youtube_dl/extractor/vidme.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     int_or_none,
   6     float_or_none,
   7     str_to_int,
   8 )
   9
  10
  11 class VidmeIE(InfoExtractor):
  12     _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z]+)'
  13     _TESTS = [{
  14         'url': 'https://vid.me/QNB',
  15         'md5': 'f42d05e7149aeaec5c037b17e5d3dc82',
  16         'info_dict': {
  17             'id': 'QNB',
  18             'ext': 'mp4',
  19             'title': 'Fishing for piranha - the easy way',
  20             'description': 'source: https://www.facebook.com/photo.php?v=312276045600871',
  21             'duration': 119.92,
  22             'timestamp': 1406313244,
  23             'upload_date': '20140725',
  24             'thumbnail': 're:^https?://.*\.jpg',
  25         },
  26     }, {
  27         # From http://naked-yogi.tumblr.com/post/118312946248/naked-smoking-stretching
  28         'url': 'https://vid.me/e/Wmur',
  29         'only_matching': True,
  30     }]
  31
  32     def _real_extract(self, url):
  33         url = url.replace('vid.me/e/', 'vid.me/')
  34         video_id = self._match_id(url)
  35         webpage = self._download_webpage(url, video_id)
  36
  37         video_url = self._html_search_regex(
  38             r'<source src="([^"]+)"', webpage, 'video URL')
  39
  40         title = self._og_search_title(webpage)
  41         description = self._og_search_description(webpage, default='')
  42         thumbnail = self._og_search_thumbnail(webpage)
  43         timestamp = int_or_none(self._og_search_property('updated_time', webpage, fatal=False))
  44         width = int_or_none(self._og_search_property('video:width', webpage, fatal=False))
  45         height = int_or_none(self._og_search_property('video:height', webpage, fatal=False))
  46         duration = float_or_none(self._html_search_regex(
  47             r'data-duration="([^"]+)"', webpage, 'duration', fatal=False))
  48         view_count = str_to_int(self._html_search_regex(
  49             r'<(?:li|span) class="video_views">\s*([\d,\.]+)\s*plays?', webpage, 'view count', fatal=False))
  50         like_count = str_to_int(self._html_search_regex(
  51             r'class="score js-video-vote-score"[^>]+data-score="([\d,\.\s]+)">',
  52             webpage, 'like count', fatal=False))
  53
  54         return {
  55             'id': video_id,
  56             'url': video_url,
  57             'title': title,
  58             'description': description,
  59             'thumbnail': thumbnail,
  60             'timestamp': timestamp,
  61             'width': width,
  62             'height': height,
  63             'duration': duration,
  64             'view_count': view_count,
  65             'like_count': like_count,
  66         }