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

   1 # -*- coding: utf-8 -*-
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6 from urllib import unquote
   7
   8 from .common import InfoExtractor
   9 from ..utils import (
  10     compat_urllib_request,
  11     compat_urllib_parse,
  12     compat_urllib_parse_urlparse
  13 )
  14
  15
  16 class CeskaTelevizeIE(InfoExtractor):
  17     _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(.+/)?(?P<id>[^?#]+)'
  18
  19     _TESTS = [{
  20         'url': 'http://www.ceskatelevize.cz/ivysilani/10532695142-prvni-republika/213512120230004-spanelska-chripka',
  21         'info_dict': {
  22             'id': '213512120230004',
  23             'ext': 'flv',
  24             'title': 'První republika: Španělská chřipka',
  25             'duration': 3107.4,
  26         },
  27         'params': {
  28             'skip_download': True,  # requires rtmpdump
  29         },
  30         'skip': 'Works only from Czech Republic.',
  31     }, {
  32         'url': 'http://www.ceskatelevize.cz/ivysilani/1030584952-tsatsiki-maminka-a-policajt',
  33         'info_dict': {
  34             'id': '20138143440',
  35             'ext': 'flv',
  36             'title': 'Tsatsiki, maminka a policajt',
  37             'duration': 6754.1,
  38         },
  39         'params': {
  40             'skip_download': True,  # requires rtmpdump
  41         },
  42         'skip': 'Works only from Czech Republic.',
  43     }]
  44
  45     def _real_extract(self, url):
  46         url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
  47
  48         mobj = re.match(self._VALID_URL, url)
  49         video_id = mobj.group('id')
  50
  51         webpage = self._download_webpage(url, video_id)
  52
  53         typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
  54         episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
  55
  56         data = {
  57             'playlist[0][type]': typ,
  58             'playlist[0][id]': episode_id,
  59             'requestUrl': compat_urllib_parse_urlparse(url).path,
  60             'requestSource': 'iVysilani',
  61         }
  62
  63         req = compat_urllib_request.Request('http://www.ceskatelevize.cz/ivysilani/ajax/get-playlist-url',
  64                                             data=compat_urllib_parse.urlencode(data))
  65
  66         req.add_header('Content-type', 'application/x-www-form-urlencoded')
  67         req.add_header('x-addr', '127.0.0.1')
  68         req.add_header('X-Requested-With', 'XMLHttpRequest')
  69         req.add_header('Referer', url)
  70
  71         playlistpage = self._download_webpage(req, video_id)
  72
  73         req = compat_urllib_request.Request(unquote(json.loads(playlistpage)['url']))
  74         req.add_header('Referer', url)
  75
  76         playlist = self._download_xml(req, video_id)
  77
  78         formats = []
  79         for i in playlist.find('smilRoot/body'):
  80             if 'AD' not in i.attrib['id']:
  81                 base_url = i.attrib['base']
  82                 parsedurl = compat_urllib_parse_urlparse(base_url)
  83                 duration = i.attrib['duration']
  84
  85                 for video in i.findall('video'):
  86                     if video.attrib['label'] != 'AD':
  87                         format_id = video.attrib['label']
  88                         play_path = video.attrib['src']
  89                         vbr = int(video.attrib['system-bitrate'])
  90
  91                         formats.append({
  92                             'format_id': format_id,
  93                             'url': base_url,
  94                             'vbr': vbr,
  95                             'play_path': play_path,
  96                             'app': parsedurl.path[1:] + '?' + parsedurl.query,
  97                             'rtmp_live': True,
  98                             'ext': 'flv',
  99                         })
 100
 101         self._sort_formats(formats)
 102
 103         return {
 104             'id': episode_id,
 105             'title': self._html_search_regex(r'<title>(.+?) — iVysílání — Česká televize</title>', webpage, 'title'),
 106             'duration': float(duration),
 107             'formats': formats,
 108         }