用 python 抓取知乎指定回答下的視頻

前言

如今知乎容許上傳視頻，奈何不能下載視頻，好氣哦，無奈之下研究一下了，而後擼了代碼，方便下載視頻保存。

接下來以 貓爲何一點也不怕蛇？ 回答爲例，分享一下整個下載過程。

調試一下

打開 F12, 找到光標，以下圖：

而後將光標移動到視頻上。以下圖：

咦這是什麼？視野中出現了一條神祕的連接: https://www.zhihu.com/video/xxxxx，讓咱們將這條連接複製到瀏覽器上，而後打開：

彷佛這就是咱們要找的視頻，不要着急，讓咱們看一看，網頁的請求，而後你會發現一個頗有意思的請求(重點來了):

讓咱們本身看一下數據吧:

{
    "playlist": {
        "ld": {
            "width": 360,
            "format": "mp4",
            "play_url": "https://vdn.vzuu.com/LD/05fc411e-d8e0-11e8-bb8b-0242ac112a0b.mp4?auth_key=1541477643-0-0-987c2c504d14ab1165ce2ed47759d927&expiration=1541477643&disable_local_cache=1",
            "duration": 17,
            "size": 1123111,
            "bitrate": 509,
            "height": 640
        },
        "hd": {
            "width": 720,
            "format": "mp4",
            "play_url": "https://vdn.vzuu.com/HD/05fc411e-d8e0-11e8-bb8b-0242ac112a0b.mp4?auth_key=1541477643-0-0-8b8024a22a62f097ca31b8b06b7233a1&expiration=1541477643&disable_local_cache=1",
            "duration": 17,
            "size": 4354364,
            "bitrate": 1974,
            "height": 1280
        },
        "sd": {
            "width": 480,
            "format": "mp4",
            "play_url": "https://vdn.vzuu.com/SD/05fc411e-d8e0-11e8-bb8b-0242ac112a0b.mp4?auth_key=1541477643-0-0-5948c2562d817218c9a9fc41abad1df8&expiration=1541477643&disable_local_cache=1",
            "duration": 17,
            "size": 1920976,
            "bitrate": 871,
            "height": 848
        }
    },
    "title": "",
    "duration": 17,
    "cover_info": {
        "width": 720,
        "thumbnail": "https://pic2.zhimg.com/80/v2-97b9435a0c32d01c7c931bd00120327d_b.jpg",
        "height": 1280
    },
    "type": "video",
    "id": "1039146361396174848",
    "misc_info": {}
}

沒錯了，咱們要下載的視頻就在這裏面，其中 ld 表明普清,sd 表明標清, hd 表明高清，把相應連接再次在瀏覽器打開，而後右鍵保存就能夠下載視頻了。

代碼

知道整個流程是什麼樣子，接下來擼代碼的過程就簡單了，這裏就不過再作過多解釋了，直接上代碼:

# -*- encoding: utf-8 -*-

import re
import requests
import uuid
import datetime


class DownloadVideo:

    __slots__ = [
        'url', 'video_name', 'url_format', 'download_url', 'video_number',
        'video_api', 'clarity_list', 'clarity'
    ]

    def __init__(self, url, clarity='ld', video_name=None):
        self.url = url
        self.video_name = video_name
        self.url_format = "https://www.zhihu.com/question/\d+/answer/\d+"
        self.clarity = clarity
        self.clarity_list = ['ld', 'sd', 'hd']
        self.video_api = 'https://lens.zhihu.com/api/videos'

    def check_url_format(self):
        pattern = re.compile(self.url_format)
        matches = re.match(pattern, self.url)
        if matches is None:
            raise ValueError(
                "連接格式應符合:https://www.zhihu.com/question/{number}/answer/{number}"
            )
        return True

    def get_video_number(self):
        try:
            headers = {
                'User-Agent':
                'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/66.0.3359.181 Safari/537.36'
            }
            response = requests.get(self.url, headers=headers)
            response.encoding = 'utf-8'
            html = response.text
            video_ids = re.findall(r'data-lens-id="(\d+)"', html)
            if video_ids:
                video_id_list = list(set([video_id for video_id in video_ids]))
                self.video_number = video_id_list[0]
                return self
            raise ValueError("獲取視頻編號異常:{}".format(self.url))
        except Exception as e:
            raise Exception(e)

    def get_video_url_by_number(self):
        url = "{}/{}".format(self.video_api, self.video_number)

        headers = {}
        headers['Referer'] = 'https://v.vzuu.com/video/{}'.format(
            self.video_number)
        headers['Origin'] = 'https://v.vzuu.com'
        headers[
            'User-Agent'] = 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.67 Safari/537.36'
        headers['Content-Type'] = 'application/json'

        try:
            response = requests.get(url, headers=headers)
            response_dict = response.json()
            if self.clarity in response_dict['playlist']:
                self.download_url = response_dict['playlist'][
                    self.clarity]['play_url']
            else:
                for clarity in self.clarity_list:
                    if clarity in response_dict['playlist']:
                        self.download_url = response_dict['playlist'][
                            self.clarity]['play_url']
                        break
            return self
        except Exception as e:
            raise Exception(e)

    def get_video_by_video_url(self):
        response = requests.get(self.download_url)
        datetime_str = datetime.datetime.now().strftime("%Y-%m-%d %H-%M-%S")
        if self.video_name is not None:
            video_name = "{}-{}.mp4".format(self.video_name, datetime_str)
        else:
            video_name = "{}-{}.mp4".format(str(uuid.uuid1()), datetime_str)
        path = "{}".format(video_name)
        with open(path, 'wb') as f:
            f.write(response.content)

    def download_video(self):

        if self.clarity not in self.clarity_list:
            raise ValueError("清晰度參數異常,僅支持:ld(普清),sd(標清),hd(高清)")

        if self.check_url_format():
            return self.get_video_number().get_video_url_by_number().get_video_by_video_url()


if __name__ == '__main__':
    a = DownloadVideo('https://www.zhihu.com/question/53031925/answer/524158069')
    print(a.download_video())

結語

代碼還有優化空間，這裏面我只是下載了回答中的第一個視頻，理論上應該存在一個回答下能夠有多個視頻的。若是還有什麼疑問或者建議，能夠多多交流。