_ Git - youtube-dl/blob - youtube_dl/extractor/daum.py

   1 # encoding: utf-8
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_urllib_parse,
   7     determine_ext,
   8 )
   9
  10
  11 class DaumIE(InfoExtractor):
  12     _VALID_URL = r'https?://tvpot\.daum\.net/.*?clipid=(?P<id>\d+)'
  13     IE_NAME = u'daum.net'
  14
  15     _TEST = {
  16         u'url': u'http://tvpot.daum.net/clip/ClipView.do?clipid=52554690',
  17         u'file': u'52554690.mp4',
  18         u'info_dict': {
  19             u'title': u'DOTA 2GETHER 시즌2 6회 - 2부',
  20             u'description': u'DOTA 2GETHER 시즌2 6회 - 2부',
  21             u'upload_date': u'20130831',
  22             u'duration': 3868,
  23         },
  24     }
  25
  26     def _real_extract(self, url):
  27         mobj = re.match(self._VALID_URL, url)
  28         video_id = mobj.group(1)
  29         canonical_url = 'http://tvpot.daum.net/v/%s' % video_id
  30         webpage = self._download_webpage(canonical_url, video_id)
  31         full_id = self._search_regex(r'<link rel="video_src" href=".+?vid=(.+?)"',
  32             webpage, u'full id')
  33         query = compat_urllib_parse.urlencode({'vid': full_id})
  34         info = self._download_xml(
  35             'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
  36             u'Downloading video info')
  37         urls = self._download_xml(
  38             'http://videofarm.daum.net/controller/api/open/v1_2/MovieData.apixml?' + query,
  39             video_id, u'Downloading video formats info')
  40
  41         self.to_screen(u'%s: Getting video urls' % video_id)
  42         formats = []
  43         for format_el in urls.findall('result/output_list/output_list'):
  44             profile = format_el.attrib['profile']
  45             format_query = compat_urllib_parse.urlencode({
  46                 'vid': full_id,
  47                 'profile': profile,
  48             })
  49             url_doc = self._download_xml(
  50                 'http://videofarm.daum.net/controller/api/open/v1_2/MovieLocation.apixml?' + format_query,
  51                 video_id, note=False)
  52             format_url = url_doc.find('result/url').text
  53             formats.append({
  54                 'url': format_url,
  55                 'ext': determine_ext(format_url),
  56                 'format_id': profile,
  57             })
  58
  59         info = {
  60             'id': video_id,
  61             'title': info.find('TITLE').text,
  62             'formats': formats,
  63             'thumbnail': self._og_search_thumbnail(webpage),
  64             'description': info.find('CONTENTS').text,
  65             'duration': int(info.find('DURATION').text),
  66             'upload_date': info.find('REGDTTM').text[:8],
  67         }
  68         # TODO: Remove when #980 has been merged
  69         info.update(formats[-1])
  70         return info