youtube_dl/extractor/gamespot.py

   1 import re
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_urllib_parse,
   7     compat_urlparse,
   8     unescapeHTML,
   9     get_meta_content,
  10 )
  11
  12
  13 class GameSpotIE(InfoExtractor):
  14     _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
  15     _TEST = {
  16         u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
  17         u"file": u"gs-2300-6410818.mp4",
  18         u"md5": u"b2a30deaa8654fcccd43713a6b6a4825",
  19         u"info_dict": {
  20             u"title": u"Arma 3 - Community Guide: SITREP I",
  21             u'description': u'Check out this video where some of the basics of Arma 3 is explained.',
  22         }
  23     }
  24
  25     def _real_extract(self, url):
  26         mobj = re.match(self._VALID_URL, url)
  27         page_id = mobj.group('page_id')
  28         webpage = self._download_webpage(url, page_id)
  29         data_video_json = self._search_regex(r'data-video=\'(.*?)\'', webpage, u'data video')
  30         data_video = json.loads(unescapeHTML(data_video_json))
  31
  32         # Transform the manifest url to a link to the mp4 files
  33         # they are used in mobile devices.
  34         f4m_url = data_video['videoStreams']['f4m_stream']
  35         f4m_path = compat_urlparse.urlparse(f4m_url).path
  36         QUALITIES_RE = r'((,\d+)+,?)'
  37         qualities = self._search_regex(QUALITIES_RE, f4m_path, u'qualities').strip(',').split(',')
  38         http_path = f4m_path[1:].split('/', 1)[1]
  39         http_template = re.sub(QUALITIES_RE, r'%s', http_path)
  40         http_template = http_template.replace('.csmil/manifest.f4m', '')
  41         http_template = compat_urlparse.urljoin('http://video.gamespotcdn.com/', http_template)
  42         formats = []
  43         for q in qualities:
  44             formats.append({
  45                 'url': http_template % q,
  46                 'ext': 'mp4',
  47                 'format_id': q,
  48             })
  49
  50         info = {
  51             'id': data_video['guid'],
  52             'title': compat_urllib_parse.unquote(data_video['title']),
  53             'formats': formats,
  54             'description': get_meta_content('description', webpage),
  55             'thumbnail': self._og_search_thumbnail(webpage),
  56         }
  57         # TODO: Remove when #980 has been merged
  58         info.update(formats[-1])
  59         return info