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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urllib_request,
   8     compat_urlparse,
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12     NO_DEFAULT,
  13     encode_dict,
  14     urlencode_postdata,
  15 )
  16
  17
  18 class NovaMovIE(InfoExtractor):
  19     IE_NAME = 'novamov'
  20     IE_DESC = 'NovaMov'
  21
  22     _VALID_URL_TEMPLATE = r'http://(?:(?:www\.)?%(host)s/(?:file|video)/|(?:(?:embed|www)\.)%(host)s/embed\.php\?(?:.*?&)?v=)(?P<id>[a-z\d]{13})'
  23     _VALID_URL = _VALID_URL_TEMPLATE % {'host': 'novamov\.com'}
  24
  25     _HOST = 'www.novamov.com'
  26
  27     _FILE_DELETED_REGEX = r'This file no longer exists on our servers!</h2>'
  28     _FILEKEY_REGEX = r'flashvars\.filekey="(?P<filekey>[^"]+)";'
  29     _TITLE_REGEX = r'(?s)<div class="v_tab blockborder rounded5" id="v_tab1">\s*<h3>([^<]+)</h3>'
  30     _DESCRIPTION_REGEX = r'(?s)<div class="v_tab blockborder rounded5" id="v_tab1">\s*<h3>[^<]+</h3><p>([^<]+)</p>'
  31
  32     _TEST = {
  33         'url': 'http://www.novamov.com/video/4rurhn9x446jj',
  34         'md5': '7205f346a52bbeba427603ba10d4b935',
  35         'info_dict': {
  36             'id': '4rurhn9x446jj',
  37             'ext': 'flv',
  38             'title': 'search engine optimization',
  39             'description': 'search engine optimization is used to rank the web page in the google search engine'
  40         },
  41         'skip': '"Invalid token" errors abound (in web interface as well as youtube-dl, there is nothing we can do about it.)'
  42     }
  43
  44     def _real_extract(self, url):
  45         video_id = self._match_id(url)
  46
  47         url = 'http://%s/video/%s' % (self._HOST, video_id)
  48
  49         webpage = self._download_webpage(
  50             url, video_id, 'Downloading video page')
  51
  52         if re.search(self._FILE_DELETED_REGEX, webpage) is not None:
  53             raise ExtractorError('Video %s does not exist' % video_id, expected=True)
  54
  55         def extract_filekey(default=NO_DEFAULT):
  56             return self._search_regex(
  57                 self._FILEKEY_REGEX, webpage, 'filekey', default=default)
  58
  59         filekey = extract_filekey(default=None)
  60
  61         if not filekey:
  62             fields = self._hidden_inputs(webpage)
  63             post_url = self._search_regex(
  64                 r'<form[^>]+action=(["\'])(?P<url>.+?)\1', webpage,
  65                 'post url', default=url, group='url')
  66             if not post_url.startswith('http'):
  67                 post_url = compat_urlparse.urljoin(url, post_url)
  68             request = compat_urllib_request.Request(
  69                 post_url, urlencode_postdata(encode_dict(fields)))
  70             request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  71             request.add_header('Referer', post_url)
  72             webpage = self._download_webpage(
  73                 request, video_id, 'Downloading continue to the video page')
  74
  75         filekey = extract_filekey()
  76
  77         title = self._html_search_regex(self._TITLE_REGEX, webpage, 'title', fatal=False)
  78         description = self._html_search_regex(self._DESCRIPTION_REGEX, webpage, 'description', default='', fatal=False)
  79
  80         api_response = self._download_webpage(
  81             'http://%s/api/player.api.php?key=%s&file=%s' % (self._HOST, filekey, video_id), video_id,
  82             'Downloading video api response')
  83
  84         response = compat_urlparse.parse_qs(api_response)
  85
  86         if 'error_msg' in response:
  87             raise ExtractorError('%s returned error: %s' % (self.IE_NAME, response['error_msg'][0]), expected=True)
  88
  89         video_url = response['url'][0]
  90
  91         return {
  92             'id': video_id,
  93             'url': video_url,
  94             'title': title,
  95             'description': description
  96         }