yt_dlp/extractor/naver.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     clean_html,
   9     dict_get,
  10     ExtractorError,
  11     int_or_none,
  12     parse_duration,
  13     try_get,
  14     update_url_query,
  15 )
  16
  17
  18 class NaverBaseIE(InfoExtractor):
  19     _CAPTION_EXT_RE = r'\.(?:ttml|vtt)'
  20
  21     def _extract_video_info(self, video_id, vid, key):
  22         video_data = self._download_json(
  23             'http://play.rmcnmv.naver.com/vod/play/v2.0/' + vid,
  24             video_id, query={
  25                 'key': key,
  26             })
  27         meta = video_data['meta']
  28         title = meta['subject']
  29         formats = []
  30         get_list = lambda x: try_get(video_data, lambda y: y[x + 's']['list'], list) or []
  31
  32         def extract_formats(streams, stream_type, query={}):
  33             for stream in streams:
  34                 stream_url = stream.get('source')
  35                 if not stream_url:
  36                     continue
  37                 stream_url = update_url_query(stream_url, query)
  38                 encoding_option = stream.get('encodingOption', {})
  39                 bitrate = stream.get('bitrate', {})
  40                 formats.append({
  41                     'format_id': '%s_%s' % (stream.get('type') or stream_type, dict_get(encoding_option, ('name', 'id'))),
  42                     'url': stream_url,
  43                     'width': int_or_none(encoding_option.get('width')),
  44                     'height': int_or_none(encoding_option.get('height')),
  45                     'vbr': int_or_none(bitrate.get('video')),
  46                     'abr': int_or_none(bitrate.get('audio')),
  47                     'filesize': int_or_none(stream.get('size')),
  48                     'protocol': 'm3u8_native' if stream_type == 'HLS' else None,
  49                 })
  50
  51         extract_formats(get_list('video'), 'H264')
  52         for stream_set in video_data.get('streams', []):
  53             query = {}
  54             for param in stream_set.get('keys', []):
  55                 query[param['name']] = param['value']
  56             stream_type = stream_set.get('type')
  57             videos = stream_set.get('videos')
  58             if videos:
  59                 extract_formats(videos, stream_type, query)
  60             elif stream_type == 'HLS':
  61                 stream_url = stream_set.get('source')
  62                 if not stream_url:
  63                     continue
  64                 formats.extend(self._extract_m3u8_formats(
  65                     update_url_query(stream_url, query), video_id,
  66                     'mp4', 'm3u8_native', m3u8_id=stream_type, fatal=False))
  67         self._sort_formats(formats)
  68
  69         replace_ext = lambda x, y: re.sub(self._CAPTION_EXT_RE, '.' + y, x)
  70
  71         def get_subs(caption_url):
  72             if re.search(self._CAPTION_EXT_RE, caption_url):
  73                 return [{
  74                     'url': replace_ext(caption_url, 'ttml'),
  75                 }, {
  76                     'url': replace_ext(caption_url, 'vtt'),
  77                 }]
  78             else:
  79                 return [{'url': caption_url}]
  80
  81         automatic_captions = {}
  82         subtitles = {}
  83         for caption in get_list('caption'):
  84             caption_url = caption.get('source')
  85             if not caption_url:
  86                 continue
  87             sub_dict = automatic_captions if caption.get('type') == 'auto' else subtitles
  88             sub_dict.setdefault(dict_get(caption, ('locale', 'language')), []).extend(get_subs(caption_url))
  89
  90         user = meta.get('user', {})
  91
  92         return {
  93             'id': video_id,
  94             'title': title,
  95             'formats': formats,
  96             'subtitles': subtitles,
  97             'automatic_captions': automatic_captions,
  98             'thumbnail': try_get(meta, lambda x: x['cover']['source']),
  99             'view_count': int_or_none(meta.get('count')),
 100             'uploader_id': user.get('id'),
 101             'uploader': user.get('name'),
 102             'uploader_url': user.get('url'),
 103         }
 104
 105
 106 class NaverIE(NaverBaseIE):
 107     _VALID_URL = r'https?://(?:m\.)?tv(?:cast)?\.naver\.com/(?:v|embed)/(?P<id>\d+)'
 108     _GEO_BYPASS = False
 109     _TESTS = [{
 110         'url': 'http://tv.naver.com/v/81652',
 111         'info_dict': {
 112             'id': '81652',
 113             'ext': 'mp4',
 114             'title': '[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
 115             'description': '메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
 116             'timestamp': 1378200754,
 117             'upload_date': '20130903',
 118             'uploader': '메가스터디, 합격불변의 법칙',
 119             'uploader_id': 'megastudy',
 120         },
 121     }, {
 122         'url': 'http://tv.naver.com/v/395837',
 123         'md5': '8a38e35354d26a17f73f4e90094febd3',
 124         'info_dict': {
 125             'id': '395837',
 126             'ext': 'mp4',
 127             'title': '9년이 지나도 아픈 기억, 전효성의 아버지',
 128             'description': 'md5:eb6aca9d457b922e43860a2a2b1984d3',
 129             'timestamp': 1432030253,
 130             'upload_date': '20150519',
 131             'uploader': '4가지쇼 시즌2',
 132             'uploader_id': 'wrappinguser29',
 133         },
 134         'skip': 'Georestricted',
 135     }, {
 136         'url': 'http://tvcast.naver.com/v/81652',
 137         'only_matching': True,
 138     }]
 139
 140     def _real_extract(self, url):
 141         video_id = self._match_id(url)
 142         content = self._download_json(
 143             'https://tv.naver.com/api/json/v/' + video_id,
 144             video_id, headers=self.geo_verification_headers())
 145         player_info_json = content.get('playerInfoJson') or {}
 146         current_clip = player_info_json.get('currentClip') or {}
 147
 148         vid = current_clip.get('videoId')
 149         in_key = current_clip.get('inKey')
 150
 151         if not vid or not in_key:
 152             player_auth = try_get(player_info_json, lambda x: x['playerOption']['auth'])
 153             if player_auth == 'notCountry':
 154                 self.raise_geo_restricted(countries=['KR'])
 155             elif player_auth == 'notLogin':
 156                 self.raise_login_required()
 157             raise ExtractorError('couldn\'t extract vid and key')
 158         info = self._extract_video_info(video_id, vid, in_key)
 159         info.update({
 160             'description': clean_html(current_clip.get('description')),
 161             'timestamp': int_or_none(current_clip.get('firstExposureTime'), 1000),
 162             'duration': parse_duration(current_clip.get('displayPlayTime')),
 163             'like_count': int_or_none(current_clip.get('recommendPoint')),
 164             'age_limit': 19 if current_clip.get('adult') else None,
 165         })
 166         return info
 167
 168
 169 class NaverLiveIE(InfoExtractor):
 170     IE_NAME = 'Naver:live'
 171     _VALID_URL = r'https?://(?:m\.)?tv(?:cast)?\.naver\.com/l/(?P<id>\d+)'
 172     _GEO_BYPASS = False
 173     _TESTS = [{
 174         'url': 'https://tv.naver.com/l/52010',
 175         'info_dict': {
 176             'id': '52010',
 177             'ext': 'm3u8',
 178             'title': '[LIVE] 뉴스특보 : "수도권 거리두기, 2주간 2단계로 조정"',
 179             'description': 'md5:df7f0c237a5ed5e786ce5c91efbeaab3',
 180             'channel_id': 'NTV-ytnnews24-0',
 181             'start_time': 1597026780000,
 182         },
 183     }, {
 184         'url': 'https://tv.naver.com/l/51549',
 185         'info_dict': {
 186             'id': '51549',
 187             'ext': 'm3u8',
 188             'title': '연합뉴스TV - 코로나19 뉴스특보',
 189             'description': 'md5:c655e82091bc21e413f549c0eaccc481',
 190             'channel_id': 'NTV-yonhapnewstv-0',
 191             'start_time': 1596406380000,
 192         },
 193     }, {
 194         'url': 'https://tv.naver.com/l/54887',
 195         'only_matching': True,
 196     }]
 197
 198     def _real_extract(self, url):
 199         video_id = self._match_id(url)
 200         page = self._download_webpage(url, video_id, 'Downloading Page', 'Unable to download Page')
 201         secure_url = self._search_regex(r'sApiF:\s+(?:"|\')([^"\']+)', page, 'secureurl')
 202
 203         info = self._extract_video_info(video_id, secure_url)
 204         info.update({
 205             'description': self._og_search_description(page)
 206         })
 207
 208         return info
 209
 210     def _extract_video_info(self, video_id, url):
 211         video_data = self._download_json(url, video_id, headers=self.geo_verification_headers())
 212         meta = video_data.get('meta')
 213         status = meta.get('status')
 214
 215         if status == 'CLOSED':
 216             raise ExtractorError('Stream is offline.', expected=True)
 217         elif status != 'OPENED':
 218             raise ExtractorError('Unknown status %s' % status)
 219
 220         title = meta.get('title')
 221         stream_list = video_data.get('streams')
 222
 223         if stream_list is None:
 224             raise ExtractorError('Could not get stream data.', expected=True)
 225
 226         formats = []
 227         for quality in stream_list:
 228             if not quality.get('url'):
 229                 continue
 230
 231             prop = quality.get('property')
 232             if prop.get('abr'):  # This abr doesn't mean Average audio bitrate.
 233                 continue
 234
 235             formats.extend(self._extract_m3u8_formats(
 236                 quality.get('url'), video_id, 'm3u8',
 237                 m3u8_id=quality.get('qualityId'), live=True
 238             ))
 239         self._sort_formats(formats)
 240
 241         return {
 242             'id': video_id,
 243             'title': title,
 244             'formats': formats,
 245             'channel_id': meta.get('channelId'),
 246             'channel_url': meta.get('channelUrl'),
 247             'thumbnail': meta.get('imgUrl'),
 248             'start_time': meta.get('startTime'),
 249             'categories': [meta.get('categoryId')],
 250             'is_live': True
 251         }