yt_dlp/extractor/vuclip.py

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