python爬蟲-b站視頻

bilibili視頻爬取

package

import requests
import json
import os
from tqdm import tqdm

如果想要調整爬取視頻的頁碼

    def Sync_run(self):
        self.num = self._cid_name()
        tn = tqdm(self.cid_list[:10])  # 如果想進行測驗, 在這里切片操作只獲取前面兩個視頻即可

在 Sync_run中將tqdm()中的cid_list進行切片操作

RUN

1. 打開想要爬取視頻的bilibili網站

2. 比如
   https://www.bilibili.com/video/BV1R7411F7JV?from=search&seid=8624852726982174897&spm_id_from=333.337.0.0
   

   • 上述連接中BV1R7411F7JV即為 需要在init函式中修改的變數self.avid, 修改該變數即可爬取這個鏈接目錄的所有視頻

運行該程式只需要將spyder中初始化的資料, 然后創建物件run即可

保存后的視頻

保存后的視頻如果想要播放需要H.264視頻解碼器, 經過測驗, 只保存videos檔案也有音頻, 可以通過potplayer 播放器，或者其他支持H.264視頻解碼器的播放器進行播放

此代碼只做學習交流使用, 請勿給服務器太大壓力

完整代碼

import requests
import json
import os
from tqdm import tqdm
import re


class spyder(object):
    def __init__(self):
        self.bvid = "BV17z4y1D7Yj"  # 在打開b站后會獲取bvid 具體使用查看readme.md
        self.video_list = "https://api.bilibili.com/x/player/pagelist?bvid={}"  # 根據bvid 可以獲取當前視頻的串列
        end = "bvid={}".format(self.bvid)  # 跟下面的self.video進行拼接, 方便后面的format只對cid進行format
        self.video = "https://api.bilibili.com/x/player/playurl?cid={}&qn=80&type=&otype=json&fourk=1&" + end
        self.headers = {
            'User-agent': "Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/94.0.4606.71 Safari/537.36",
            'referer': "https://www.bilibili.com/",
        }
        # 指定保存視頻的目錄
        self.save_file = "./data/qfjy/html"
        # 如果沒有保存檔案的目錄, 新建目錄
        if not os.path.exists(self.save_file):
            os.makedirs(self.save_file)

    # 獲取視頻串列所有視頻的cid 和name
    def _cid_name(self):
        res = requests.get(url=self.video_list.format(self.bvid), headers=self.headers).text
        j = json.loads(res)
        n = len(j['data'])
        cid_dir = [cid for cid in j['data']]
        self.cid_list = [cid['cid'] for cid in cid_dir]
        self.name_list = [cid['part'] for cid in cid_dir]
        return n

    # 決議單個cid 對應的視頻中的video和 voice， 以及視頻質量
    def _parse_cid(self, cid: str):
        u_json = self.video.format(cid)
        res = requests.get(url=u_json, headers=self.headers).text
        j = json.loads(res)
        l = j['data']['durl'][0]['backup_url']
        quality = {}
        quality['accept_description'] = j['data']['accept_description'][0]  # 默認視頻質量為最高
        quality['accept_quality'] = j['data']['accept_quality'][0]  # 默認視頻質量為最高
        video = str(l[0])
        voice = str(l[1])
        return video, voice, quality

    def Sync_run(self):
        self.num = self._cid_name()
        tn = tqdm(self.cid_list)  # 如果想進行測驗, 在這里切片操作只獲取前面兩個視頻即可
        all = []
        rstr = r"[\/\\\:\*\?\"\<\>\|]"  # 非法路徑名稱
        for i, data in enumerate(tn):
            dic = {}
            dic['name'] = self.name_list[i]  # Video name
            dic['cid'] = self.cid_list[i]  # Video cid
            dic['video'] = self._parse_cid(str(self.cid_list[i]))[0]  # video link
            dic['voice'] = self._parse_cid(str(self.cid_list[i]))[1]  # voice link
            dic['vq'] = self._parse_cid(str(self.cid_list[i]))[2]  # video quality
            all.append(dic)
            print("\n正在保存第{}個視頻\n".format(i + 1))
            if os.path.exists(os.path.join(self.save_file, re.sub(rstr, "_", all[i]['name'])) + '.mp4'):
                print(all[i]['name'] + "檔案已存在")
                continue
            video_res = requests.get(all[i]['video'], headers=self.headers, params=all[i]['vq'])
            # voice_res = requests.get(self.all[i]['voice'], headers=self.headers)
            # print(self.save_file + self.all[i]['name']+'.mp4')

            with open(os.path.join(self.save_file, re.sub(rstr, "_", all[i]['name'])) + '.mp4', 'wb+')as f:
                f.write(video_res.content)


spyder().Sync_run()