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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     xpath_with_ns,
   9     parse_iso8601
  10 )
  11
  12 NAMESPACE_MAP = {
  13     'media': 'http://search.yahoo.com/mrss/',
  14 }
  15
  16 # URL prefix to download the mp4 files directly instead of streaming via rtmp
  17 # Credits go to XBox-Maniac http://board.jdownloader.org/showpost.php?p=185835&postcount=31
  18 RAW_MP4_URL = 'http://cdn.riptide-mtvn.com/'
  19
  20 PUB_DATE_FORMAT = '%Y-%m-%d %H:%M:%S %z'
  21
  22 class GameOneIE(InfoExtractor):
  23     _VALID_URL = r'https?://(?:www\.)?gameone\.de/tv/(?P<id>\d+)'
  24     _TEST = {
  25         'url': 'http://www.gameone.de/tv/288',
  26         'md5': '136656b7fb4c9cb4a8e2d500651c499b',
  27         'info_dict': {
  28             'id': '288',
  29             'ext': 'mp4',
  30             'title': 'Game One - Folge 288',
  31             'duration': 1238,
  32             'thumbnail': 'http://s3.gameone.de/gameone/assets/video_metas/teaser_images/000/643/636/big/640x360.jpg',
  33             'description': 'FIFA-Pressepokal 2014, Star Citizen, Kingdom Come: Deliverance, Project Cars, Schöner Trants Nerdquiz Folge 2 Runde 1',
  34             'age_limit': 16,
  35             'upload_date': '20140513',
  36             'timestamp': 1399980122,
  37         }
  38     }
  39
  40     def _real_extract(self, url):
  41         mobj = re.match(self._VALID_URL, url)
  42         video_id = mobj.group('id')
  43
  44         webpage = self._download_webpage(url, video_id)
  45         og_video = self._og_search_video_url(webpage, secure=False)
  46         description = self._html_search_meta('description', webpage)
  47         age_limit = int(self._search_regex(r'age=(\d+)', self._html_search_meta('age-de-meta-label', webpage), 'age_limit', '0'))
  48         mrss_url = self._search_regex(r'mrss=([^&]+)', og_video, 'mrss')
  49
  50         mrss = self._download_xml(mrss_url, video_id, 'Downloading mrss')
  51         title = mrss.find('.//item/title').text
  52         thumbnail = mrss.find('.//item/image').get('url')
  53         timestamp = parse_iso8601(mrss.find('.//pubDate').text, delimiter=' ')
  54         content = mrss.find(xpath_with_ns('.//media:content', NAMESPACE_MAP))
  55         content_url = content.get('url')
  56
  57         content = self._download_xml(content_url, video_id, 'Downloading media:content')
  58         rendition_items = content.findall('.//rendition')
  59         duration = int(rendition_items[0].get('duration'))
  60         formats = [
  61                 {
  62                     'url': re.sub(r'.*/(r2)', RAW_MP4_URL + r'\1', r.find('./src').text),
  63                     'width': int(r.get('width')),
  64                     'height': int(r.get('height')),
  65                     'tbr': int(r.get('bitrate')),
  66                 }
  67             for r in rendition_items
  68         ]
  69
  70         return {
  71             'id': video_id,
  72             'title': title,
  73             'thumbnail': thumbnail,
  74             'duration': duration,
  75             'formats': formats,
  76             'description': description,
  77             'age_limit': age_limit,
  78             'timestamp': timestamp,
  79         }