AI网络爬虫:用deepseek提取百度文心一言的智能体数据

2024-06-21 07:16:05 浏览数 (1)

真实网址:https://agents.baidu.com/lingjing/experhub/search/list?pageSize=36&pageNo=1&tagId=-99

返回的json数据:{

"errno": 0,

"msg": "success",

"data": {

"total": 36,

"pageNo": 1,

"pageSize": 36,

"plugins": [

{

"name": "零基础学习路径规划",

"description": "你好,请你给出一个主题,我将给你一份完整的学习路径规划",

"logoUrl": "https://now.bdstatic.com/stash/v1/6f672d5/lingjing-fe/07ccbd4/agent-logo/logo-1.png",

"logoText": {

"bgImage": "",

"labelValue": "1",

"color": 0,

"labelType": 2

},

"previewUrl": "https://t6acl8.smartapps.baidu.com",

"appId": "UjRqa5SBZzwvbRH8I06qp19H6OZdzjCR",

"pluginSource": 3,

"useNum": "786492",

"buildType": 1,

"isOwner": null,

"packageStatus": 6,

"dialogueCount": "2030481",

"developerName": "好想出去玩2020",

"isFavorited": false

},

deepseek中输入提示词:

你是一个Python编程专家,完成一个Python脚本编写的任务,具体步骤如下:

在F盘新建一个Excel文件:baiduaiagent20240619.xlsx

请求网址:

https://agents.baidu.com/lingjing/experhub/search/list?pageSize=36&pageNo=1&tagId=-99请求方法:

GET

状态代码:

200 OK

获取网页的响应,这是一个嵌套的json数据;

获取json数据中"data"键的值,然后获取其中"plugins"键的值,这是一个json数据,提取这个json数据中所有的键写入Excel文件的表头 ,提取这个json数据中所有键对应的值写入Excel文件的列 ;

保存Excel文件;

注意:每一步都输出信息到屏幕;

每爬取1页数据后暂停5-9秒;

需要对 JSON 数据进行预处理,将嵌套的字典和列表转换成适合写入 Excel 的格式,比如将嵌套的字典转换为字符串;

在较新的Pandas版本中,append方法已被弃用。我们应该使用pd.concat来代替。

源代码:

import requests

import pandas as pd

import time

import json

# 请求URL

url = "https://agents.baidu.com/lingjing/experhub/search/list?pageSize=36&pageNo=1&tagId=-99"

# 请求头

headers = {

"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/125.0.0.0 Safari/537.36"

}

# 创建Excel文件

file_path = "F:/baiduaiagent20240619.xlsx"

df = pd.DataFrame()

# 发送GET请求

response = requests.get(url, headers=headers)

if response.status_code == 200:

data = response.json()

products = data['data']['plugins']

# 提取所有产品的键作为表头

headers = set()

for product in products:

headers.update(product.keys())

# 创建DataFrame并填充数据

for product in products:

product_data = {header: product.get(header, '') for header in headers}

new_data = pd.DataFrame([product_data])

df = pd.concat([df, new_data], ignore_index=True)

print("Data processed.")

else:

print(f"Failed to retrieve data. Status code: {response.status_code}")

# 保存Excel文件

df.to_excel(file_path, index=False)

print(f"Data saved to {file_path}")

# 暂停5-9秒

time.sleep(5 (1 % 5))

0 人点赞