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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import xml.etree.ElementTree as ET
   6
   7 from .common import InfoExtractor
   8 from ..utils import xpath_with_ns
   9
  10 NAMESPACE_MAP = {
  11     'media': 'http://search.yahoo.com/mrss/',
  12 }
  13
  14 # URL prefix to download the mp4 files directly instead of streaming via rtmp
  15 # Credits go to XBox-Maniac http://board.jdownloader.org/showpost.php?p=185835&postcount=31
  16 RAW_MP4_URL = 'http://cdn.riptide-mtvn.com/'
  17
  18 class GameOneIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:www\.)?gameone\.de/tv/(?P<id>\d+)'
  20     _TEST = {
  21         'url': 'http://www.gameone.de/tv/288',
  22         'md5': '136656b7fb4c9cb4a8e2d500651c499b',
  23         'info_dict': {
  24             'id': '288',
  25             'ext': 'mp4',
  26             'title': 'Game One - Folge 288',
  27             'duration': 1238,
  28             'thumbnail': 'http://s3.gameone.de/gameone/assets/video_metas/teaser_images/000/643/636/big/640x360.jpg',
  29             'description': 'FIFA-Pressepokal 2014, Star Citizen, Kingdom Come: Deliverance, Project Cars, Schöner Trants Nerdquiz Folge 2 Runde 1',
  30         }
  31     }
  32
  33     def _real_extract(self, url):
  34         mobj = re.match(self._VALID_URL, url)
  35         video_id = mobj.group('id')
  36
  37         webpage = self._download_webpage(url, video_id)
  38         og_video = self._og_search_video_url(webpage, secure=False)
  39         description = self._html_search_meta('description', webpage)
  40         mrss_url = self._search_regex(r'mrss=([^&]+)', og_video, 'mrss')
  41
  42         mrss = self._download_xml(mrss_url, video_id, 'Downloading mrss')
  43         title = mrss.find('.//item/title').text
  44         thumbnail = mrss.find('.//item/image').get('url')
  45         content = mrss.find(xpath_with_ns('.//media:content', NAMESPACE_MAP))
  46         content_url = content.get('url')
  47
  48         content = self._download_xml(content_url, video_id, 'Downloading media:content')
  49         rendition_items = content.findall('.//rendition')
  50         duration = int(rendition_items[0].get('duration'))
  51         formats = [
  52                 {
  53                     'url': re.sub(r'.*/(r2)', RAW_MP4_URL + r'\1', r.find('./src').text),
  54                     'width': int(r.get('width')),
  55                     'height': int(r.get('height')),
  56                     'tbr': int(r.get('bitrate')),
  57                 }
  58             for r in rendition_items
  59         ]
  60
  61         return {
  62             'id': video_id,
  63             'title': title,
  64             'thumbnail': thumbnail,
  65             'duration': duration,
  66             'formats': formats,
  67             'description': description,
  68         }