python 爬虫 获取bilibili search搜索数据，返回json格式

话不多说直接上代码

测试用例 :

搜索java为关键词的 时长在60分钟以上的视频

import requests
import json
import os

headers = {
    "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/88.0.4302.0 Safari/537.36"
}


def get_json(url):
    req = requests.get(url, headers=headers).text
    start_with_one = req.find('window.__INITIAL_STATE__=')
    end_with_one = req.find("""(function(){var s;(s=document.currentScript""")

    result_one = req[start_with_one: end_with_one]

    start_with_two = result_one.find('result')
    end_with_two = result_one.find(',"noMore"')

    result_two = result_one[start_with_two:end_with_two]

    start_with_three = result_two.find('[')
    result_json = result_two[start_with_three:]

    return result_json


def json_txt(dic_json):
    for i in range(0, len(dic_json)):
        for key in dic_json[i]:
            print("%s : %s" % (key, dic_json[i][key]))


if __name__ == "__main__":
    # 关键字 keyword
    keyword = "java"
    # 时长 duration  1: 10分钟以下 2: 10-30分钟 3: 30-60分钟 4: 60分钟以上
    duration = 4
    # 排序 order click： 最多点击    pubdate： 最新发布    dm： 最多弹幕    stow： 最多收藏
    order = "pubdate"
    # 分区 1=0  1 代表一级分类 0 代表全部
    tids_ = "1=0"
    page = 1
    url = "https://search.bilibili.com/all?keyword={}&order={}&duration={}&tids_{}&page={}".format(keyword, order,
                                                                                                   duration, tids_,
                                                                                                   page)
    # 获取json数据
    result = get_json(url)
    # 格式化成 str转为json
    json_value = json.loads(result)
    # 遍历打印json
    json_txt(json_value)
    os.system("pause");

输出文件：

import requests
import time


headers = {
    "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/88.0.4302.0 Safari/537.36"
}


def get_json(url):
    req = requests.get(url, headers=headers).text
    start_with_one = req.find('window.__INITIAL_STATE__=')
    end_with_one = req.find("""(function(){var s;(s=document.currentScript""")

    result_one = req[start_with_one: end_with_one]

    start_with_two = result_one.find('result')
    end_with_two = result_one.find(',"noMore"')

    result_two = result_one[start_with_two:end_with_two]

    start_with_three = result_two.find('[')
    result_json = result_two[start_with_three:]

    return result_json


def json_write(keyword,result):
    file_name = "bilibili-search-{}-{}.json".format(keyword, time.strftime("%Y-%m-%d", time.localtime()))
    with open(file_name, 'a+', encoding='utf-8') as  file:
        file.write(result)

if __name__ == "__main__":
    # 关键字 k+eyword
    keyword = input("请输入搜索关键词:  ")
    # 时长 duration  1: 10分钟以下 2: 10-30分钟 3: 30-60分钟 4: 60分钟以上
    duration = input("时长 1: 10分钟以下 2: 10-30分钟 3: 30-60分钟 4: 60分钟以上 : ")
    # 排序 order click： 最多点击    pubdate： 最新发布    dm： 最多弹幕    stow： 最多收藏
    order_list = ['click', 'pubdate', 'dm', 'stow']
    order_in = int(input("排序规则： 1 .最多点击 2.最新发布 3.最多弹幕 4.最多收藏 : "))
    if order_in<1 and order_in> 4:
        order_in =1
    order = order_list[order_in-1]
    # 分区 1=0  1 代表一级分类 0 代表全部
    tids_ = "1=0"
    page_totol = int(input("需要获取多少页 的数据 : "))+1

    for page in range(1,page_totol):
        url = "https://search.bilibili.com/all?keyword={}&order={}&duration={}&tids_{}&page={}".format(keyword, order,
                                                                                                       duration, tids_,
                                                                                                       page)
        result_json = get_json(url)
        if page ==(page_totol-1):
            result = ','+result_json[1:]
        elif page ==1:
            result =result_json[:-1]+','
        else:
            result =result_json[1:-1]
        print(result)
        json_write(keyword,result)