yt_dlp/extractor/zhihu.py

   1 from .common import InfoExtractor
   2 from ..utils import float_or_none, format_field, int_or_none
   3
   4
   5 class ZhihuIE(InfoExtractor):
   6     _VALID_URL = r'https?://(?:www\.)?zhihu\.com/zvideo/(?P<id>[0-9]+)'
   7     _TEST = {
   8         'url': 'https://www.zhihu.com/zvideo/1342930761977176064',
   9         'md5': 'c8d4c9cd72dd58e6f9bc9c2c84266464',
  10         'info_dict': {
  11             'id': '1342930761977176064',
  12             'ext': 'mp4',
  13             'title': '写春联也太难了吧！',
  14             'thumbnail': r're:^https?://.*\.jpg',
  15             'uploader': '桥半舫',
  16             'timestamp': 1612959715,
  17             'upload_date': '20210210',
  18             'uploader_id': '244ecb13b0fd7daf92235288c8ca3365',
  19             'duration': 146.333,
  20             'view_count': int,
  21             'like_count': int,
  22             'comment_count': int,
  23         },
  24     }
  25
  26     def _real_extract(self, url):
  27         video_id = self._match_id(url)
  28         zvideo = self._download_json(
  29             'https://www.zhihu.com/api/v4/zvideos/' + video_id, video_id)
  30         title = zvideo['title']
  31         video = zvideo.get('video') or {}
  32
  33         formats = []
  34         for format_id, q in (video.get('playlist') or {}).items():
  35             play_url = q.get('url') or q.get('play_url')
  36             if not play_url:
  37                 continue
  38             formats.append({
  39                 'asr': int_or_none(q.get('sample_rate')),
  40                 'filesize': int_or_none(q.get('size')),
  41                 'format_id': format_id,
  42                 'fps': int_or_none(q.get('fps')),
  43                 'height': int_or_none(q.get('height')),
  44                 'tbr': float_or_none(q.get('bitrate')),
  45                 'url': play_url,
  46                 'width': int_or_none(q.get('width')),
  47             })
  48
  49         author = zvideo.get('author') or {}
  50         url_token = author.get('url_token')
  51
  52         return {
  53             'id': video_id,
  54             'title': title,
  55             'formats': formats,
  56             'thumbnail': video.get('thumbnail') or zvideo.get('image_url'),
  57             'uploader': author.get('name'),
  58             'timestamp': int_or_none(zvideo.get('published_at')),
  59             'uploader_id': author.get('id'),
  60             'uploader_url': format_field(url_token, None, 'https://www.zhihu.com/people/%s'),
  61             'duration': float_or_none(video.get('duration')),
  62             'view_count': int_or_none(zvideo.get('play_count')),
  63             'like_count': int_or_none(zvideo.get('liked_count')),
  64             'comment_count': int_or_none(zvideo.get('comment_count')),
  65         }