youtube_dl/extractor/nytimes.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import parse_iso8601
   7
   8
   9 class NYTimesIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?nytimes\.com/video/(?:[^/]+/)+(?P<id>\d+)'
  11
  12     _TEST = {
  13         'url': 'http://www.nytimes.com/video/opinion/100000002847155/verbatim-what-is-a-photocopier.html?playlistId=100000001150263',
  14         'md5': '18a525a510f942ada2720db5f31644c0',
  15         'info_dict': {
  16             'id': '100000002847155',
  17             'ext': 'mov',
  18             'title': 'Verbatim: What Is a Photocopier?',
  19             'description': 'md5:93603dada88ddbda9395632fdc5da260',
  20             'timestamp': 1398631707,
  21             'upload_date': '20140427',
  22             'uploader': 'Brett Weiner',
  23             'duration': 419,
  24         }
  25     }
  26
  27     def _real_extract(self, url):
  28         mobj = re.match(self._VALID_URL, url)
  29         video_id = mobj.group('id')
  30
  31         video_data = self._download_json(
  32             'http://www.nytimes.com/svc/video/api/v2/video/%s' % video_id, video_id, 'Downloading video JSON')
  33
  34         title = video_data['headline']
  35         description = video_data['summary']
  36         duration = video_data['duration'] / 1000.0
  37
  38         uploader = video_data['byline']
  39         timestamp = parse_iso8601(video_data['publication_date'][:-8])
  40
  41         def get_file_size(file_size):
  42             if isinstance(file_size, int):
  43                 return file_size
  44             elif isinstance(file_size, dict):
  45                 return int(file_size.get('value', 0))
  46             else:
  47                 return 0
  48
  49         formats = [
  50             {
  51                 'url': video['url'],
  52                 'format_id': video['type'],
  53                 'vcodec': video['video_codec'],
  54                 'width': video['width'],
  55                 'height': video['height'],
  56                 'filesize': get_file_size(video['fileSize']),
  57             } for video in video_data['renditions']
  58         ]
  59         self._sort_formats(formats)
  60
  61         thumbnails = [
  62             {
  63                 'url': 'http://www.nytimes.com/%s' % image['url'],
  64                 'resolution': '%dx%d' % (image['width'], image['height']),
  65             } for image in video_data['images']
  66         ]
  67
  68         return {
  69             'id': video_id,
  70             'title': title,
  71             'description': description,
  72             'timestamp': timestamp,
  73             'uploader': uploader,
  74             'duration': duration,
  75             'formats': formats,
  76             'thumbnails': thumbnails,
  77         }