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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import datetime
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     xpath_with_ns,
  10     parse_iso8601
  11 )
  12
  13 NAMESPACE_MAP = {
  14     'media': 'http://search.yahoo.com/mrss/',
  15 }
  16
  17 # URL prefix to download the mp4 files directly instead of streaming via rtmp
  18 # Credits go to XBox-Maniac
  19 # http://board.jdownloader.org/showpost.php?p=185835&postcount=31
  20 RAW_MP4_URL = 'http://cdn.riptide-mtvn.com/'
  21
  22
  23 class GameOneIE(InfoExtractor):
  24     _VALID_URL = r'https?://(?:www\.)?gameone\.de/tv/(?P<id>\d+)'
  25     _TEST = {
  26         'url': 'http://www.gameone.de/tv/288',
  27         'md5': '136656b7fb4c9cb4a8e2d500651c499b',
  28         'info_dict': {
  29             'id': '288',
  30             'ext': 'mp4',
  31             'title': 'Game One - Folge 288',
  32             'duration': 1238,
  33             'thumbnail': 'http://s3.gameone.de/gameone/assets/video_metas/teaser_images/000/643/636/big/640x360.jpg',
  34             'description': 'FIFA-Pressepokal 2014, Star Citizen, Kingdom Come: Deliverance, Project Cars, Schöner Trants Nerdquiz Folge 2 Runde 1',
  35             'age_limit': 16,
  36             'upload_date': '20140513',
  37             'timestamp': 1399980122,
  38         }
  39     }
  40
  41     def _real_extract(self, url):
  42         mobj = re.match(self._VALID_URL, url)
  43         video_id = mobj.group('id')
  44
  45         webpage = self._download_webpage(url, video_id)
  46         og_video = self._og_search_video_url(webpage, secure=False)
  47         description = self._html_search_meta('description', webpage)
  48         age_limit = int(
  49             self._search_regex(
  50                 r'age=(\d+)',
  51                 self._html_search_meta(
  52                     'age-de-meta-label',
  53                     webpage),
  54                 'age_limit',
  55                 '0'))
  56         mrss_url = self._search_regex(r'mrss=([^&]+)', og_video, 'mrss')
  57
  58         mrss = self._download_xml(mrss_url, video_id, 'Downloading mrss')
  59         title = mrss.find('.//item/title').text
  60         thumbnail = mrss.find('.//item/image').get('url')
  61         timestamp = parse_iso8601(mrss.find('.//pubDate').text, delimiter=' ')
  62         content = mrss.find(xpath_with_ns('.//media:content', NAMESPACE_MAP))
  63         content_url = content.get('url')
  64
  65         content = self._download_xml(
  66             content_url,
  67             video_id,
  68             'Downloading media:content')
  69         rendition_items = content.findall('.//rendition')
  70         duration = int(rendition_items[0].get('duration'))
  71         formats = [
  72             {
  73                 'url': re.sub(r'.*/(r2)', RAW_MP4_URL + r'\1', r.find('./src').text),
  74                 'width': int(r.get('width')),
  75                 'height': int(r.get('height')),
  76                 'tbr': int(r.get('bitrate')),
  77             }
  78             for r in rendition_items
  79         ]
  80         self._sort_formats(formats)
  81
  82         return {
  83             'id': video_id,
  84             'title': title,
  85             'thumbnail': thumbnail,
  86             'duration': duration,
  87             'formats': formats,
  88             'description': description,
  89             'age_limit': age_limit,
  90             'timestamp': timestamp,
  91         }
  92
  93 class GameOnePlaylistIE(InfoExtractor):
  94     _VALID_URL = r'https?://(?:www\.)?gameone\.de(?:/tv)?/?$'
  95
  96     def _real_extract(self, url):
  97         this_year = datetime.date.today().year
  98         webpage = self._download_webpage('http://www.gameone.de/tv/year/%d' % this_year, this_year)
  99         max_id = max(map(int, re.findall(r'<a href="/tv/(\d+)"', webpage)))
 100         entries = [self.url_result('http://www.gameone.de/tv/%d' % video_id, 'GameOne') for video_id in range(max_id, 0, -1)]
 101
 102         return {
 103             '_type': 'playlist',
 104             'title': 'GameOne',
 105             'entries': entries,
 106         }