://www.nicovideo.jp', | |
384 | 'Referer': 'https://www.nicovideo.jp/watch/' + video_id, | |
385 | } | |
ee6a6116 YCH |
386 | } |
387 | ||
52ad14ae | 388 | def _real_extract(self, url): |
937daef4 | 389 | video_id = self._match_id(url) |
52ad14ae | 390 | |
7bdcb4a4 LNO |
391 | try: |
392 | webpage, handle = self._download_webpage_handle( | |
393 | 'http://www.nicovideo.jp/watch/' + video_id, video_id) | |
394 | if video_id.startswith('so'): | |
395 | video_id = self._match_id(handle.geturl()) | |
fb198a8a | 396 | |
7bdcb4a4 LNO |
397 | api_data = self._parse_json(self._html_search_regex( |
398 | 'data-api-data="([^"]+)"', webpage, | |
399 | 'API data', default='{}'), video_id) | |
400 | except ExtractorError as e: | |
401 | try: | |
402 | api_data = self._download_json( | |
403 | 'https://www.nicovideo.jp/api/watch/v3/%s?_frontendId=6&_frontendVersion=0&actionTrackId=AAAAAAAAAA_%d' % (video_id, round(time.time() * 1000)), video_id, | |
404 | note='Downloading API JSON', errnote='Unable to fetch data')['data'] | |
405 | except ExtractorError: | |
406 | if not isinstance(e.cause, compat_HTTPError): | |
407 | raise | |
408 | webpage = e.cause.read().decode('utf-8', 'replace') | |
409 | error_msg = self._html_search_regex( | |
410 | r'(?s)<section\s+class="(?:(?:ErrorMessage|WatchExceptionPage-message)\s*)+">(.+?)</section>', | |
411 | webpage, 'error reason', default=None) | |
412 | if not error_msg: | |
413 | raise | |
414 | raise ExtractorError(re.sub(r'\s+', ' ', error_msg), expected=True) | |
7978e172 | 415 | |
7bdcb4a4 | 416 | formats = [] |
7978e172 | 417 | |
7bdcb4a4 LNO |
418 | def get_video_info(*items, get_first=True, **kwargs): |
419 | return traverse_obj(api_data, ('video', *items), get_all=not get_first, **kwargs) | |
7978e172 | 420 | |
7bdcb4a4 LNO |
421 | quality_info = api_data['media']['delivery']['movie'] |
422 | session_api_data = quality_info['session'] | |
423 | for (audio_quality, video_quality, protocol) in itertools.product(quality_info['audios'], quality_info['videos'], session_api_data['protocols']): | |
424 | fmt = self._extract_format_for_quality(video_id, audio_quality, video_quality, protocol) | |
425 | if fmt: | |
426 | formats.append(fmt) | |
fb198a8a | 427 | |
fb198a8a | 428 | self._sort_formats(formats) |
ee6a6116 | 429 | |
52ad14ae | 430 | # Start extracting information |
7bdcb4a4 LNO |
431 | tags = None |
432 | if webpage: | |
433 | # use og:video:tag (not logged in) | |
434 | og_video_tags = re.finditer(r'<meta\s+property="og:video:tag"\s*content="(.*?)">', webpage) | |
435 | tags = list(filter(None, (clean_html(x.group(1)) for x in og_video_tags))) | |
436 | if not tags: | |
437 | # use keywords and split with comma (not logged in) | |
438 | kwds = self._html_search_meta('keywords', webpage, default=None) | |
439 | if kwds: | |
440 | tags = [x for x in kwds.split(',') if x] | |
441 | if not tags: | |
442 | # find in json (logged in) | |
443 | tags = traverse_obj(api_data, ('tag', 'items', ..., 'name')) | |
52ad14ae | 444 | |
5d45484c LNO |
445 | thumb_prefs = qualities(['url', 'middleUrl', 'largeUrl', 'player', 'ogp']) |
446 | ||
b2e8e7da | 447 | return { |
214c22c7 | 448 | 'id': video_id, |
2291dbce | 449 | '_api_data': api_data, |
7bdcb4a4 | 450 | 'title': get_video_info(('originalTitle', 'title')) or self._og_search_title(webpage, default=None), |
ee6a6116 | 451 | 'formats': formats, |
5d45484c LNO |
452 | 'thumbnails': [{ |
453 | 'id': key, | |
454 | 'url': url, | |
455 | 'ext': 'jpg', | |
456 | 'preference': thumb_prefs(key), | |
457 | **parse_resolution(url, lenient=True), | |
458 | } for key, url in (get_video_info('thumbnail') or {}).items() if url], | |
7bdcb4a4 | 459 | 'description': clean_html(get_video_info('description')), |
5d45484c LNO |
460 | 'uploader': traverse_obj(api_data, ('owner', 'nickname'), ('channel', 'name'), ('community', 'name')), |
461 | 'uploader_id': str_or_none(traverse_obj(api_data, ('owner', 'id'), ('channel', 'id'), ('community', 'id'))), | |
7bdcb4a4 LNO |
462 | 'timestamp': parse_iso8601(get_video_info('registeredAt')) or parse_iso8601( |
463 | self._html_search_meta('video:release_date', webpage, 'date published', default=None)), | |
7bdcb4a4 LNO |
464 | 'channel': traverse_obj(api_data, ('channel', 'name'), ('community', 'name')), |
465 | 'channel_id': traverse_obj(api_data, ('channel', 'id'), ('community', 'id')), | |
466 | 'view_count': int_or_none(get_video_info('count', 'view')), | |
467 | 'tags': tags, | |
468 | 'genre': traverse_obj(api_data, ('genre', 'label'), ('genre', 'key')), | |
469 | 'comment_count': get_video_info('count', 'comment', expected_type=int), | |
470 | 'duration': ( | |
471 | parse_duration(self._html_search_meta('video:duration', webpage, 'video duration', default=None)) | |
472 | or get_video_info('duration')), | |
473 | 'webpage_url': url_or_none(url) or f'https://www.nicovideo.jp/watch/{video_id}', | |
474 | 'subtitles': self.extract_subtitles(video_id, api_data, session_api_data), | |
52ad14ae | 475 | } |
a9bad429 | 476 | |
7bdcb4a4 LNO |
477 | def _get_subtitles(self, video_id, api_data, session_api_data): |
478 | comment_user_key = traverse_obj(api_data, ('comment', 'keys', 'userKey')) | |
479 | user_id_str = session_api_data.get('serviceUserId') | |
480 | ||
a44ca5a4 | 481 | thread_ids = traverse_obj(api_data, ('comment', 'threads', lambda _, v: v['isActive'])) |
7bdcb4a4 LNO |
482 | raw_danmaku = self._extract_all_comments(video_id, thread_ids, user_id_str, comment_user_key) |
483 | if not raw_danmaku: | |
484 | self.report_warning(f'Failed to get comments. {bug_reports_message()}') | |
485 | return | |
486 | return { | |
487 | 'comments': [{ | |
488 | 'ext': 'json', | |
489 | 'data': json.dumps(raw_danmaku), | |
490 | }], | |
491 | } | |
492 | ||
493 | def _extract_all_comments(self, video_id, threads, user_id, user_key): | |
494 | auth_data = { | |
495 | 'user_id': user_id, | |
496 | 'userkey': user_key, | |
497 | } if user_id and user_key else {'user_id': ''} | |
498 | ||
499 | # Request Start | |
500 | post_data = [{'ping': {'content': 'rs:0'}}] | |
501 | for i, thread in enumerate(threads): | |
502 | thread_id = thread['id'] | |
503 | thread_fork = thread['fork'] | |
504 | # Post Start (2N) | |
505 | post_data.append({'ping': {'content': f'ps:{i * 2}'}}) | |
506 | post_data.append({'thread': { | |
507 | 'fork': thread_fork, | |
508 | 'language': 0, | |
509 | 'nicoru': 3, | |
510 | 'scores': 1, | |
511 | 'thread': thread_id, | |
512 | 'version': '20090904', | |
513 | 'with_global': 1, | |
514 | **auth_data, | |
515 | }}) | |
516 | # Post Final (2N) | |
517 | post_data.append({'ping': {'content': f'pf:{i * 2}'}}) | |
518 | ||
519 | # Post Start (2N+1) | |
520 | post_data.append({'ping': {'content': f'ps:{i * 2 + 1}'}}) | |
521 | post_data.append({'thread_leaves': { | |
522 | # format is '<bottom of minute range>-<top of minute range>:<comments per minute>,<total last comments' | |
523 | # unfortunately NND limits (deletes?) comment returns this way, so you're only able to grab the last 1000 per language | |
524 | 'content': '0-999999:999999,999999,nicoru:999999', | |
525 | 'fork': thread_fork, | |
526 | 'language': 0, | |
527 | 'nicoru': 3, | |
528 | 'scores': 1, | |
529 | 'thread': thread_id, | |
530 | **auth_data, | |
531 | }}) | |
532 | # Post Final (2N+1) | |
533 | post_data.append({'ping': {'content': f'pf:{i * 2 + 1}'}}) | |
534 | # Request Final | |
535 | post_data.append({'ping': {'content': 'rf:0'}}) | |
536 | ||
537 | for api_url in self._COMMENT_API_ENDPOINTS: | |
538 | comments = self._download_json( | |
539 | api_url, video_id, data=json.dumps(post_data).encode(), fatal=False, | |
540 | headers={ | |
541 | 'Referer': 'https://www.nicovideo.jp/watch/%s' % video_id, | |
542 | 'Origin': 'https://www.nicovideo.jp', | |
543 | 'Content-Type': 'text/plain;charset=UTF-8', | |
544 | }, | |
545 | note='Downloading comments', errnote=f'Failed to access endpoint {api_url}') | |
546 | if comments: | |
547 | return comments | |
548 | ||
a9bad429 | 549 | |
f494ddad LNO |
550 | class NiconicoPlaylistBaseIE(InfoExtractor): |
551 | _PAGE_SIZE = 100 | |
552 | ||
553 | _API_HEADERS = { | |
554 | 'X-Frontend-ID': '6', | |
555 | 'X-Frontend-Version': '0', | |
556 | 'X-Niconico-Language': 'en-us' | |
557 | } | |
558 | ||
559 | def _call_api(self, list_id, resource, query): | |
2fa669f7 | 560 | raise NotImplementedError('Must be implemented in subclasses') |
f494ddad LNO |
561 | |
562 | @staticmethod | |
563 | def _parse_owner(item): | |
564 | return { | |
565 | 'uploader': traverse_obj(item, ('owner', 'name')), | |
566 | 'uploader_id': traverse_obj(item, ('owner', 'id')), | |
567 | } | |
568 | ||
569 | def _fetch_page(self, list_id, page): | |
570 | page += 1 | |
571 | resp = self._call_api(list_id, 'page %d' % page, { | |
572 | 'page': page, | |
573 | 'pageSize': self._PAGE_SIZE, | |
574 | }) | |
575 | # this is needed to support both mylist and user | |
576 | for video in traverse_obj(resp, ('items', ..., ('video', None))) or []: | |
577 | video_id = video.get('id') | |
578 | if not video_id: | |
579 | # skip {"video": {"id": "blablabla", ...}} | |
580 | continue | |
581 | count = video.get('count') or {} | |
582 | get_count = lambda x: int_or_none(count.get(x)) | |
583 | yield { | |
584 | '_type': 'url', | |
585 | 'id': video_id, | |
586 | 'title': video.get('title'), | |
587 | 'url': f'https://www.nicovideo.jp/watch/{video_id}', | |
588 | 'description': video.get('shortDescription'), | |
589 | 'duration': int_or_none(video.get('duration')), | |
590 | 'view_count': get_count('view'), | |
591 | 'comment_count': get_count('comment'), | |
592 | 'thumbnail': traverse_obj(video, ('thumbnail', ('nHdUrl', 'largeUrl', 'listingUrl', 'url'))), | |
593 | 'ie_key': NiconicoIE.ie_key(), | |
594 | **self._parse_owner(video), | |
595 | } | |
596 | ||
597 | def _entries(self, list_id): | |
598 | return OnDemandPagedList(functools.partial(self._fetch_page, list_id), self._PAGE_SIZE) | |
599 | ||
600 | ||
601 | class NiconicoPlaylistIE(NiconicoPlaylistBaseIE): | |
602 | IE_NAME = 'niconico:playlist' | |
603 | _VALID_URL = r'https?://(?:(?:www\.|sp\.)?nicovideo\.jp|nico\.ms)/(?:user/\d+/)?(?:my/)?mylist/(?:#/)?(?P<id>\d+)' | |
a9bad429 | 604 | |
29f7c58a | 605 | _TESTS = [{ |
a9bad429 JMF |
606 | 'url': 'http://www.nicovideo.jp/mylist/27411728', |
607 | 'info_dict': { | |
608 | 'id': '27411728', | |
609 | 'title': 'AKB48のオールナイトニッポン', | |
29f7c58a | 610 | 'description': 'md5:d89694c5ded4b6c693dea2db6e41aa08', |
611 | 'uploader': 'のっく', | |
612 | 'uploader_id': '805442', | |
a9bad429 | 613 | }, |
f494ddad | 614 | 'playlist_mincount': 291, |
29f7c58a | 615 | }, { |
616 | 'url': 'https://www.nicovideo.jp/user/805442/mylist/27411728', | |
617 | 'only_matching': True, | |
f494ddad LNO |
618 | }, { |
619 | 'url': 'https://www.nicovideo.jp/my/mylist/#/68048635', | |
620 | 'only_matching': True, | |
29f7c58a | 621 | }] |
a9bad429 | 622 | |
f494ddad LNO |
623 | def _call_api(self, list_id, resource, query): |
624 | return self._download_json( | |
625 | f'https://nvapi.nicovideo.jp/v2/mylists/{list_id}', list_id, | |
626 | f'Downloading {resource}', query=query, | |
627 | headers=self._API_HEADERS)['data']['mylist'] | |
2291dbce | 628 | |
a9bad429 JMF |
629 | def _real_extract(self, url): |
630 | list_id = self._match_id(url) | |
f494ddad LNO |
631 | mylist = self._call_api(list_id, 'list', { |
632 | 'pageSize': 1, | |
633 | }) | |
634 | return self.playlist_result( | |
635 | self._entries(list_id), list_id, | |
636 | mylist.get('name'), mylist.get('description'), **self._parse_owner(mylist)) | |
fb198a8a | 637 | |
fb198a8a | 638 | |
f494ddad LNO |
639 | class NiconicoSeriesIE(InfoExtractor): |
640 | IE_NAME = 'niconico:series' | |
641 | _VALID_URL = r'https?://(?:(?:www\.|sp\.)?nicovideo\.jp|nico\.ms)/series/(?P<id>\d+)' | |
642 | ||
643 | _TESTS = [{ | |
644 | 'url': 'https://www.nicovideo.jp/series/110226', | |
645 | 'info_dict': { | |
646 | 'id': '110226', | |
647 | 'title': 'ご立派ァ!のシリーズ', | |
648 | }, | |
ac05fb93 | 649 | 'playlist_mincount': 10, |
f494ddad LNO |
650 | }, { |
651 | 'url': 'https://www.nicovideo.jp/series/12312/', | |
652 | 'info_dict': { | |
653 | 'id': '12312', | |
654 | 'title': 'バトルスピリッツ お勧めカード紹介(調整中)', | |
655 | }, | |
ac05fb93 | 656 | 'playlist_mincount': 103, |
f494ddad LNO |
657 | }, { |
658 | 'url': 'https://nico.ms/series/203559', | |
659 | 'only_matching': True, | |
660 | }] | |
661 | ||
662 | def _real_extract(self, url): | |
663 | list_id = self._match_id(url) | |
664 | webpage = self._download_webpage(f'https://www.nicovideo.jp/series/{list_id}', list_id) | |
665 | ||
666 | title = self._search_regex( | |
667 | (r'<title>「(.+)(全', | |
668 | r'<div class="TwitterShareButton"\s+data-text="(.+)\s+https:'), | |
669 | webpage, 'title', fatal=False) | |
670 | if title: | |
671 | title = unescapeHTML(title) | |
672 | playlist = [ | |
673 | self.url_result(f'https://www.nicovideo.jp/watch/{v_id}', video_id=v_id) | |
ac05fb93 | 674 | for v_id in re.findall(r'data-href=[\'"](?:https://www\.nicovideo\.jp)?/watch/([a-z0-9]+)', webpage)] |
f494ddad LNO |
675 | return self.playlist_result(playlist, list_id, title) |
676 | ||
677 | ||
678 | class NiconicoHistoryIE(NiconicoPlaylistBaseIE): | |
679 | IE_NAME = 'niconico:history' | |
680 | IE_DESC = 'NicoNico user history. Requires cookies.' | |
681 | _VALID_URL = r'https?://(?:www\.|sp\.)?nicovideo\.jp/my/history' | |
682 | ||
683 | _TESTS = [{ | |
684 | 'note': 'PC page, with /video', | |
685 | 'url': 'https://www.nicovideo.jp/my/history/video', | |
686 | 'only_matching': True, | |
687 | }, { | |
688 | 'note': 'PC page, without /video', | |
689 | 'url': 'https://www.nicovideo.jp/my/history', | |
690 | 'only_matching': True, | |
691 | }, { | |
692 | 'note': 'mobile page, with /video', | |
693 | 'url': 'https://sp.nicovideo.jp/my/history/video', | |
694 | 'only_matching': True, | |
695 | }, { | |
696 | 'note': 'mobile page, without /video', | |
697 | 'url': 'https://sp.nicovideo.jp/my/history', | |
698 | 'only_matching': True, | |
699 | }] | |
700 | ||
701 | def _call_api(self, list_id, resource, query): | |
702 | return self._download_json( | |
703 | 'https://nvapi.nicovideo.jp/v1/users/me/watch/history', 'history', | |
704 | f'Downloading {resource}', query=query, | |
705 | headers=self._API_HEADERS)['data'] | |
706 | ||
707 | def _real_extract(self, url): | |
708 | list_id = 'history' | |
709 | try: | |
710 | mylist = self._call_api(list_id, 'list', { | |
711 | 'pageSize': 1, | |
712 | }) | |
713 | except ExtractorError as e: | |
714 | if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401: | |
715 | self.raise_login_required('You have to be logged in to get your watch history') | |
716 | raise | |
717 | return self.playlist_result(self._entries(list_id), list_id, **self._parse_owner(mylist)) | |
c1d3a4a8 | 718 | |
719 | ||
63ccf4ff | 720 | class NicovideoSearchBaseIE(InfoExtractor): |
9a5b0125 LNO |
721 | _SEARCH_TYPE = 'search' |
722 | ||
63ccf4ff | 723 | def _entries(self, url, item_id, query=None, note='Downloading page %(page)s'): |
724 | query = query or {} | |
725 | pages = [query['page']] if 'page' in query else itertools.count(1) | |
726 | for page_num in pages: | |
727 | query['page'] = str(page_num) | |
728 | webpage = self._download_webpage(url, item_id, query=query, note=note % {'page': page_num}) | |
729 | results = re.findall(r'(?<=data-video-id=)["\']?(?P<videoid>.*?)(?=["\'])', webpage) | |
730 | for item in results: | |
731 | yield self.url_result(f'http://www.nicovideo.jp/watch/{item}', 'Niconico', item) | |
732 | if not results: | |
733 | break | |
734 | ||
146cc411 | 735 | def _search_results(self, query): |
736 | return self._entries( | |
9a5b0125 | 737 | self._proto_relative_url(f'//www.nicovideo.jp/{self._SEARCH_TYPE}/{query}'), query) |
146cc411 | 738 | |
63ccf4ff | 739 | |
740 | class NicovideoSearchIE(NicovideoSearchBaseIE, SearchInfoExtractor): | |
741 | IE_DESC = 'Nico video search' | |
742 | IE_NAME = 'nicovideo:search' | |
743 | _SEARCH_KEY = 'nicosearch' | |
abafce59 | 744 | |
63ccf4ff | 745 | |
746 | class NicovideoSearchURLIE(NicovideoSearchBaseIE): | |
747 | IE_NAME = f'{NicovideoSearchIE.IE_NAME}_url' | |
abafce59 | 748 | IE_DESC = 'Nico video search URLs' |
749 | _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/search/(?P<id>[^?#&]+)?' | |
750 | _TESTS = [{ | |
751 | 'url': 'http://www.nicovideo.jp/search/sm9', | |
752 | 'info_dict': { | |
753 | 'id': 'sm9', | |
754 | 'title': 'sm9' | |
755 | }, | |
756 | 'playlist_mincount': 40, | |
757 | }, { | |
758 | 'url': 'https://www.nicovideo.jp/search/sm9?sort=h&order=d&end=2020-12-31&start=2020-01-01', | |
759 | 'info_dict': { | |
760 | 'id': 'sm9', | |
761 | 'title': 'sm9' | |
762 | }, | |
763 | 'playlist_count': 31, | |
764 | }] | |
765 | ||
abafce59 | 766 | def _real_extract(self, url): |
767 | query = self._match_id(url) | |
768 | return self.playlist_result(self._entries(url, query), query, query) | |
769 | ||
770 | ||
63ccf4ff | 771 | class NicovideoSearchDateIE(NicovideoSearchBaseIE, SearchInfoExtractor): |
f304da8a | 772 | IE_DESC = 'Nico video search, newest first' |
63ccf4ff | 773 | IE_NAME = f'{NicovideoSearchIE.IE_NAME}:date' |
abafce59 | 774 | _SEARCH_KEY = 'nicosearchdate' |
775 | _TESTS = [{ | |
776 | 'url': 'nicosearchdateall:a', | |
777 | 'info_dict': { | |
778 | 'id': 'a', | |
779 | 'title': 'a' | |
780 | }, | |
781 | 'playlist_mincount': 1610, | |
782 | }] | |
783 | ||
784 | _START_DATE = datetime.date(2007, 1, 1) | |
785 | _RESULTS_PER_PAGE = 32 | |
786 | _MAX_PAGES = 50 | |
787 | ||
788 | def _entries(self, url, item_id, start_date=None, end_date=None): | |
789 | start_date, end_date = start_date or self._START_DATE, end_date or datetime.datetime.now().date() | |
790 | ||
791 | # If the last page has a full page of videos, we need to break down the query interval further | |
792 | last_page_len = len(list(self._get_entries_for_date( | |
793 | url, item_id, start_date, end_date, self._MAX_PAGES, | |
794 | note=f'Checking number of videos from {start_date} to {end_date}'))) | |
795 | if (last_page_len == self._RESULTS_PER_PAGE and start_date != end_date): | |
796 | midpoint = start_date + ((end_date - start_date) // 2) | |
797 | yield from self._entries(url, item_id, midpoint, end_date) | |
798 | yield from self._entries(url, item_id, start_date, midpoint) | |
799 | else: | |
800 | self.to_screen(f'{item_id}: Downloading results from {start_date} to {end_date}') | |
801 | yield from self._get_entries_for_date( | |
802 | url, item_id, start_date, end_date, note=' Downloading page %(page)s') | |
803 | ||
804 | def _get_entries_for_date(self, url, item_id, start_date, end_date=None, page_num=None, note=None): | |
805 | query = { | |
806 | 'start': str(start_date), | |
807 | 'end': str(end_date or start_date), | |
808 | 'sort': 'f', | |
809 | 'order': 'd', | |
810 | } | |
811 | if page_num: | |
812 | query['page'] = str(page_num) | |
813 | ||
63ccf4ff | 814 | yield from super()._entries(url, item_id, query=query, note=note) |
abafce59 | 815 | |
816 | ||
9a5b0125 LNO |
817 | class NicovideoTagURLIE(NicovideoSearchBaseIE): |
818 | IE_NAME = 'niconico:tag' | |
819 | IE_DESC = 'NicoNico video tag URLs' | |
820 | _SEARCH_TYPE = 'tag' | |
821 | _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/tag/(?P<id>[^?#&]+)?' | |
822 | _TESTS = [{ | |
823 | 'url': 'https://www.nicovideo.jp/tag/ドキュメンタリー淫夢', | |
824 | 'info_dict': { | |
825 | 'id': 'ドキュメンタリー淫夢', | |
826 | 'title': 'ドキュメンタリー淫夢' | |
827 | }, | |
828 | 'playlist_mincount': 400, | |
829 | }] | |
830 | ||
831 | def _real_extract(self, url): | |
832 | query = self._match_id(url) | |
833 | return self.playlist_result(self._entries(url, query), query, query) | |
834 | ||
835 | ||
c1d3a4a8 | 836 | class NiconicoUserIE(InfoExtractor): |
837 | _VALID_URL = r'https?://(?:www\.)?nicovideo\.jp/user/(?P<id>\d+)/?(?:$|[#?])' | |
838 | _TEST = { | |
839 | 'url': 'https://www.nicovideo.jp/user/419948', | |
840 | 'info_dict': { | |
841 | 'id': '419948', | |
842 | }, | |
843 | 'playlist_mincount': 101, | |
844 | } | |
845 | _API_URL = "https://nvapi.nicovideo.jp/v1/users/%s/videos?sortKey=registeredAt&sortOrder=desc&pageSize=%s&page=%s" | |
2291dbce | 846 | _PAGE_SIZE = 100 |
847 | ||
848 | _API_HEADERS = { | |
c1d3a4a8 | 849 | 'X-Frontend-ID': '6', |
2291dbce | 850 | 'X-Frontend-Version': '0' |
c1d3a4a8 | 851 | } |
c1d3a4a8 | 852 | |
abafce59 | 853 | def _entries(self, list_id): |
c1d3a4a8 | 854 | total_count = 1 |
855 | count = page_num = 0 | |
856 | while count < total_count: | |
857 | json_parsed = self._download_json( | |
858 | self._API_URL % (list_id, self._PAGE_SIZE, page_num + 1), list_id, | |
2291dbce | 859 | headers=self._API_HEADERS, |
c1d3a4a8 | 860 | note='Downloading JSON metadata%s' % (' page %d' % page_num if page_num else '')) |
861 | if not page_num: | |
862 | total_count = int_or_none(json_parsed['data'].get('totalCount')) | |
863 | for entry in json_parsed["data"]["items"]: | |
864 | count += 1 | |
865 | yield self.url_result('https://www.nicovideo.jp/watch/%s' % entry['id']) | |
866 | page_num += 1 | |
867 | ||
868 | def _real_extract(self, url): | |
869 | list_id = self._match_id(url) | |
870 | return self.playlist_result(self._entries(list_id), list_id, ie=NiconicoIE.ie_key()) |