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

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_urllib_parse,
   9     compat_urlparse,
  10     unescapeHTML,
  11 )
  12
  13
  14 class GameSpotIE(InfoExtractor):
  15     _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?'
  16     _TEST = {
  17         'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',
  18         'md5': 'b2a30deaa8654fcccd43713a6b6a4825',
  19         'info_dict': {
  20             'id': 'gs-2300-6410818',
  21             'ext': 'mp4',
  22             'title': 'Arma 3 - Community Guide: SITREP I',
  23             'description': 'Check out this video where some of the basics of Arma 3 is explained.',
  24         }
  25     }
  26
  27     def _real_extract(self, url):
  28         page_id = self._match_id(url)
  29         webpage = self._download_webpage(url, page_id)
  30         data_video_json = self._search_regex(
  31             r'data-video=["\'](.*?)["\']', webpage, 'data video')
  32         data_video = json.loads(unescapeHTML(data_video_json))
  33
  34         # Transform the manifest url to a link to the mp4 files
  35         # they are used in mobile devices.
  36         f4m_url = data_video['videoStreams']['f4m_stream']
  37         f4m_path = compat_urlparse.urlparse(f4m_url).path
  38         QUALITIES_RE = r'((,\d+)+,?)'
  39         qualities = self._search_regex(QUALITIES_RE, f4m_path, 'qualities').strip(',').split(',')
  40         http_path = f4m_path[1:].split('/', 1)[1]
  41         http_template = re.sub(QUALITIES_RE, r'%s', http_path)
  42         http_template = http_template.replace('.csmil/manifest.f4m', '')
  43         http_template = compat_urlparse.urljoin(
  44             'http://video.gamespotcdn.com/', http_template)
  45         formats = []
  46         for q in qualities:
  47             formats.append({
  48                 'url': http_template % q,
  49                 'ext': 'mp4',
  50                 'format_id': q,
  51             })
  52
  53         return {
  54             'id': data_video['guid'],
  55             'display_id': page_id,
  56             'title': compat_urllib_parse.unquote(data_video['title']),
  57             'formats': formats,
  58             'description': self._html_search_meta('description', webpage),
  59             'thumbnail': self._og_search_thumbnail(webpage),
  60         }