yt_dlp/extractor/vuclip.py

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