youtube_dl/extractor/vine.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import unified_strdate
   8
   9
  10 class VineIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?vine\.co/v/(?P<id>\w+)'
  12     _TEST = {
  13         'url': 'https://vine.co/v/b9KOOWX7HUx',
  14         'md5': '2f36fed6235b16da96ce9b4dc890940d',
  15         'info_dict': {
  16             'id': 'b9KOOWX7HUx',
  17             'ext': 'mp4',
  18             'title': 'Chicken.',
  19             'description': 'Chicken.',
  20             'upload_date': '20130519',
  21             'uploader': 'Jack Dorsey',
  22             'uploader_id': '76',
  23         },
  24     }
  25
  26     def _real_extract(self, url):
  27         mobj = re.match(self._VALID_URL, url)
  28         video_id = mobj.group('id')
  29
  30         webpage = self._download_webpage('https://vine.co/v/' + video_id, video_id)
  31
  32         data = json.loads(self._html_search_regex(
  33             r'window\.POST_DATA = { %s: ({.+?}) }' % video_id, webpage, 'vine data'))
  34
  35         formats = [
  36             {
  37                 'url': data['videoLowURL'],
  38                 'ext': 'mp4',
  39                 'format_id': 'low',
  40             },
  41             {
  42                 'url': data['videoUrl'],
  43                 'ext': 'mp4',
  44                 'format_id': 'standard',
  45             }
  46         ]
  47
  48         return {
  49             'id': video_id,
  50             'title': self._og_search_title(webpage),
  51             'description': data['description'],
  52             'thumbnail': data['thumbnailUrl'],
  53             'upload_date': unified_strdate(data['created']),
  54             'uploader': data['username'],
  55             'uploader_id': data['userIdStr'],
  56             'like_count': data['likes']['count'],
  57             'comment_count': data['comments']['count'],
  58             'repost_count': data['reposts']['count'],
  59             'formats': formats,
  60         }