Raphaël G. Git Repositories - youtubedl/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 ..compat import (
   8     compat_urllib_parse,
   9     compat_urlparse,
  10 )
  11 from ..utils import (
  12     unescapeHTML,
  13 )
  14
  15
  16 class GameSpotIE(InfoExtractor):
  17     _VALID_URL = r'http://(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?'
  18     _TESTS = [{
  19         'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',
  20         'md5': 'b2a30deaa8654fcccd43713a6b6a4825',
  21         'info_dict': {
  22             'id': 'gs-2300-6410818',
  23             'ext': 'mp4',
  24             'title': 'Arma 3 - Community Guide: SITREP I',
  25             'description': 'Check out this video where some of the basics of Arma 3 is explained.',
  26         },
  27     }, {
  28         'url': 'http://www.gamespot.com/videos/the-witcher-3-wild-hunt-xbox-one-now-playing/2300-6424837/',
  29         'info_dict': {
  30             'id': 'gs-2300-6424837',
  31             'ext': 'flv',
  32             'title': 'The Witcher 3: Wild Hunt [Xbox ONE]  - Now Playing',
  33             'description': 'Join us as we take a look at the early hours of The Witcher 3: Wild Hunt and more.',
  34         },
  35     }]
  36
  37     def _real_extract(self, url):
  38         page_id = self._match_id(url)
  39         webpage = self._download_webpage(url, page_id)
  40         data_video_json = self._search_regex(
  41             r'data-video=["\'](.*?)["\']', webpage, 'data video')
  42         data_video = json.loads(unescapeHTML(data_video_json))
  43         streams = data_video['videoStreams']
  44
  45         formats = []
  46         f4m_url = streams.get('f4m_stream')
  47         if f4m_url is not None:
  48             # Transform the manifest url to a link to the mp4 files
  49             # they are used in mobile devices.
  50             f4m_path = compat_urlparse.urlparse(f4m_url).path
  51             QUALITIES_RE = r'((,\d+)+,?)'
  52             qualities = self._search_regex(QUALITIES_RE, f4m_path, 'qualities').strip(',').split(',')
  53             http_path = f4m_path[1:].split('/', 1)[1]
  54             http_template = re.sub(QUALITIES_RE, r'%s', http_path)
  55             http_template = http_template.replace('.csmil/manifest.f4m', '')
  56             http_template = compat_urlparse.urljoin(
  57                 'http://video.gamespotcdn.com/', http_template)
  58             for q in qualities:
  59                 formats.append({
  60                     'url': http_template % q,
  61                     'ext': 'mp4',
  62                     'format_id': q,
  63                 })
  64         else:
  65             for quality in ['sd', 'hd']:
  66                 # It's actually a link to a flv file
  67                 flv_url = streams.get('f4m_{0}'.format(quality))
  68                 if flv_url is not None:
  69                     formats.append({
  70                         'url': flv_url,
  71                         'ext': 'flv',
  72                         'format_id': quality,
  73                     })
  74
  75         return {
  76             'id': data_video['guid'],
  77             'display_id': page_id,
  78             'title': compat_urllib_parse.unquote(data_video['title']),
  79             'formats': formats,
  80             'description': self._html_search_meta('description', webpage),
  81             'thumbnail': self._og_search_thumbnail(webpage),
  82         }