一只爬虫

1. 流程分析

1.1 分析目标地址分页的情况

• 第一页：https://域名/分类/index.html
• 第二页：https://域名/分类/index_2.html
• 第三页：https://域名/分类/index_3.html

即可得出目标分页的规律

1.2 分析某一页各个项的特征

观察得出每一项的url地址在单页各个项的ul下面的li中的a标签中保存，进入这个url地址即可得到此项中所有图片的地址；

此外这个a标签的alt属性中还保存了此项的摘要描述。

1.3 进入某一目标项获取图片

通过以上步骤，我们要想获取图片的url的路径，需要访问每一页的每一项的URL，这样即可加载获取所有图片的URL地址；

拿到图片地址后保存，然后即可开启多线程下载图片。

2. 代码实现

2.1 代码

import os, re
import requests
import logging
from bs4 import BeautifulSoup
from urllib.request import urlretrieve
from concurrent.futures import ThreadPoolExecutor

logging.captureWarnings(True)     # 忽略所有警告信息

class fuck_photo():
    xxoo = """
             1:  Asian   
             2:  Cartoon 
             3:  Cute    
             4:  Secret 
             5:  Silk   
         """

    def __init__(self, server, base_url, dirname, thread_num, headers, ):
        self.server = server
        self.base_url = base_url
        self.dirname = dirname
        self.headers = headers
        self.set_path(dirname)
        self.thread_poll = ThreadPoolExecutor(thread_num)
        self.photo_type = {
            1: "yazhousetu/",
            2: "katong/",
            3: "mengmeizi/",
            4: "zipai/",
            5: "qingchun/",
        }

    def get_page_list(self, target, start_page, max_page):
        """获取所有页的url"""
        if start_page <= 1:
            page_list = [target, ]
            print("从第一页开始爬...")
            start_page += 1
        else:
            page_list = []
        for i in range(start_page, max_page + 1):
            page_link = target[:-5] + "_" + str(i) + ".html"
            page_list.append(page_link)
        print('
所有图片会保存在【%s】文件夹中...
' % self.dirname)
        return page_list

    def get_list(self, target):
        """获取每页所有项目的url"""
        per_page_link_list = []
        response = requests.get(url=target, headers=self.headers, verify=False)
        content = BeautifulSoup(response.text, features="html.parser")
        the_list = content.find('ul', class_='update_area_lists cl').find_all('a')
        for i in the_list:
            per_page_link_list.append(self.server[:-1] + i.get('href'))
        return per_page_link_list

    def get_photo_link_list(self, link_target):
        """获取某页中某个项中所有图片的URL"""
        photo_link_dick = {}
        response = requests.get(url=link_target, headers=self.headers, verify=False)
        content = BeautifulSoup(response.text, features="html.parser")
        the_list = content.find('div', class_='content_left').find_all('img')
        for i in the_list:
            photo_link_dick[i.get('src')] = i.get('alt')
        return photo_link_dick

    def set_path(self, dirname):
        self.dirpath = os.getcwd() + os.sep + dirname
        if not os.path.isdir(self.dirpath):
            os.mkdir(dirname)

    def download_photo(self, photo_links, ret):
        """下载图片"""
        try:
            # filename = re.findall('/(w*?.jpg)', photo_links)[0]
            filename = re.findall('/(w*?.jpg)', photo_links)[0][-12:]
            path = self.dirpath + os.sep + ret[photo_links] + filename
            urlretrieve(photo_links, path)
        except Exception as e:
            print("sorry，图片下载时出现了一个错误: %s 
【已忽略...】" % e)
            pass

    def download_thread(self, i):
        ret = photo.get_photo_link_list(i)
        for item in ret:
            self.thread_poll.submit(photo.download_photo, item, ret)

    def start_fuck(self):
        type_num = int(input(self.xxoo + '

' + "请输入图片类型："))
        while not type_num:
            print("请输入对应的数字！" + '

' + self.xxoo)
            type_num = int(input("请输入图片类型："))
        target = self.base_url + self.photo_type.get(type_num) + 'index.html'
        startpage = int(input("从哪一页开始:"))
        maxpage = int(input("到哪一页为止:"))

        # 获取指定范围内所有项的URL
        all_page_list = self.get_page_list(target, startpage, maxpage)
        for item in all_page_list:
            try:
                content_list = photo.get_list(item)
                for i in content_list:
                    photo.download_thread(i)
            except Exception as e:
                print("出现了一个错误: %s 
【已忽略】" % e)
                pass


if __name__ == '__main__':
    photo = fuck_photo(
        "https://htx5.com/",      # 目标主机
        "https://htx5.com/",      # 目标地址
        "精美壁纸",                 # 设置保存图片的目录，更改爬取类型时，要修改这里的目录名称
        64,                         # 线程数
        {                           # 配置请求头
            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.81 Safari/537.36 SE 2.X MetaSr 1.0',
            'Referer': "https://htx5.com/",
        }
    )
    photo.start_fuck()

2.2 运行效果