yt_dlp/extractor/niconico.py

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