_ Git - youtube-dl/blob - youtube_dl/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     get_element_by_class,
  12     int_or_none,
  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             'upload_date': '20130903',
 117             'uploader': '메가스터디, 합격불변의 법칙',
 118             'uploader_id': 'megastudy',
 119         },
 120     }, {
 121         'url': 'http://tv.naver.com/v/395837',
 122         'md5': '8a38e35354d26a17f73f4e90094febd3',
 123         'info_dict': {
 124             'id': '395837',
 125             'ext': 'mp4',
 126             'title': '9년이 지나도 아픈 기억, 전효성의 아버지',
 127             'description': 'md5:eb6aca9d457b922e43860a2a2b1984d3',
 128             'upload_date': '20150519',
 129             'uploader': '4가지쇼 시즌2',
 130             'uploader_id': 'wrappinguser29',
 131         },
 132         'skip': 'Georestricted',
 133     }, {
 134         'url': 'http://tvcast.naver.com/v/81652',
 135         'only_matching': True,
 136     }]
 137
 138     def _real_extract(self, url):
 139         video_id = self._match_id(url)
 140         content = self._download_json(
 141             'https://tv.naver.com/api/contents/json/v/' + video_id,
 142             video_id, headers=self.geo_verification_headers())
 143         player_json = content.get('playerJson') or {}
 144
 145         vid = player_json.get('videoId')
 146         in_key = player_json.get('inKey')
 147
 148         if not vid or not in_key:
 149             player_auth = player_json.get('playerAuth')
 150             if player_auth == 'notCountry':
 151                 self.raise_geo_restricted(countries=['KR'])
 152             elif player_auth == 'notLogin':
 153                 self.raise_login_required()
 154             raise ExtractorError('couldn\'t extract vid and key')
 155         info = self._extract_video_info(video_id, vid, in_key)
 156
 157         clip_info_html = content.get('clipInfoHtml')
 158         if clip_info_html:
 159             info['description'] = clean_html(get_element_by_class('desc', clip_info_html))
 160             upload_date = self._search_regex(
 161                 r'<span[^>]+class="date".*?(\d{4}\.\d{2}\.\d{2})',
 162                 clip_info_html, 'upload date', fatal=False)
 163             if upload_date:
 164                 info['upload_date'] = upload_date.replace('.', '')
 165
 166         return info