_ 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
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     compat_urllib_request,
  10     compat_urllib_parse,
  11     compat_urllib_parse_urlparse,
  12     ExtractorError,
  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         if '<p class="title">Chyba konfigurace prohlížeče.</p>' not in webpage:
  54             msg = self._html_search_regex(r'<p class="title">(.+?)</p>', webpage, 'error-message')
  55             raise ExtractorError(msg.replace('<br />', ' '))
  56
  57         typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
  58         episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
  59
  60         data = {
  61             'playlist[0][type]': typ,
  62             'playlist[0][id]': episode_id,
  63             'requestUrl': compat_urllib_parse_urlparse(url).path,
  64             'requestSource': 'iVysilani',
  65         }
  66
  67         req = compat_urllib_request.Request('http://www.ceskatelevize.cz/ivysilani/ajax/get-playlist-url',
  68                                             data=compat_urllib_parse.urlencode(data))
  69
  70         req.add_header('Content-type', 'application/x-www-form-urlencoded')
  71         req.add_header('x-addr', '127.0.0.1')
  72         req.add_header('X-Requested-With', 'XMLHttpRequest')
  73         req.add_header('Referer', url)
  74
  75         playlistpage = self._download_webpage(req, video_id)
  76
  77         req = compat_urllib_request.Request(compat_urllib_parse.unquote(json.loads(playlistpage)['url']))
  78         req.add_header('Referer', url)
  79
  80         playlist = self._download_xml(req, video_id)
  81
  82         formats = []
  83         for i in playlist.find('smilRoot/body'):
  84             if 'AD' not in i.attrib['id']:
  85                 base_url = i.attrib['base']
  86                 parsedurl = compat_urllib_parse_urlparse(base_url)
  87                 duration = i.attrib['duration']
  88
  89                 for video in i.findall('video'):
  90                     if video.attrib['label'] != 'AD':
  91                         format_id = video.attrib['label']
  92                         play_path = video.attrib['src']
  93                         vbr = int(video.attrib['system-bitrate'])
  94
  95                         formats.append({
  96                             'format_id': format_id,
  97                             'url': base_url,
  98                             'vbr': vbr,
  99                             'play_path': play_path,
 100                             'app': parsedurl.path[1:] + '?' + parsedurl.query,
 101                             'rtmp_live': True,
 102                             'ext': 'flv',
 103                         })
 104
 105         self._sort_formats(formats)
 106
 107         return {
 108             'id': episode_id,
 109             'title': self._html_search_regex(r'<title>(.+?) — iVysílání — Česká televize</title>', webpage, 'title'),
 110             'duration': float(duration),
 111             'formats': formats,
 112         }