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

   1 import re
   2 import xml.etree.ElementTree
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     unified_strdate,
   7     compat_urllib_parse,
   8 )
   9
  10 class GameSpotIE(InfoExtractor):
  11     _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
  12     _TEST = {
  13         u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
  14         u"file": u"6410818.mp4",
  15         u"md5": u"b2a30deaa8654fcccd43713a6b6a4825",
  16         u"info_dict": {
  17             u"title": u"Arma 3 - Community Guide: SITREP I",
  18             u"upload_date": u"20130627",
  19         }
  20     }
  21
  22
  23     def _real_extract(self, url):
  24         mobj = re.match(self._VALID_URL, url)
  25         page_id = mobj.group('page_id')
  26         webpage = self._download_webpage(url, page_id)
  27         video_id = self._html_search_regex([r'"og:video" content=".*?\?id=(\d+)"',
  28                                             r'http://www\.gamespot\.com/videoembed/(\d+)'],
  29                                            webpage, 'video id')
  30         data = compat_urllib_parse.urlencode({'id': video_id, 'newplayer': '1'})
  31         info_url = 'http://www.gamespot.com/pages/video_player/xml.php?' + data
  32         info_xml = self._download_webpage(info_url, video_id)
  33         doc = xml.etree.ElementTree.fromstring(info_xml)
  34         clip_el = doc.find('./playList/clip')
  35
  36         http_urls = [{'url': node.find('filePath').text,
  37                       'rate': int(node.find('rate').text)}
  38             for node in clip_el.find('./httpURI')]
  39         best_quality = sorted(http_urls, key=lambda f: f['rate'])[-1]
  40         video_url = best_quality['url']
  41         title = clip_el.find('./title').text
  42         ext = video_url.rpartition('.')[2]
  43         thumbnail_url = clip_el.find('./screenGrabURI').text
  44         view_count = int(clip_el.find('./views').text)
  45         upload_date = unified_strdate(clip_el.find('./postDate').text)
  46
  47         return [{
  48             'id'          : video_id,
  49             'url'         : video_url,
  50             'ext'         : ext,
  51             'title'       : title,
  52             'thumbnail'   : thumbnail_url,
  53             'upload_date' : upload_date,
  54             'view_count'  : view_count,
  55         }]