yt_dlp/extractor/flickr.py

   1 import urllib.parse
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     ExtractorError,
   6     format_field,
   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             'uploader_url': 'https://www.flickr.com/photos/forestwander-nature-pictures/',
  28             'comment_count': int,
  29             'view_count': int,
  30             'tags': list,
  31             'license': 'Attribution-ShareAlike',
  32         },
  33     }
  34     _API_BASE_URL = 'https://api.flickr.com/services/rest?'
  35     # https://help.yahoo.com/kb/flickr/SLN25525.html
  36     _LICENSES = {
  37         '0': 'All Rights Reserved',
  38         '1': 'Attribution-NonCommercial-ShareAlike',
  39         '2': 'Attribution-NonCommercial',
  40         '3': 'Attribution-NonCommercial-NoDerivs',
  41         '4': 'Attribution',
  42         '5': 'Attribution-ShareAlike',
  43         '6': 'Attribution-NoDerivs',
  44         '7': 'No known copyright restrictions',
  45         '8': 'United States government work',
  46         '9': 'Public Domain Dedication (CC0)',
  47         '10': 'Public Domain Work',
  48     }
  49
  50     def _call_api(self, method, video_id, api_key, note, secret=None):
  51         query = {
  52             'photo_id': video_id,
  53             'method': f'flickr.{method}',
  54             'api_key': api_key,
  55             'format': 'json',
  56             'nojsoncallback': 1,
  57         }
  58         if secret:
  59             query['secret'] = secret
  60         data = self._download_json(self._API_BASE_URL + urllib.parse.urlencode(query), video_id, note)
  61         if data['stat'] != 'ok':
  62             raise ExtractorError(data['message'])
  63         return data
  64
  65     def _real_extract(self, url):
  66         video_id = self._match_id(url)
  67
  68         api_key = self._download_json(
  69             'https://www.flickr.com/hermes_error_beacon.gne', video_id,
  70             'Downloading api key')['site_key']
  71
  72         video_info = self._call_api(
  73             'photos.getInfo', video_id, api_key, 'Downloading video info')['photo']
  74         if video_info['media'] == 'video':
  75             streams = self._call_api(
  76                 'video.getStreamInfo', video_id, api_key,
  77                 'Downloading streams info', video_info['secret'])['streams']
  78
  79             preference = qualities(
  80                 ['288p', 'iphone_wifi', '100', '300', '700', '360p', 'appletv', '720p', '1080p', 'orig'])
  81
  82             formats = []
  83             for stream in streams['stream']:
  84                 stream_type = str(stream.get('type'))
  85                 formats.append({
  86                     'format_id': stream_type,
  87                     'url': stream['_content'],
  88                     'quality': preference(stream_type),
  89                 })
  90
  91             owner = video_info.get('owner', {})
  92             uploader_id = owner.get('nsid')
  93             uploader_path = owner.get('path_alias') or uploader_id
  94             uploader_url = format_field(uploader_path, None, 'https://www.flickr.com/photos/%s/')
  95
  96             return {
  97                 'id': video_id,
  98                 'title': video_info['title']['_content'],
  99                 'description': video_info.get('description', {}).get('_content'),
 100                 'formats': formats,
 101                 'timestamp': int_or_none(video_info.get('dateuploaded')),
 102                 'duration': int_or_none(video_info.get('video', {}).get('duration')),
 103                 'uploader_id': uploader_id,
 104                 'uploader': owner.get('realname'),
 105                 'uploader_url': uploader_url,
 106                 'comment_count': int_or_none(video_info.get('comments', {}).get('_content')),
 107                 'view_count': int_or_none(video_info.get('views')),
 108                 'tags': [tag.get('_content') for tag in video_info.get('tags', {}).get('tag', [])],
 109                 'license': self._LICENSES.get(video_info.get('license')),
 110             }
 111         else:
 112             raise ExtractorError('not a video', expected=True)