yt_dlp/extractor/twentythreevideo.py

   1 from .common import InfoExtractor
   2 from ..utils import int_or_none
   3
   4
   5 class TwentyThreeVideoIE(InfoExtractor):
   6     IE_NAME = '23video'
   7     _VALID_URL = r'https?://(?P<domain>[^.]+\.(?:twentythree\.net|23video\.com|filmweb\.no))/v\.ihtml/player\.html\?(?P<query>.*?\bphoto(?:_|%5f)id=(?P<id>\d+).*)'
   8     _TESTS = [{
   9         'url': 'https://video.twentythree.net/v.ihtml/player.html?showDescriptions=0&source=site&photo%5fid=20448876&autoPlay=1',
  10         'md5': '75fcf216303eb1dae9920d651f85ced4',
  11         'info_dict': {
  12             'id': '20448876',
  13             'ext': 'mp4',
  14             'title': 'Video Marketing Minute: Personalized Video',
  15             'timestamp': 1513855354,
  16             'upload_date': '20171221',
  17             'uploader_id': '12258964',
  18             'uploader': 'Rasmus Bysted',
  19         },
  20     }, {
  21         'url': 'https://bonnier-publications-danmark.23video.com/v.ihtml/player.html?token=f0dc46476e06e13afd5a1f84a29e31e8&source=embed&photo%5fid=36137620',
  22         'only_matching': True,
  23     }]
  24
  25     def _real_extract(self, url):
  26         domain, query, photo_id = self._match_valid_url(url).groups()
  27         base_url = f'https://{domain}'
  28         photo_data = self._download_json(
  29             base_url + '/api/photo/list?' + query, photo_id, query={
  30                 'format': 'json',
  31             }, transform_source=lambda s: self._search_regex(r'(?s)({.+})', s, 'photo data'))['photo']
  32         title = photo_data['title']
  33
  34         formats = []
  35
  36         audio_path = photo_data.get('audio_download')
  37         if audio_path:
  38             formats.append({
  39                 'format_id': 'audio',
  40                 'url': base_url + audio_path,
  41                 'filesize': int_or_none(photo_data.get('audio_size')),
  42                 'vcodec': 'none',
  43             })
  44
  45         def add_common_info_to_list(l, template, id_field, id_value):
  46             f_base = template % id_value
  47             f_path = photo_data.get(f_base + 'download')
  48             if not f_path:
  49                 return
  50             l.append({
  51                 id_field: id_value,
  52                 'url': base_url + f_path,
  53                 'width': int_or_none(photo_data.get(f_base + 'width')),
  54                 'height': int_or_none(photo_data.get(f_base + 'height')),
  55                 'filesize': int_or_none(photo_data.get(f_base + 'size')),
  56             })
  57
  58         for f in ('mobile_high', 'medium', 'hd', '1080p', '4k'):
  59             add_common_info_to_list(formats, 'video_%s_', 'format_id', f)
  60
  61         thumbnails = []
  62         for t in ('quad16', 'quad50', 'quad75', 'quad100', 'small', 'portrait', 'standard', 'medium', 'large', 'original'):
  63             add_common_info_to_list(thumbnails, '%s_', 'id', t)
  64
  65         return {
  66             'id': photo_id,
  67             'title': title,
  68             'timestamp': int_or_none(photo_data.get('creation_date_epoch')),
  69             'duration': int_or_none(photo_data.get('video_length')),
  70             'view_count': int_or_none(photo_data.get('view_count')),
  71             'comment_count': int_or_none(photo_data.get('number_of_comments')),
  72             'uploader_id': photo_data.get('user_id'),
  73             'uploader': photo_data.get('display_name'),
  74             'thumbnails': thumbnails,
  75             'formats': formats,
  76         }