youtube_dl/extractor/flickr.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..compat import (
   5     compat_str,
   6     compat_urllib_parse_urlencode,
   7 )
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     qualities,
  12 )
  13
  14
  15 class FlickrIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.|secure\.)?flickr\.com/photos/[\w\-_@]+/(?P<id>\d+)'
  17     _TEST = {
  18         'url': 'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
  19         'md5': '164fe3fa6c22e18d448d4d5af2330f31',
  20         'info_dict': {
  21             'id': '5645318632',
  22             'ext': 'mpg',
  23             '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.',
  24             'title': 'Dark Hollow Waterfalls',
  25             'duration': 19,
  26             'timestamp': 1303528740,
  27             'upload_date': '20110423',
  28             'uploader_id': '10922353@N03',
  29             'uploader': 'Forest Wander',
  30             'uploader_url': 'https://www.flickr.com/photos/forestwander-nature-pictures/',
  31             'comment_count': int,
  32             'view_count': int,
  33             'tags': list,
  34             'license': 'Attribution-ShareAlike',
  35         }
  36     }
  37     _API_BASE_URL = 'https://api.flickr.com/services/rest?'
  38     # https://help.yahoo.com/kb/flickr/SLN25525.html
  39     _LICENSES = {
  40         '0': 'All Rights Reserved',
  41         '1': 'Attribution-NonCommercial-ShareAlike',
  42         '2': 'Attribution-NonCommercial',
  43         '3': 'Attribution-NonCommercial-NoDerivs',
  44         '4': 'Attribution',
  45         '5': 'Attribution-ShareAlike',
  46         '6': 'Attribution-NoDerivs',
  47         '7': 'No known copyright restrictions',
  48         '8': 'United States government work',
  49         '9': 'Public Domain Dedication (CC0)',
  50         '10': 'Public Domain Work',
  51     }
  52
  53     def _call_api(self, method, video_id, api_key, note, secret=None):
  54         query = {
  55             'photo_id': video_id,
  56             'method': 'flickr.%s' % method,
  57             'api_key': api_key,
  58             'format': 'json',
  59             'nojsoncallback': 1,
  60         }
  61         if secret:
  62             query['secret'] = secret
  63         data = self._download_json(self._API_BASE_URL + compat_urllib_parse_urlencode(query), video_id, note)
  64         if data['stat'] != 'ok':
  65             raise ExtractorError(data['message'])
  66         return data
  67
  68     def _real_extract(self, url):
  69         video_id = self._match_id(url)
  70
  71         api_key = self._download_json(
  72             'https://www.flickr.com/hermes_error_beacon.gne', video_id,
  73             'Downloading api key')['site_key']
  74
  75         video_info = self._call_api(
  76             'photos.getInfo', video_id, api_key, 'Downloading video info')['photo']
  77         if video_info['media'] == 'video':
  78             streams = self._call_api(
  79                 'video.getStreamInfo', video_id, api_key,
  80                 'Downloading streams info', video_info['secret'])['streams']
  81
  82             preference = qualities(
  83                 ['288p', 'iphone_wifi', '100', '300', '700', '360p', 'appletv', '720p', '1080p', 'orig'])
  84
  85             formats = []
  86             for stream in streams['stream']:
  87                 stream_type = compat_str(stream.get('type'))
  88                 formats.append({
  89                     'format_id': stream_type,
  90                     'url': stream['_content'],
  91                     'preference': preference(stream_type),
  92                 })
  93             self._sort_formats(formats)
  94
  95             owner = video_info.get('owner', {})
  96             uploader_id = owner.get('nsid')
  97             uploader_path = owner.get('path_alias') or uploader_id
  98             uploader_url = 'https://www.flickr.com/photos/%s/' % uploader_path if uploader_path else None
  99
 100             return {
 101                 'id': video_id,
 102                 'title': video_info['title']['_content'],
 103                 'description': video_info.get('description', {}).get('_content'),
 104                 'formats': formats,
 105                 'timestamp': int_or_none(video_info.get('dateuploaded')),
 106                 'duration': int_or_none(video_info.get('video', {}).get('duration')),
 107                 'uploader_id': uploader_id,
 108                 'uploader': owner.get('realname'),
 109                 'uploader_url': uploader_url,
 110                 'comment_count': int_or_none(video_info.get('comments', {}).get('_content')),
 111                 'view_count': int_or_none(video_info.get('views')),
 112                 'tags': [tag.get('_content') for tag in video_info.get('tags', {}).get('tag', [])],
 113                 'license': self._LICENSES.get(video_info.get('license')),
 114             }
 115         else:
 116             raise ExtractorError('not a video', expected=True)