前言

今天我们通过Python爬取小姐姐图片网站上的美图，零基础学会通用爬虫，当然我们还可以实现多线程爬虫，加快爬虫速度

环境介绍

python 3.6
pycharm
requests >>> pip install requests
re
time
concurrent.futures

爬虫最基本思路

爬取单个相册内容:

找到目标 https://https://www.kanxiaojiejie.com/img/6509
发送请求 (人为操作: 访问网站)
获取数据 (HTML代码就是服务器返回的数据)
数据提取 (筛选里面的内容)
HTML网页代码
保存数据 (把图片下载下来)

目标网站

简单的通用爬虫代码

import requests
import parsel
import re
import os

page_html = requests.get('https://www.kanxiaojiejie.com/page/1').text
pages = parsel.Selector(page_html).css('.last::attr(href)').get().split('/')[-1]
for page in range(1, int(pages) + 1):
    print(f'==================正在爬取第{page}页==================')
    response = requests.get(f'https://www.kanxiaojiejie.com/page/{page}')
    data_html = response.text
    # 提取详情页
    zip_data = re.findall('<a href="(.*?)" target="_blank"rel="bookmark">(.*?)</a>', data_html)
    for url, title in zip_data:
        print(f'----------------正在爬取{title}----------------')
        if not os.path.exists('img/' + title):
            os.mkdir('img/' + title)
        resp = requests.get(url)
        url_data = resp.text
        selector = parsel.Selector(url_data)
        img_list = selector.css('p>img::attr(src)').getall()

        for img in img_list:
            img_data = requests.get(img).content
            img_name = img.split('/')[-1]
            with open(f"img/{title}/{img_name}", mode='wb') as f:
                f.write(img_data)
            print(img_name, '爬取成功！！！')
        print(title,'爬取成功！！！')

升级多线程版本

把每一块都封装一个函数, 每个函数都有它特定的功能

先导入模块

import requests # 第三方模块 pip install requests
import re # 正则表达式模块 内置模块
import time
import concurrent.futures
import os
import parsel

发送请求

def get_response(html_url):
    headers = {
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/90.0.4430.212 Safari/537.36'
    }
    # 为什么这里要 requests.get()  post() 请求会更安全...
    response = requests.get(url=html_url, headers=headers)
    return response

保存数据

def save(title, img_url):
    img_data = requests.get(img_url).content
    img_name = img_url.split('/')[-1]
    with open("img\\" + title + '\\' + img_name, mode='wb') as f:
        f.write(img_data)

解析数据获取图片url地址以及标题

def parse_1(data_html):
    zip_data = re.findall('<a href="(.*?)" target="_blank"rel="bookmark">(.*?)</a>', data_html, re.S)
    return zip_data

解析数据获取图片url地址以及标题

def parse_2(html_data):
    selector = parsel.Selector(html_data)
    img_list = selector.css('p>img::attr(src)').getall()
    return img_list

创建文件夹

def mkdir_img(title):
    if not os.path.exists('img\\' + title):
        os.mkdir('img\\' + title)

主函数

def main(html_url):
    html_data = requests.get(html_url).text
    zip_data = parse_1(html_data)
    for url, title in zip_data:
        mkdir_img(title)
        html_data_2 = get_response(url).text
        img_list = parse_2(html_data_2)
        for img in img_list:
            save(title, img)
        print(title, '爬取成功！！！')

程序的入口

if __name__ == '__main__':
    time_1 = time.time()
    exe = concurrent.futures.ThreadPoolExecutor(max_workers=10)
    for page in range(1, 11):
        url = f'https://www.kanxiaojiejie.com/page/{page}'
        exe.submit(main, url)
    exe.shutdown()
    time_2 = time.time()
    use_time = int(time_2) - int(time_1)
    print(f'总计耗时:{use_time}秒')