Raphaël G. Git Repositories - youtubedl/blob - 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 ..compat import (
   8     compat_urllib_parse_urlencode,
   9     compat_urlparse,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13 )
  14
  15
  16 class NaverIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:m\.)?tvcast\.naver\.com/v/(?P<id>\d+)'
  18
  19     _TESTS = [{
  20         'url': 'http://tvcast.naver.com/v/81652',
  21         'info_dict': {
  22             'id': '81652',
  23             'ext': 'mp4',
  24             'title': '[9월 모의고사 해설강의][수학_김상희] 수학 A형 16~20번',
  25             'description': '합격불변의 법칙 메가스터디 | 메가스터디 수학 김상희 선생님이 9월 모의고사 수학A형 16번에서 20번까지 해설강의를 공개합니다.',
  26             'upload_date': '20130903',
  27         },
  28     }, {
  29         'url': 'http://tvcast.naver.com/v/395837',
  30         'md5': '638ed4c12012c458fefcddfd01f173cd',
  31         'info_dict': {
  32             'id': '395837',
  33             'ext': 'mp4',
  34             'title': '9년이 지나도 아픈 기억, 전효성의 아버지',
  35             'description': 'md5:5bf200dcbf4b66eb1b350d1eb9c753f7',
  36             'upload_date': '20150519',
  37         },
  38         'skip': 'Georestricted',
  39     }]
  40
  41     def _real_extract(self, url):
  42         video_id = self._match_id(url)
  43         webpage = self._download_webpage(url, video_id)
  44
  45         m_id = re.search(r'var rmcPlayer = new nhn.rmcnmv.RMCVideoPlayer\("(.+?)", "(.+?)"',
  46                          webpage)
  47         if m_id is None:
  48             error = self._html_search_regex(
  49                 r'(?s)<div class="(?:nation_error|nation_box|error_box)">\s*(?:<!--.*?-->)?\s*<p class="[^"]+">(?P<msg>.+?)</p>\s*</div>',
  50                 webpage, 'error', default=None)
  51             if error:
  52                 raise ExtractorError(error, expected=True)
  53             raise ExtractorError('couldn\'t extract vid and key')
  54         vid = m_id.group(1)
  55         key = m_id.group(2)
  56         query = compat_urllib_parse_urlencode({'vid': vid, 'inKey': key, })
  57         query_urls = compat_urllib_parse_urlencode({
  58             'masterVid': vid,
  59             'protocol': 'p2p',
  60             'inKey': key,
  61         })
  62         info = self._download_xml(
  63             'http://serviceapi.rmcnmv.naver.com/flash/videoInfo.nhn?' + query,
  64             video_id, 'Downloading video info')
  65         urls = self._download_xml(
  66             'http://serviceapi.rmcnmv.naver.com/flash/playableEncodingOption.nhn?' + query_urls,
  67             video_id, 'Downloading video formats info')
  68
  69         formats = []
  70         for format_el in urls.findall('EncodingOptions/EncodingOption'):
  71             domain = format_el.find('Domain').text
  72             uri = format_el.find('uri').text
  73             f = {
  74                 'url': compat_urlparse.urljoin(domain, uri),
  75                 'ext': 'mp4',
  76                 'width': int(format_el.find('width').text),
  77                 'height': int(format_el.find('height').text),
  78             }
  79             if domain.startswith('rtmp'):
  80                 # urlparse does not support custom schemes
  81                 # https://bugs.python.org/issue18828
  82                 f.update({
  83                     'url': domain + uri,
  84                     'ext': 'flv',
  85                     'rtmp_protocol': '1',  # rtmpt
  86                 })
  87             formats.append(f)
  88         self._sort_formats(formats)
  89
  90         return {
  91             'id': video_id,
  92             'title': info.find('Subject').text,
  93             'formats': formats,
  94             'description': self._og_search_description(webpage),
  95             'thumbnail': self._og_search_thumbnail(webpage),
  96             'upload_date': info.find('WriteDate').text.replace('.', ''),
  97             'view_count': int(info.find('PlayCount').text),
  98         }