youtube_dlc/extractor/twentythreevideo.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import int_or_none
   7
   8
   9 class TwentyThreeVideoIE(InfoExtractor):
  10     IE_NAME = '23video'
  11     _VALID_URL = r'https?://(?P<domain>[^.]+\.(?:twentythree\.net|23video\.com|filmweb\.no))/v\.ihtml/player\.html\?(?P<query>.*?\bphoto(?:_|%5f)id=(?P<id>\d+).*)'
  12     _TESTS = [{
  13         'url': 'https://video.twentythree.net/v.ihtml/player.html?showDescriptions=0&source=site&photo%5fid=20448876&autoPlay=1',
  14         'md5': '75fcf216303eb1dae9920d651f85ced4',
  15         'info_dict': {
  16             'id': '20448876',
  17             'ext': 'mp4',
  18             'title': 'Video Marketing Minute: Personalized Video',
  19             'timestamp': 1513855354,
  20             'upload_date': '20171221',
  21             'uploader_id': '12258964',
  22             'uploader': 'Rasmus Bysted',
  23         }
  24     }, {
  25         'url': 'https://bonnier-publications-danmark.23video.com/v.ihtml/player.html?token=f0dc46476e06e13afd5a1f84a29e31e8&source=embed&photo%5fid=36137620',
  26         'only_matching': True,
  27     }]
  28
  29     def _real_extract(self, url):
  30         domain, query, photo_id = re.match(self._VALID_URL, url).groups()
  31         base_url = 'https://%s' % domain
  32         photo_data = self._download_json(
  33             base_url + '/api/photo/list?' + query, photo_id, query={
  34                 'format': 'json',
  35             }, transform_source=lambda s: self._search_regex(r'(?s)({.+})', s, 'photo data'))['photo']
  36         title = photo_data['title']
  37
  38         formats = []
  39
  40         audio_path = photo_data.get('audio_download')
  41         if audio_path:
  42             formats.append({
  43                 'format_id': 'audio',
  44                 'url': base_url + audio_path,
  45                 'filesize': int_or_none(photo_data.get('audio_size')),
  46                 'vcodec': 'none',
  47             })
  48
  49         def add_common_info_to_list(l, template, id_field, id_value):
  50             f_base = template % id_value
  51             f_path = photo_data.get(f_base + 'download')
  52             if not f_path:
  53                 return
  54             l.append({
  55                 id_field: id_value,
  56                 'url': base_url + f_path,
  57                 'width': int_or_none(photo_data.get(f_base + 'width')),
  58                 'height': int_or_none(photo_data.get(f_base + 'height')),
  59                 'filesize': int_or_none(photo_data.get(f_base + 'size')),
  60             })
  61
  62         for f in ('mobile_high', 'medium', 'hd', '1080p', '4k'):
  63             add_common_info_to_list(formats, 'video_%s_', 'format_id', f)
  64
  65         thumbnails = []
  66         for t in ('quad16', 'quad50', 'quad75', 'quad100', 'small', 'portrait', 'standard', 'medium', 'large', 'original'):
  67             add_common_info_to_list(thumbnails, '%s_', 'id', t)
  68
  69         return {
  70             'id': photo_id,
  71             'title': title,
  72             'timestamp': int_or_none(photo_data.get('creation_date_epoch')),
  73             'duration': int_or_none(photo_data.get('video_length')),
  74             'view_count': int_or_none(photo_data.get('view_count')),
  75             'comment_count': int_or_none(photo_data.get('number_of_comments')),
  76             'uploader_id': photo_data.get('user_id'),
  77             'uploader': photo_data.get('display_name'),
  78             'thumbnails': thumbnails,
  79             'formats': formats,
  80         }