_ 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     }, {
  31         'url': 'http://www.ceskatelevize.cz/ivysilani/1030584952-tsatsiki-maminka-a-policajt',
  32         'info_dict': {
  33             'id': '20138143440',
  34             'ext': 'flv',
  35             'title': 'Tsatsiki, maminka a policajt',
  36             'duration': 6754.1,
  37         },
  38         'params': {
  39             'skip_download': True,  # requires rtmpdump
  40         },
  41     }]
  42
  43     def _real_extract(self, url):
  44         url = url.replace('/porady/', '/ivysilani/').replace('/video/', '')
  45
  46         mobj = re.match(self._VALID_URL, url)
  47         video_id = mobj.group('id')
  48
  49         webpage = self._download_webpage(url, video_id)
  50
  51         typ = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":"(.+?)","id":".+?"\}\],', webpage, 'type')
  52         episode_id = self._html_search_regex(r'getPlaylistUrl\(\[\{"type":".+?","id":"(.+?)"\}\],', webpage, 'episode_id')
  53
  54         data = {
  55             'playlist[0][type]': typ,
  56             'playlist[0][id]': episode_id,
  57             'requestUrl': compat_urllib_parse_urlparse(url).path,
  58             'requestSource': 'iVysilani',
  59         }
  60
  61         req = compat_urllib_request.Request('http://www.ceskatelevize.cz/ivysilani/ajax/get-playlist-url',
  62                                             data=compat_urllib_parse.urlencode(data))
  63
  64         req.add_header('Content-type', 'application/x-www-form-urlencoded')
  65         req.add_header('x-addr', '127.0.0.1')
  66         req.add_header('X-Requested-With', 'XMLHttpRequest')
  67         req.add_header('Referer', url)
  68
  69         playlistpage = self._download_webpage(req, video_id)
  70
  71         req = compat_urllib_request.Request(unquote(json.loads(playlistpage)['url']))
  72         req.add_header('Referer', url)
  73
  74         playlist = self._download_xml(req, video_id)
  75
  76         formats = []
  77         for i in playlist.find('smilRoot/body'):
  78             if 'AD' not in i.attrib['id']:
  79                 base_url = i.attrib['base']
  80                 parsedurl = compat_urllib_parse_urlparse(base_url)
  81                 duration = i.attrib['duration']
  82
  83                 for video in i.findall('video'):
  84                     if video.attrib['label'] != 'AD':
  85                         format_id = video.attrib['label']
  86                         play_path = video.attrib['src']
  87                         vbr = int(video.attrib['system-bitrate'])
  88
  89                         formats.append({
  90                             'format_id': format_id,
  91                             'url': base_url,
  92                             'vbr': vbr,
  93                             'play_path': play_path,
  94                             'app': parsedurl.path[1:] + '?' + parsedurl.query,
  95                             'rtmp_live': True,
  96                             'ext': 'flv',
  97                         })
  98
  99         self._sort_formats(formats)
 100
 101         return {
 102             'id': episode_id,
 103             'title': self._html_search_regex(r'<title>(.+?) — iVysílání — Česká televize</title>', webpage, 'title'),
 104             'duration': float(duration),
 105             'formats': formats,
 106         }