youtube_dl/extractor/naver.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     update_url_query,
  11 )
  12
  13
  14 class NaverIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:m\.)?tvcast\.naver\.com/v/(?P<id>\d+)'
  16
  17     _TESTS = [{
  18         'url': 'http://tvcast.naver.com/v/81652',
  19         'info_dict': {
  20             'id': '81652',
  21             'ext': 'mp4',
  22             'title': '[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
  23             'description': '합격불변의 법칙 메가스터디 | 메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
  24         },
  25     }, {
  26         'url': 'http://tvcast.naver.com/v/395837',
  27         'md5': '638ed4c12012c458fefcddfd01f173cd',
  28         'info_dict': {
  29             'id': '395837',
  30             'ext': 'mp4',
  31             'title': '9년이 지나도 아픈 기억, 전효성의 아버지',
  32             'description': 'md5:5bf200dcbf4b66eb1b350d1eb9c753f7',
  33         },
  34         'skip': 'Georestricted',
  35     }]
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39         webpage = self._download_webpage(url, video_id)
  40
  41         m_id = re.search(r'var rmcPlayer = new nhn.rmcnmv.RMCVideoPlayer\("(.+?)", "(.+?)"',
  42                          webpage)
  43         if m_id is None:
  44             error = self._html_search_regex(
  45                 r'(?s)<div class="(?:nation_error|nation_box|error_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>',
  46                 webpage, 'error', default=None)
  47             if error:
  48                 raise ExtractorError(error, expected=True)
  49             raise ExtractorError('couldn\'t extract vid and key')
  50         video_data = self._download_json('http://play.rmcnmv.naver.com/vod/play/v2.0/' + m_id.group(1), video_id, query={
  51             'key': m_id.group(2),
  52         })
  53         meta = video_data['meta']
  54         title = meta['subject']
  55         formats = []
  56
  57         def extract_formats(streams, stream_type, query={}):
  58             for stream in streams:
  59                 stream_url = stream.get('source')
  60                 if not stream_url:
  61                     continue
  62                 stream_url = update_url_query(stream_url, query)
  63                 encoding_option = stream.get('encodingOption', {})
  64                 bitrate = stream.get('bitrate', {})
  65                 formats.append({
  66                     'format_id': '%s_%s' % (stream.get('type') or stream_type, encoding_option.get('id') or encoding_option.get('name')),
  67                     'url': stream_url,
  68                     'width': int_or_none(encoding_option.get('width')),
  69                     'height': int_or_none(encoding_option.get('height')),
  70                     'vbr': int_or_none(bitrate.get('video')),
  71                     'abr': int_or_none(bitrate.get('audio')),
  72                     'filesize': int_or_none(stream.get('size')),
  73                     'protocol': 'm3u8_native' if stream_type == 'HLS' else None,
  74                 })
  75
  76         extract_formats(video_data.get('videos', {}).get('list', []), 'H264')
  77         for stream_set in video_data.get('streams', []):
  78             query = {}
  79             for param in stream_set.get('keys', []):
  80                 query[param['name']] = param['value']
  81             stream_type = stream_set.get('type')
  82             videos = stream_set.get('videos')
  83             if videos:
  84                 extract_formats(videos, stream_type, query)
  85             elif stream_type == 'HLS':
  86                 stream_url = stream_set.get('source')
  87                 if not stream_url:
  88                     continue
  89                 formats.extend(self._extract_m3u8_formats(
  90                     update_url_query(stream_url, query), video_id,
  91                     'mp4', 'm3u8_native', m3u8_id=stream_type, fatal=False))
  92         self._sort_formats(formats)
  93
  94         subtitles = {}
  95         for caption in video_data.get('captions', {}).get('list', []):
  96             caption_url = caption.get('source')
  97             if not caption_url:
  98                 continue
  99             subtitles.setdefault(caption.get('language') or caption.get('locale'), []).append({
 100                 'url': caption_url,
 101             })
 102
 103         return {
 104             'id': video_id,
 105             'title': title,
 106             'formats': formats,
 107             'subtitles': subtitles,
 108             'description': self._og_search_description(webpage),
 109             'thumbnail': meta.get('cover', {}).get('source') or self._og_search_thumbnail(webpage),
 110             'view_count': int_or_none(meta.get('count')),
 111         }