_ Git - youtube-dl/blob - youtube_dl/extractor/nytimes.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     float_or_none,
   6     int_or_none,
   7     parse_iso8601,
   8 )
   9
  10
  11 class NYTimesIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:(?:www\.)?nytimes\.com/video/(?:[^/]+/)+?|graphics8\.nytimes\.com/bcvideo/\d+(?:\.\d+)?/iframe/embed\.html\?videoId=)(?P<id>\d+)'
  13
  14     _TESTS = [{
  15         'url': 'http://www.nytimes.com/video/opinion/100000002847155/verbatim-what-is-a-photocopier.html?playlistId=100000001150263',
  16         'md5': '18a525a510f942ada2720db5f31644c0',
  17         'info_dict': {
  18             'id': '100000002847155',
  19             'ext': 'mov',
  20             'title': 'Verbatim: What Is a Photocopier?',
  21             'description': 'md5:93603dada88ddbda9395632fdc5da260',
  22             'timestamp': 1398631707,
  23             'upload_date': '20140427',
  24             'uploader': 'Brett Weiner',
  25             'duration': 419,
  26         }
  27     }, {
  28         'url': 'http://www.nytimes.com/video/travel/100000003550828/36-hours-in-dubai.html',
  29         'only_matching': True,
  30     }]
  31
  32     def _real_extract(self, url):
  33         video_id = self._match_id(url)
  34
  35         video_data = self._download_json(
  36             'http://www.nytimes.com/svc/video/api/v2/video/%s' % video_id,
  37             video_id, 'Downloading video JSON')
  38
  39         title = video_data['headline']
  40         description = video_data.get('summary')
  41         duration = float_or_none(video_data.get('duration'), 1000)
  42
  43         uploader = video_data['byline']
  44         timestamp = parse_iso8601(video_data['publication_date'][:-8])
  45
  46         def get_file_size(file_size):
  47             if isinstance(file_size, int):
  48                 return file_size
  49             elif isinstance(file_size, dict):
  50                 return int(file_size.get('value', 0))
  51             else:
  52                 return 0
  53
  54         formats = [
  55             {
  56                 'url': video['url'],
  57                 'format_id': video.get('type'),
  58                 'vcodec': video.get('video_codec'),
  59                 'width': int_or_none(video.get('width')),
  60                 'height': int_or_none(video.get('height')),
  61                 'filesize': get_file_size(video.get('fileSize')),
  62             } for video in video_data['renditions']
  63         ]
  64         self._sort_formats(formats)
  65
  66         thumbnails = [
  67             {
  68                 'url': 'http://www.nytimes.com/%s' % image['url'],
  69                 'width': int_or_none(image.get('width')),
  70                 'height': int_or_none(image.get('height')),
  71             } for image in video_data['images']
  72         ]
  73
  74         return {
  75             'id': video_id,
  76             'title': title,
  77             'description': description,
  78             'timestamp': timestamp,
  79             'uploader': uploader,
  80             'duration': duration,
  81             'formats': formats,
  82             'thumbnails': thumbnails,
  83         }