Raphaël G. Git Repositories - youtubedl/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
  18 # http://board.jdownloader.org/showpost.php?p=185835&postcount=31
  19 RAW_MP4_URL = 'http://cdn.riptide-mtvn.com/'
  20
  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(
  48             self._search_regex(
  49                 r'age=(\d+)',
  50                 self._html_search_meta(
  51                     'age-de-meta-label',
  52                     webpage),
  53                 'age_limit',
  54                 '0'))
  55         mrss_url = self._search_regex(r'mrss=([^&]+)', og_video, 'mrss')
  56
  57         mrss = self._download_xml(mrss_url, video_id, 'Downloading mrss')
  58         title = mrss.find('.//item/title').text
  59         thumbnail = mrss.find('.//item/image').get('url')
  60         timestamp = parse_iso8601(mrss.find('.//pubDate').text, delimiter=' ')
  61         content = mrss.find(xpath_with_ns('.//media:content', NAMESPACE_MAP))
  62         content_url = content.get('url')
  63
  64         content = self._download_xml(
  65             content_url,
  66             video_id,
  67             'Downloading media:content')
  68         rendition_items = content.findall('.//rendition')
  69         duration = int(rendition_items[0].get('duration'))
  70         formats = [
  71             {
  72                 'url': re.sub(r'.*/(r2)', RAW_MP4_URL + r'\1', r.find('./src').text),
  73                 'width': int(r.get('width')),
  74                 'height': int(r.get('height')),
  75                 'tbr': int(r.get('bitrate')),
  76             }
  77             for r in rendition_items
  78         ]
  79         self._sort_formats(formats)
  80
  81         return {
  82             'id': video_id,
  83             'title': title,
  84             'thumbnail': thumbnail,
  85             'duration': duration,
  86             'formats': formats,
  87             'description': description,
  88             'age_limit': age_limit,
  89             'timestamp': timestamp,
  90         }
  91
  92
  93 class GameOnePlaylistIE(InfoExtractor):
  94     _VALID_URL = r'https?://(?:www\.)?gameone\.de(?:/tv)?/?$'
  95     IE_NAME = 'gameone:playlist'
  96     _TEST = {
  97         'url': 'http://www.gameone.de/tv',
  98         'info_dict': {
  99             'title': 'GameOne',
 100         },
 101         'playlist_mincount': 294,
 102     }
 103
 104     def _real_extract(self, url):
 105         webpage = self._download_webpage('http://www.gameone.de/tv', 'TV')
 106         max_id = max(map(int, re.findall(r'<a href="/tv/(\d+)"', webpage)))
 107         entries = [
 108             self.url_result('http://www.gameone.de/tv/%d' % video_id, 'GameOne')
 109             for video_id in range(max_id, 0, -1)]
 110
 111         return {
 112             '_type': 'playlist',
 113             'title': 'GameOne',
 114             'entries': entries,
 115         }