yt_dlp/extractor/flickr.py

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