youtube_dl/extractor/vidme.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     int_or_none,
   8     float_or_none,
   9     str_to_int,
  10 )
  11
  12
  13 class VidmeIE(InfoExtractor):
  14     _VALID_URL = r'https?://vid\.me/(?:e/)?(?P<id>[\da-zA-Z]+)'
  15     _TEST = {
  16         'url': 'https://vid.me/QNB',
  17         'md5': 'f42d05e7149aeaec5c037b17e5d3dc82',
  18         'info_dict': {
  19             'id': 'QNB',
  20             'ext': 'mp4',
  21             'title': 'Fishing for piranha - the easy way',
  22             'description': 'source: https://www.facebook.com/photo.php?v=312276045600871',
  23             'duration': 119.92,
  24             'timestamp': 1406313244,
  25             'upload_date': '20140725',
  26             'thumbnail': 're:^https?://.*\.jpg',
  27         },
  28     }
  29
  30     def _real_extract(self, url):
  31         mobj = re.match(self._VALID_URL, url)
  32         video_id = mobj.group('id')
  33
  34         webpage = self._download_webpage(url, video_id)
  35
  36         video_url = self._html_search_regex(r'<source src="([^"]+)"', webpage, 'video URL')
  37
  38         title = self._og_search_title(webpage)
  39         description = self._og_search_description(webpage, default='')
  40         thumbnail = self._og_search_thumbnail(webpage)
  41         timestamp = int_or_none(self._og_search_property('updated_time', webpage, fatal=False))
  42         width = int_or_none(self._og_search_property('video:width', webpage, fatal=False))
  43         height = int_or_none(self._og_search_property('video:height', webpage, fatal=False))
  44         duration = float_or_none(self._html_search_regex(
  45             r'data-duration="([^"]+)"', webpage, 'duration', fatal=False))
  46         view_count = str_to_int(self._html_search_regex(
  47             r'<span class="video_views">\s*([\d,\.]+)\s*plays?', webpage, 'view count', fatal=False))
  48         like_count = str_to_int(self._html_search_regex(
  49             r'class="score js-video-vote-score"[^>]+data-score="([\d,\.\s]+)">',
  50             webpage, 'like count', fatal=False))
  51         comment_count = str_to_int(self._html_search_regex(
  52             r'class="js-comment-count"[^>]+data-count="([\d,\.\s]+)">',
  53             webpage, 'comment count', fatal=False))
  54
  55         return {
  56             'id': video_id,
  57             'url': video_url,
  58             'title': title,
  59             'description': description,
  60             'thumbnail': thumbnail,
  61             'timestamp': timestamp,
  62             'width': width,
  63             'height': height,
  64             'duration': duration,
  65             'view_count': view_count,
  66             'like_count': like_count,
  67             'comment_count': comment_count,
  68         }