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