summaryrefslogtreecommitdiff
path: root/youtube_dl/extractor/zhihu.py
blob: d1ed55be35a00e898b1a5daaee7797e3a7adfadf (plain)
    1 # coding: utf-8
    2 from __future__ import unicode_literals
    3 
    4 from .common import InfoExtractor
    5 from ..utils import float_or_none, int_or_none
    6 
    7 
    8 class ZhihuIE(InfoExtractor):
    9     _VALID_URL = r'https?://(?:www\.)?zhihu\.com/zvideo/(?P<id>[0-9]+)'
   10     _TEST = {
   11         'url': 'https://www.zhihu.com/zvideo/1342930761977176064',
   12         'md5': 'c8d4c9cd72dd58e6f9bc9c2c84266464',
   13         'info_dict': {
   14             'id': '1342930761977176064',
   15             'ext': 'mp4',
   16             'title': '写春联也太难了吧!',
   17             'thumbnail': r're:^https?://.*\.jpg',
   18             'uploader': '桥半舫',
   19             'timestamp': 1612959715,
   20             'upload_date': '20210210',
   21             'uploader_id': '244ecb13b0fd7daf92235288c8ca3365',
   22             'duration': 146.333,
   23             'view_count': int,
   24             'like_count': int,
   25             'comment_count': int,
   26         }
   27     }
   28 
   29     def _real_extract(self, url):
   30         video_id = self._match_id(url)
   31         zvideo = self._download_json(
   32             'https://www.zhihu.com/api/v4/zvideos/' + video_id, video_id)
   33         title = zvideo['title']
   34         video = zvideo.get('video') or {}
   35 
   36         formats = []
   37         for format_id, q in (video.get('playlist') or {}).items():
   38             play_url = q.get('url') or q.get('play_url')
   39             if not play_url:
   40                 continue
   41             formats.append({
   42                 'asr': int_or_none(q.get('sample_rate')),
   43                 'filesize': int_or_none(q.get('size')),
   44                 'format_id': format_id,
   45                 'fps': int_or_none(q.get('fps')),
   46                 'height': int_or_none(q.get('height')),
   47                 'tbr': float_or_none(q.get('bitrate')),
   48                 'url': play_url,
   49                 'width': int_or_none(q.get('width')),
   50             })
   51         self._sort_formats(formats)
   52 
   53         author = zvideo.get('author') or {}
   54         url_token = author.get('url_token')
   55 
   56         return {
   57             'id': video_id,
   58             'title': title,
   59             'formats': formats,
   60             'thumbnail': video.get('thumbnail') or zvideo.get('image_url'),
   61             'uploader': author.get('name'),
   62             'timestamp': int_or_none(zvideo.get('published_at')),
   63             'uploader_id': author.get('id'),
   64             'uploader_url': 'https://www.zhihu.com/people/' + url_token if url_token else None,
   65             'duration': float_or_none(video.get('duration')),
   66             'view_count': int_or_none(zvideo.get('play_count')),
   67             'like_count': int_or_none(zvideo.get('liked_count')),
   68             'comment_count': int_or_none(zvideo.get('comment_count')),
   69         }

Generated by cgit