youtube_dl/extractor/naver.py

   1 # encoding: utf-8
   2 import re
   3 import xml.etree.ElementTree
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_urllib_parse,
   8     ExtractorError,
   9 )
  10
  11
  12 class NaverIE(InfoExtractor):
  13     _VALID_URL = r'https?://tvcast\.naver\.com/v/(?P<id>\d+)'
  14
  15     _TEST = {
  16         u'url': u'http://tvcast.naver.com/v/81652',
  17         u'file': u'81652.mp4',
  18         u'info_dict': {
  19             u'title': u'[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
  20             u'description': u'합격불변의 법칙 메가스터디 | 메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
  21             u'upload_date': u'20130903',
  22         },
  23     }
  24
  25     def _real_extract(self, url):
  26         mobj = re.match(self._VALID_URL, url)
  27         video_id = mobj.group(1)
  28         webpage = self._download_webpage(url, video_id)
  29         m_id = re.search(r'var rmcPlayer = new nhn.rmcnmv.RMCVideoPlayer\("(.+?)", "(.+?)"',
  30             webpage)
  31         if m_id is None:
  32             raise ExtractorError(u'couldn\'t extract vid and key')
  33         vid = m_id.group(1)
  34         key = m_id.group(2)
  35         query = compat_urllib_parse.urlencode({'vid': vid, 'inKey': key,})
  36         query_urls = compat_urllib_parse.urlencode({
  37             'masterVid': vid,
  38             'protocol': 'p2p',
  39             'inKey': key,
  40         })
  41         info_xml = self._download_webpage(
  42             'http://serviceapi.rmcnmv.naver.com/flash/videoInfo.nhn?' + query,
  43             video_id, u'Downloading video info')
  44         urls_xml = self._download_webpage(
  45             'http://serviceapi.rmcnmv.naver.com/flash/playableEncodingOption.nhn?' + query_urls,
  46             video_id, u'Downloading video formats info')
  47         info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8'))
  48         urls = xml.etree.ElementTree.fromstring(urls_xml.encode('utf-8'))
  49
  50         formats = []
  51         for format_el in urls.findall('EncodingOptions/EncodingOption'):
  52             domain = format_el.find('Domain').text
  53             if domain.startswith('rtmp'):
  54                 continue
  55             formats.append({
  56                 'url': domain + format_el.find('uri').text,
  57                 'ext': 'mp4',
  58                 'width': int(format_el.find('width').text),
  59                 'height': int(format_el.find('height').text),
  60             })
  61
  62         info = {
  63             'id': video_id,
  64             'title': info.find('Subject').text,
  65             'formats': formats,
  66             'description': self._og_search_description(webpage),
  67             'thumbnail': self._og_search_thumbnail(webpage),
  68             'upload_date': info.find('WriteDate').text.replace('.', ''),
  69             'view_count': int(info.find('PlayCount').text),
  70         }
  71         # TODO: Remove when #980 has been merged
  72         info.update(formats[-1])
  73         return info