youtube_dl/extractor/playvid.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_urllib_parse,
   8 )
   9
  10
  11 class PlayvidIE(InfoExtractor):
  12     _VALID_URL = r'^https?://www\.playvid\.com/watch(\?v=|/)(?P<id>.+?)(?:#|$)'
  13     _TEST = {
  14         'url': 'http://www.playvid.com/watch/agbDDi7WZTV',
  15         'md5': '44930f8afa616efdf9482daf4fe53e1e',
  16         'info_dict': {
  17             'id': 'agbDDi7WZTV',
  18             'ext': 'mp4',
  19             'title': 'Michelle Lewin in Miami Beach',
  20             'duration': 240,
  21             'age_limit': 18,
  22         }
  23     }
  24
  25     def _real_extract(self, url):
  26         mobj = re.match(self._VALID_URL, url)
  27         video_id = mobj.group('id')
  28
  29         webpage = self._download_webpage(url, video_id)
  30
  31         video_title = None
  32         duration = None
  33         video_thumbnail = None
  34         formats = []
  35
  36         # most of the information is stored in the flashvars
  37         flashvars = self._html_search_regex(
  38             r'flashvars="(.+?)"', webpage, 'flashvars')
  39
  40         infos = compat_urllib_parse.unquote(flashvars).split(r'&')
  41         for info in infos:
  42             videovars_match = re.match(r'^video_vars\[(.+?)\]=(.+?)$', info)
  43             if videovars_match:
  44                 key = videovars_match.group(1)
  45                 val = videovars_match.group(2)
  46
  47                 if key == 'title':
  48                     video_title = compat_urllib_parse.unquote_plus(val)
  49                 if key == 'duration':
  50                     try:
  51                         duration = int(val)
  52                     except ValueError:
  53                         pass
  54                 if key == 'big_thumb':
  55                     video_thumbnail = val
  56
  57                 videourl_match = re.match(
  58                     r'^video_urls\]\[(?P<resolution>[0-9]+)p', key)
  59                 if videourl_match:
  60                     height = int(videourl_match.group('resolution'))
  61                     formats.append({
  62                         'height': height,
  63                         'url': val,
  64                     })
  65         self._sort_formats(formats)
  66
  67         # Extract title - should be in the flashvars; if not, look elsewhere
  68         if video_title is None:
  69             video_title = self._html_search_regex(
  70                 r'<title>(.*?)</title', webpage, 'title')
  71
  72         return {
  73             'id': video_id,
  74             'formats': formats,
  75             'title': video_title,
  76             'thumbnail': video_thumbnail,
  77             'duration': duration,
  78             'description': None,
  79             'age_limit': 18
  80         }