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