_ Git - youtube-dl/blob - youtube_dl/extractor/rtl2.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     ExtractorError,
  10     clean_html,
  11     unified_strdate,
  12     int_or_none,
  13 )
  14
  15
  16 class RTL2IE(InfoExtractor):
  17     """Information Extractor for RTL NOW, RTL2 NOW, RTL NITRO, SUPER RTL NOW, VOX NOW and n-tv NOW"""
  18     _VALID_URL = r'http?://(?P<url>(?P<domain>(www\.)?rtl2\.de)/.*/(?P<video_id>.*))'
  19     _TEST = {
  20         'url': 'http://www.rtl2.de/sendung/grip-das-motormagazin/folge/folge-203-0',
  21         'md5': 'dsadasdada',
  22         'info_dict': {
  23             'id': 'folge-203-0',
  24             'ext': 'f4v',
  25             'title': 'GRIP sucht den Sommerk\xf6nig',
  26             'description' : 'Matthias, Det und Helge treten gegeneinander an.'
  27             # TODO more properties, either as:
  28             # * A value
  29             # * MD5 checksum; start the string with md5:
  30             # * A regular expression; start the string with re:
  31             # * Any Python type (for example int or float)
  32         },
  33         #'params': {
  34                 # rtmp download
  35         #       'skip_download': True,
  36         #},
  37     }
  38
  39     def _real_extract(self, url):
  40         mobj = re.match(self._VALID_URL, url)
  41         video_page_url = 'http://%s/' % mobj.group('domain')
  42         video_id = mobj.group('video_id')
  43
  44         webpage = self._download_webpage('http://' + mobj.group('url'), video_id)
  45
  46         vico_id = self._html_search_regex(r'vico_id: ([0-9]+)', webpage, '%s');
  47         vivi_id = self._html_search_regex(r'vivi_id: ([0-9]+)', webpage, '%s');
  48
  49         info_url = 'http://www.rtl2.de/video/php/get_video.php?vico_id=' + vico_id + '&vivi_id=' + vivi_id
  50         webpage = self._download_webpage(info_url, '')
  51
  52         video_info = json.loads(webpage.decode("latin1"))
  53         print video_info
  54
  55
  56         #self._download_webpage('http://cp108781.edgefcs.net/crossdomain.xml', '')
  57
  58         download_url = video_info["video"]["streamurl"] # self._html_search_regex(r'streamurl\":\"(.*?)\"', webpage, '%s');
  59         title = video_info["video"]["titel"] # self._html_search_regex(r'titel\":\"(.*?)\"', webpage, '%s');
  60         description = video_info["video"]["beschreibung"] # self._html_search_regex(r'beschreibung\":\"(.*?)\"', webpage, '%s');
  61         #ext = self._html_search_regex(r'streamurl\":\".*?(\..{2,4})\"', webpage, '%s');
  62
  63         thumbnail = video_info["video"]["image"]
  64
  65         download_url = download_url.replace("\\", "")
  66
  67         stream_url = 'mp4:' + self._html_search_regex(r'ondemand/(.*)', download_url, '%s');
  68
  69         #upload_date = self._html_search_regex(r'property=\"dc:date\".*?datatype=\"xsd:dateTime\".*?content=\"(.*?)\"', webpage, 'title')
  70         #download_url += " -y " + stream_url
  71
  72         #print stream_url
  73         #print download_url
  74         #print description
  75         #print title
  76         #print ext
  77
  78         formats = []
  79
  80         fmt = {
  81             'url' : download_url,
  82             #'app': 'ondemand?_fcs_vhost=cp108781.edgefcs.net',
  83             'play_path': stream_url,
  84             #'player_url': 'http://www.cbsnews.com/[[IMPORT]]/vidtech.cbsinteractive.com/player/3_3_0/CBSI_PLAYER_HD.swf',
  85             #'page_url': 'http://www.cbsnews.com',
  86             #'ext': ext,
  87         }
  88
  89         formats.append(fmt)
  90
  91
  92         return {
  93             'id': video_id,
  94             'title': title,
  95             'thumbnail' : thumbnail,
  96             'description' : description,
  97             'formats': formats,
  98         }