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