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