youtube_dl/extractor/vuclip.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urllib_parse_urlparse,
   8 )
   9 from ..utils import (
  10     ExtractorError,
  11     parse_duration,
  12     remove_end,
  13 )
  14
  15
  16 class VuClipIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:m\.)?vuclip\.com/w\?.*?cid=(?P<id>[0-9]+)'
  18
  19     _TEST = {
  20         'url': 'http://m.vuclip.com/w?cid=1129900602&bu=8589892792&frm=w&z=34801&op=0&oc=843169247&section=recommend',
  21         'info_dict': {
  22             'id': '1129900602',
  23             'ext': '3gp',
  24             'title': 'Top 10 TV Convicts',
  25             'duration': 733,
  26         }
  27     }
  28
  29     def _real_extract(self, url):
  30         video_id = self._match_id(url)
  31         webpage = self._download_webpage(url, video_id)
  32
  33         ad_m = re.search(
  34             r'''value="No.*?" onClick="location.href='([^"']+)'"''', webpage)
  35         if ad_m:
  36             urlr = compat_urllib_parse_urlparse(url)
  37             adfree_url = urlr.scheme + '://' + urlr.netloc + ad_m.group(1)
  38             webpage = self._download_webpage(
  39                 adfree_url, video_id, note='Download post-ad page')
  40
  41         error_msg = self._html_search_regex(
  42             r'<p class="message">(.*?)</p>', webpage, 'error message',
  43             default=None)
  44         if error_msg:
  45             raise ExtractorError(
  46                 '%s said: %s' % (self.IE_NAME, error_msg), expected=True)
  47
  48         # These clowns alternate between two page types
  49         video_url = self._search_regex(
  50             r'<a[^>]+href="([^"]+)"[^>]*><img[^>]+src="[^"]*/play\.gif',
  51             webpage, 'video URL', default=None)
  52         if video_url:
  53             formats = [{
  54                 'url': video_url,
  55             }]
  56         else:
  57             formats = self._parse_html5_media_entries(url, webpage, video_id)[0]['formats']
  58
  59         title = remove_end(self._html_search_regex(
  60             r'<title>(.*?)-\s*Vuclip</title>', webpage, 'title').strip(), ' - Video')
  61
  62         duration = parse_duration(self._html_search_regex(
  63             r'[(>]([0-9]+:[0-9]+)(?:<span|\))', webpage, 'duration', fatal=False))
  64
  65         return {
  66             'id': video_id,
  67             'formats': formats,
  68             'title': title,
  69             'duration': duration,
  70         }