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

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_urllib_parse
   5 from ..utils import (
   6     ExtractorError,
   7     int_or_none,
   8     qualities,
   9 )
  10
  11
  12 class FlickrIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.|secure\.)?flickr\.com/photos/[\w\-_@]+/(?P<id>\d+)'
  14     _TEST = {
  15         'url': 'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
  16         'md5': '164fe3fa6c22e18d448d4d5af2330f31',
  17         'info_dict': {
  18             'id': '5645318632',
  19             'ext': 'mpg',
  20             'description': 'Waterfalls in the Springtime at Dark Hollow Waterfalls. These are located just off of Skyline Drive in Virginia. They are only about 6/10 of a mile hike but it is a pretty steep hill and a good climb back up.',
  21             'title': 'Dark Hollow Waterfalls',
  22             'duration': 19,
  23             'timestamp': 1303528740,
  24             'upload_date': '20110423',
  25             'uploader_id': '10922353@N03',
  26             'uploader': 'Forest Wander',
  27             'comment_count': int,
  28             'view_count': int,
  29             'tags': list,
  30         }
  31     }
  32
  33     _API_BASE_URL = 'https://api.flickr.com/services/rest?'
  34
  35     def _call_api(self, method, video_id, api_key, note, secret=None):
  36         query = {
  37             'photo_id': video_id,
  38             'method': 'flickr.%s' % method,
  39             'api_key': api_key,
  40             'format': 'json',
  41             'nojsoncallback': 1,
  42         }
  43         if secret:
  44             query['secret'] = secret
  45         data = self._download_json(self._API_BASE_URL + compat_urllib_parse.urlencode(query), video_id, note)
  46         if data['stat'] != 'ok':
  47             raise ExtractorError(data['message'])
  48         return data
  49
  50     def _real_extract(self, url):
  51         video_id = self._match_id(url)
  52
  53         api_key = self._download_json('https://www.flickr.com/hermes_error_beacon.gne', video_id, 'Downloading api key',)['site_key']
  54
  55         video_info = self._call_api('photos.getInfo', video_id, api_key, 'Downloading video info')['photo']
  56         if video_info['media'] == 'video':
  57             streams = self._call_api('video.getStreamInfo', video_id, api_key, 'Downloading streams info', video_info['secret'])['streams']
  58
  59             preference = qualities(['iphone_wifi', '700', 'appletv', 'orig'])
  60
  61             formats = []
  62             for stream in streams['stream']:
  63                 stream_type = str(stream.get('type'))
  64                 formats.append({
  65                     'format_id': stream_type,
  66                     'url': stream['_content'],
  67                     'preference': preference(stream_type),
  68                 })
  69             self._sort_formats(formats)
  70
  71             owner = video_info.get('owner', {})
  72
  73             return {
  74                 'id': video_id,
  75                 'title': video_info['title']['_content'],
  76                 'description': video_info.get('description', {}).get('_content'),
  77                 'formats': formats,
  78                 'timestamp': int_or_none(video_info.get('dateuploaded')),
  79                 'duration': int_or_none(video_info.get('video', {}).get('duration')),
  80                 'uploader_id': owner.get('nsid'),
  81                 'uploader': owner.get('realname'),
  82                 'comment_count': int_or_none(video_info.get('comments', {}).get('_content')),
  83                 'view_count': int_or_none(video_info.get('views')),
  84                 'tags': [tag.get('_content') for tag in video_info.get('tags', {}).get('tag', [])]
  85             }
  86         else:
  87             raise ExtractorError('not a video', expected=True)