yt_dlp/extractor/naver.py

   1 import itertools
   2 import re
   3 from urllib.parse import urlparse, parse_qs
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     clean_html,
   9     dict_get,
  10     int_or_none,
  11     join_nonempty,
  12     merge_dicts,
  13     parse_duration,
  14     traverse_obj,
  15     try_get,
  16     unified_timestamp,
  17     update_url_query,
  18 )
  19
  20
  21 class NaverBaseIE(InfoExtractor):
  22     _CAPTION_EXT_RE = r'\.(?:ttml|vtt)'
  23
  24     @staticmethod  # NB: Used in VLiveWebArchiveIE, WeverseIE
  25     def process_subtitles(vod_data, process_url):
  26         ret = {'subtitles': {}, 'automatic_captions': {}}
  27         for caption in traverse_obj(vod_data, ('captions', 'list', ...)):
  28             caption_url = caption.get('source')
  29             if not caption_url:
  30                 continue
  31             type_ = 'automatic_captions' if caption.get('type') == 'auto' else 'subtitles'
  32             lang = caption.get('locale') or join_nonempty('language', 'country', from_dict=caption) or 'und'
  33             if caption.get('type') == 'fan':
  34                 lang += '_fan%d' % next(i for i in itertools.count(1) if f'{lang}_fan{i}' not in ret[type_])
  35             ret[type_].setdefault(lang, []).extend({
  36                 'url': sub_url,
  37                 'name': join_nonempty('label', 'fanName', from_dict=caption, delim=' - '),
  38             } for sub_url in process_url(caption_url))
  39         return ret
  40
  41     def _extract_video_info(self, video_id, vid, key):
  42         video_data = self._download_json(
  43             'http://play.rmcnmv.naver.com/vod/play/v2.0/' + vid,
  44             video_id, query={
  45                 'key': key,
  46             })
  47         meta = video_data['meta']
  48         title = meta['subject']
  49         formats = []
  50         get_list = lambda x: try_get(video_data, lambda y: y[x + 's']['list'], list) or []
  51
  52         def extract_formats(streams, stream_type, query={}):
  53             for stream in streams:
  54                 stream_url = stream.get('source')
  55                 if not stream_url:
  56                     continue
  57                 stream_url = update_url_query(stream_url, query)
  58                 encoding_option = stream.get('encodingOption', {})
  59                 bitrate = stream.get('bitrate', {})
  60                 formats.append({
  61                     'format_id': '%s_%s' % (stream.get('type') or stream_type, dict_get(encoding_option, ('name', 'id'))),
  62                     'url': stream_url,
  63                     'ext': 'mp4',
  64                     'width': int_or_none(encoding_option.get('width')),
  65                     'height': int_or_none(encoding_option.get('height')),
  66                     'vbr': int_or_none(bitrate.get('video')),
  67                     'abr': int_or_none(bitrate.get('audio')),
  68                     'filesize': int_or_none(stream.get('size')),
  69                     'protocol': 'm3u8_native' if stream_type == 'HLS' else None,
  70                 })
  71
  72         extract_formats(get_list('video'), 'H264')
  73         for stream_set in video_data.get('streams', []):
  74             query = {}
  75             for param in stream_set.get('keys', []):
  76                 query[param['name']] = param['value']
  77             stream_type = stream_set.get('type')
  78             videos = stream_set.get('videos')
  79             if videos:
  80                 extract_formats(videos, stream_type, query)
  81             elif stream_type == 'HLS':
  82                 stream_url = stream_set.get('source')
  83                 if not stream_url:
  84                     continue
  85                 formats.extend(self._extract_m3u8_formats(
  86                     update_url_query(stream_url, query), video_id,
  87                     'mp4', 'm3u8_native', m3u8_id=stream_type, fatal=False))
  88
  89         replace_ext = lambda x, y: re.sub(self._CAPTION_EXT_RE, '.' + y, x)
  90
  91         def get_subs(caption_url):
  92             if re.search(self._CAPTION_EXT_RE, caption_url):
  93                 return [
  94                     replace_ext(caption_url, 'ttml'),
  95                     replace_ext(caption_url, 'vtt'),
  96                 ]
  97             return [caption_url]
  98
  99         user = meta.get('user', {})
 100
 101         return {
 102             'id': video_id,
 103             'title': title,
 104             'formats': formats,
 105             'thumbnail': try_get(meta, lambda x: x['cover']['source']),
 106             'view_count': int_or_none(meta.get('count')),
 107             'uploader_id': user.get('id'),
 108             'uploader': user.get('name'),
 109             'uploader_url': user.get('url'),
 110             **self.process_subtitles(video_data, get_subs),
 111         }
 112
 113
 114 class NaverIE(NaverBaseIE):
 115     _VALID_URL = r'https?://(?:m\.)?tv(?:cast)?\.naver\.com/(?:v|embed)/(?P<id>\d+)'
 116     _GEO_BYPASS = False
 117     _TESTS = [{
 118         'url': 'http://tv.naver.com/v/81652',
 119         'info_dict': {
 120             'id': '81652',
 121             'ext': 'mp4',
 122             'title': '[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
 123             'description': '메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
 124             'timestamp': 1378200754,
 125             'upload_date': '20130903',
 126             'uploader': '메가스터디, 합격불변의 법칙',
 127             'uploader_id': 'megastudy',
 128         },
 129     }, {
 130         'url': 'http://tv.naver.com/v/395837',
 131         'md5': '8a38e35354d26a17f73f4e90094febd3',
 132         'info_dict': {
 133             'id': '395837',
 134             'ext': 'mp4',
 135             'title': '9년이 지나도 아픈 기억, 전효성의 아버지',
 136             'description': 'md5:eb6aca9d457b922e43860a2a2b1984d3',
 137             'timestamp': 1432030253,
 138             'upload_date': '20150519',
 139             'uploader': '4가지쇼 시즌2',
 140             'uploader_id': 'wrappinguser29',
 141         },
 142         'skip': 'Georestricted',
 143     }, {
 144         'url': 'http://tvcast.naver.com/v/81652',
 145         'only_matching': True,
 146     }]
 147
 148     def _real_extract(self, url):
 149         video_id = self._match_id(url)
 150         content = self._download_json(
 151             'https://tv.naver.com/api/json/v/' + video_id,
 152             video_id, headers=self.geo_verification_headers())
 153         player_info_json = content.get('playerInfoJson') or {}
 154         current_clip = player_info_json.get('currentClip') or {}
 155
 156         vid = current_clip.get('videoId')
 157         in_key = current_clip.get('inKey')
 158
 159         if not vid or not in_key:
 160             player_auth = try_get(player_info_json, lambda x: x['playerOption']['auth'])
 161             if player_auth == 'notCountry':
 162                 self.raise_geo_restricted(countries=['KR'])
 163             elif player_auth == 'notLogin':
 164                 self.raise_login_required()
 165             raise ExtractorError('couldn\'t extract vid and key')
 166         info = self._extract_video_info(video_id, vid, in_key)
 167         info.update({
 168             'description': clean_html(current_clip.get('description')),
 169             'timestamp': int_or_none(current_clip.get('firstExposureTime'), 1000),
 170             'duration': parse_duration(current_clip.get('displayPlayTime')),
 171             'like_count': int_or_none(current_clip.get('recommendPoint')),
 172             'age_limit': 19 if current_clip.get('adult') else None,
 173         })
 174         return info
 175
 176
 177 class NaverLiveIE(InfoExtractor):
 178     IE_NAME = 'Naver:live'
 179     _VALID_URL = r'https?://(?:m\.)?tv(?:cast)?\.naver\.com/l/(?P<id>\d+)'
 180     _GEO_BYPASS = False
 181     _TESTS = [{
 182         'url': 'https://tv.naver.com/l/52010',
 183         'info_dict': {
 184             'id': '52010',
 185             'ext': 'mp4',
 186             'title': '[LIVE] 뉴스특보 : "수도권 거리두기, 2주간 2단계로 조정"',
 187             'description': 'md5:df7f0c237a5ed5e786ce5c91efbeaab3',
 188             'channel_id': 'NTV-ytnnews24-0',
 189             'start_time': 1597026780000,
 190         },
 191     }, {
 192         'url': 'https://tv.naver.com/l/51549',
 193         'info_dict': {
 194             'id': '51549',
 195             'ext': 'mp4',
 196             'title': '연합뉴스TV - 코로나19 뉴스특보',
 197             'description': 'md5:c655e82091bc21e413f549c0eaccc481',
 198             'channel_id': 'NTV-yonhapnewstv-0',
 199             'start_time': 1596406380000,
 200         },
 201     }, {
 202         'url': 'https://tv.naver.com/l/54887',
 203         'only_matching': True,
 204     }]
 205
 206     def _real_extract(self, url):
 207         video_id = self._match_id(url)
 208         page = self._download_webpage(url, video_id, 'Downloading Page', 'Unable to download Page')
 209         secure_url = self._search_regex(r'sApiF:\s+(?:"|\')([^"\']+)', page, 'secureurl')
 210
 211         info = self._extract_video_info(video_id, secure_url)
 212         info.update({
 213             'description': self._og_search_description(page)
 214         })
 215
 216         return info
 217
 218     def _extract_video_info(self, video_id, url):
 219         video_data = self._download_json(url, video_id, headers=self.geo_verification_headers())
 220         meta = video_data.get('meta')
 221         status = meta.get('status')
 222
 223         if status == 'CLOSED':
 224             raise ExtractorError('Stream is offline.', expected=True)
 225         elif status != 'OPENED':
 226             raise ExtractorError('Unknown status %s' % status)
 227
 228         title = meta.get('title')
 229         stream_list = video_data.get('streams')
 230
 231         if stream_list is None:
 232             raise ExtractorError('Could not get stream data.', expected=True)
 233
 234         formats = []
 235         for quality in stream_list:
 236             if not quality.get('url'):
 237                 continue
 238
 239             prop = quality.get('property')
 240             if prop.get('abr'):  # This abr doesn't mean Average audio bitrate.
 241                 continue
 242
 243             formats.extend(self._extract_m3u8_formats(
 244                 quality.get('url'), video_id, 'mp4',
 245                 m3u8_id=quality.get('qualityId'), live=True
 246             ))
 247
 248         return {
 249             'id': video_id,
 250             'title': title,
 251             'formats': formats,
 252             'channel_id': meta.get('channelId'),
 253             'channel_url': meta.get('channelUrl'),
 254             'thumbnail': meta.get('imgUrl'),
 255             'start_time': meta.get('startTime'),
 256             'categories': [meta.get('categoryId')],
 257             'is_live': True
 258         }
 259
 260
 261 class NaverNowIE(NaverBaseIE):
 262     IE_NAME = 'navernow'
 263     _VALID_URL = r'https?://now\.naver\.com/s/now\.(?P<id>\w+)'
 264     _API_URL = 'https://apis.naver.com/now_web/oldnow_web/v4'
 265     _TESTS = [{
 266         'url': 'https://now.naver.com/s/now.4759?shareReplayId=26331132#replay=',
 267         'md5': 'e05854162c21c221481de16b2944a0bc',
 268         'info_dict': {
 269             'id': '4759-26331132',
 270             'title': '아이키X노제\r\n💖꽁냥꽁냥💖(1)',
 271             'ext': 'mp4',
 272             'thumbnail': r're:^https?://.*\.jpg',
 273             'timestamp': 1650369600,
 274             'upload_date': '20220419',
 275             'uploader_id': 'now',
 276             'view_count': int,
 277             'uploader_url': 'https://now.naver.com/show/4759',
 278             'uploader': '아이키의 떰즈업',
 279         },
 280         'params': {
 281             'noplaylist': True,
 282         }
 283     }, {
 284         'url': 'https://now.naver.com/s/now.4759?shareHightlight=26601461#highlight=',
 285         'md5': '9f6118e398aa0f22b2152f554ea7851b',
 286         'info_dict': {
 287             'id': '4759-26601461',
 288             'title': '아이키: 나 리정한테 흔들렸어,,, 질투 폭발하는 노제 여보😾 [아이키의 떰즈업]ㅣ네이버 NOW.',
 289             'ext': 'mp4',
 290             'thumbnail': r're:^https?://.*\.jpg',
 291             'upload_date': '20220504',
 292             'timestamp': 1651648311,
 293             'uploader_id': 'now',
 294             'view_count': int,
 295             'uploader_url': 'https://now.naver.com/show/4759',
 296             'uploader': '아이키의 떰즈업',
 297         },
 298         'params': {
 299             'noplaylist': True,
 300         },
 301     }, {
 302         'url': 'https://now.naver.com/s/now.4759',
 303         'info_dict': {
 304             'id': '4759',
 305             'title': '아이키의 떰즈업',
 306         },
 307         'playlist_mincount': 101
 308     }, {
 309         'url': 'https://now.naver.com/s/now.4759?shareReplayId=26331132#replay',
 310         'info_dict': {
 311             'id': '4759',
 312             'title': '아이키의 떰즈업',
 313         },
 314         'playlist_mincount': 101,
 315     }, {
 316         'url': 'https://now.naver.com/s/now.4759?shareHightlight=26601461#highlight=',
 317         'info_dict': {
 318             'id': '4759',
 319             'title': '아이키의 떰즈업',
 320         },
 321         'playlist_mincount': 101,
 322     }, {
 323         'url': 'https://now.naver.com/s/now.kihyunplay?shareReplayId=30573291#replay',
 324         'only_matching': True,
 325     }]
 326
 327     def _extract_replay(self, show_id, replay_id):
 328         vod_info = self._download_json(f'{self._API_URL}/shows/now.{show_id}/vod/{replay_id}', replay_id)
 329         in_key = self._download_json(f'{self._API_URL}/shows/now.{show_id}/vod/{replay_id}/inkey', replay_id)['inKey']
 330         return merge_dicts({
 331             'id': f'{show_id}-{replay_id}',
 332             'title': traverse_obj(vod_info, ('episode', 'title')),
 333             'timestamp': unified_timestamp(traverse_obj(vod_info, ('episode', 'start_time'))),
 334             'thumbnail': vod_info.get('thumbnail_image_url'),
 335         }, self._extract_video_info(replay_id, vod_info['video_id'], in_key))
 336
 337     def _extract_show_replays(self, show_id):
 338         page_size = 15
 339         page = 1
 340         while True:
 341             show_vod_info = self._download_json(
 342                 f'{self._API_URL}/vod-shows/now.{show_id}', show_id,
 343                 query={'page': page, 'page_size': page_size},
 344                 note=f'Downloading JSON vod list for show {show_id} - page {page}'
 345             )['response']['result']
 346             for v in show_vod_info.get('vod_list') or []:
 347                 yield self._extract_replay(show_id, v['id'])
 348
 349             if len(show_vod_info.get('vod_list') or []) < page_size:
 350                 break
 351             page += 1
 352
 353     def _extract_show_highlights(self, show_id, highlight_id=None):
 354         page_size = 10
 355         page = 1
 356         while True:
 357             highlights_videos = self._download_json(
 358                 f'{self._API_URL}/shows/now.{show_id}/highlights/videos/', show_id,
 359                 query={'page': page, 'page_size': page_size},
 360                 note=f'Downloading JSON highlights for show {show_id} - page {page}')
 361
 362             for highlight in highlights_videos.get('results') or []:
 363                 if highlight_id and highlight.get('clip_no') != int(highlight_id):
 364                     continue
 365                 yield merge_dicts({
 366                     'id': f'{show_id}-{highlight["clip_no"]}',
 367                     'title': highlight.get('title'),
 368                     'timestamp': unified_timestamp(highlight.get('regdate')),
 369                     'thumbnail': highlight.get('thumbnail_url'),
 370                 }, self._extract_video_info(highlight['clip_no'], highlight['video_id'], highlight['video_inkey']))
 371
 372             if len(highlights_videos.get('results') or []) < page_size:
 373                 break
 374             page += 1
 375
 376     def _extract_highlight(self, show_id, highlight_id):
 377         try:
 378             return next(self._extract_show_highlights(show_id, highlight_id))
 379         except StopIteration:
 380             raise ExtractorError(f'Unable to find highlight {highlight_id} for show {show_id}')
 381
 382     def _real_extract(self, url):
 383         show_id = self._match_id(url)
 384         qs = parse_qs(urlparse(url).query)
 385
 386         if not self._yes_playlist(show_id, qs.get('shareHightlight')):
 387             return self._extract_highlight(show_id, qs['shareHightlight'][0])
 388         elif not self._yes_playlist(show_id, qs.get('shareReplayId')):
 389             return self._extract_replay(show_id, qs['shareReplayId'][0])
 390
 391         show_info = self._download_json(
 392             f'{self._API_URL}/shows/now.{show_id}/', show_id,
 393             note=f'Downloading JSON vod list for show {show_id}')
 394
 395         return self.playlist_result(
 396             itertools.chain(self._extract_show_replays(show_id), self._extract_show_highlights(show_id)),
 397             show_id, show_info.get('title'))