1
1
mirror of https://github.com/ytdl-org/youtube-dl synced 2024-11-30 12:32:56 +01:00
youtube-dl/youtube_dl/extractor/zhihu.py
2021-02-16 10:08:43 +01:00

70 lines
2.6 KiB
Python

# coding: utf-8
from __future__ import unicode_literals
from .common import InfoExtractor
from ..utils import float_or_none, int_or_none
class ZhihuIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?zhihu\.com/zvideo/(?P<id>[0-9]+)'
_TEST = {
'url': 'https://www.zhihu.com/zvideo/1342930761977176064',
'md5': 'c8d4c9cd72dd58e6f9bc9c2c84266464',
'info_dict': {
'id': '1342930761977176064',
'ext': 'mp4',
'title': '写春联也太难了吧!',
'thumbnail': r're:^https?://.*\.jpg',
'uploader': '桥半舫',
'timestamp': 1612959715,
'upload_date': '20210210',
'uploader_id': '244ecb13b0fd7daf92235288c8ca3365',
'duration': 146.333,
'view_count': int,
'like_count': int,
'comment_count': int,
}
}
def _real_extract(self, url):
video_id = self._match_id(url)
zvideo = self._download_json(
'https://www.zhihu.com/api/v4/zvideos/' + video_id, video_id)
title = zvideo['title']
video = zvideo.get('video') or {}
formats = []
for format_id, q in (video.get('playlist') or {}).items():
play_url = q.get('url') or q.get('play_url')
if not play_url:
continue
formats.append({
'asr': int_or_none(q.get('sample_rate')),
'filesize': int_or_none(q.get('size')),
'format_id': format_id,
'fps': int_or_none(q.get('fps')),
'height': int_or_none(q.get('height')),
'tbr': float_or_none(q.get('bitrate')),
'url': play_url,
'width': int_or_none(q.get('width')),
})
self._sort_formats(formats)
author = zvideo.get('author') or {}
url_token = author.get('url_token')
return {
'id': video_id,
'title': title,
'formats': formats,
'thumbnail': video.get('thumbnail') or zvideo.get('image_url'),
'uploader': author.get('name'),
'timestamp': int_or_none(zvideo.get('published_at')),
'uploader_id': author.get('id'),
'uploader_url': 'https://www.zhihu.com/people/' + url_token if url_token else None,
'duration': float_or_none(video.get('duration')),
'view_count': int_or_none(zvideo.get('play_count')),
'like_count': int_or_none(zvideo.get('liked_count')),
'comment_count': int_or_none(zvideo.get('comment_count')),
}