寒假大数据学习笔记四

　　今天的学习内容是利用python对图片进行爬取。

　　首先找到一个中意的图片网站，打开开发者工具，仔细寻找有关爬取内容的代码

　　

 

 　　可以很明显的找到.JPG格式的文件，然后直接爬取本网页的源代码，用正则表达式筛选出相应的.JPG文件，读取并保存就可以啦！

　　

from urllib import request
import os
import time
import re
from fake_useragent import UserAgent
import random


def url_open(url):
    # 使用代理IP的操作
    proxies = ['39.106.114.143:80', '47.99.236.251:3128', '58.222.32.77:8080',
               '101.4.136.34:81', '39.137.95.71:80', '39.80.41.0:8060']
    proxy_support = request.ProxyHandler(
        {'http': random.choice(proxies)})
    opener = request.build_opener(proxy_support)
    request.install_opener(opener)
    # 编写请求头
    header = {"User-Agent": UserAgent().random}
    if bool(re.search(r'https://www.xxx.com/\d{6}', url)):
        header = {
            "User-Agent": UserAgent().random,
            "Sec-Fetch-User": "?1",
            "Upgrade-Insecure-Requests": 1,
            "accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9",
            "cookie": "Hm_lvt_dbc355aef238b6c32b43eacbbf161c3c=1580652014; Hm_lpvt_dbc355aef238b6c32b43eacbbf161c3c=1580662027",
            "referer": 
        }
    req = request.Request(url, headers=header)
    response = request.urlopen(req, timeout=300)
    html = response.read()

    # print(url)
    return html

# 得到图片所在网页的网址
def get_page(url):
    html = url_open(url).decode("utf-8")
    pattern = re.compile(r'https://www.xxx.com/\d{6}')
    result = pattern.findall(html)
    # 去重并转换为集合
    result = set(result)
    # 集合不支持索引调用，转换为列表
    list_url = list(result)
    # 返回包含地址的列表
    return list_url

# 得到图片网址下的每一张图片的网址
def find_image(image_page_url):
    html = url_open(image_page_url).decode("utf-8")
    pattern = re.compile(image_page_url + r"(\d\d\d|\d\d|\d{0,1})")
    result = pattern.findall(html)
    # 去重并转换为集合
    result = set(result)
    # 集合不支持索引调用，转换为列表
    list_url = list(result)
    list_url = [image_page_url + x for x in list_url]
    # for i in list_url:
    #     print(i)
    return list_url

# 找到相应网址下的图片
def find_images_jpg(image_addr):
    images_addr = []
    images_URL = []
    for each in image_addr:
        html = url_open(each).decode("utf-8")
        pattern = re.compile(r'((https):[^\s]*?(jpge|jpg|png|PNG|JPG))')
        images_addr.append(pattern.findall(html))

    for i in images_addr:
        images_URL.append(i[0][0])

    # for j in images_URL:
    #     print(j)

    return images_URL

# 将图片保存到本地
def save_image(images):
    i = 1
    for each in images:
        with open(str(i) + ".jpg", "wb") as p:
            html = url_open(each)
            p.write(html)
        i += 1
        time.sleep(1)
        print(each)

# 主程序体
def downloadimg(folder="meiimages"):
    os.mkdir(folder)
    os.chdir(folder)

    url = "https://www.xxx.com/"
    # 得到相应的图片id
    image_page_url = get_page(url)
    # 循环访问相应路径
    for i in image_page_url:
        image_addr = find_image(i + "/")
        time.sleep(2)
        images = find_images_jpg(image_addr)
        save_image(images)
        time.sleep(2)

# 程序入口
if __name__ == '__main__':
    downloadimg()

　　本程序语法、逻辑无误，但本次爬取工作却是失败的：

　　

 

 　　403错误，程序触发了反爬取机制，导致程序运行失败，接下来的任务是学习编写请求头欺骗服务器。

 

来源：https://www.cnblogs.com/YXSZ/p/12254476.html