python爬虫爬美女图片：“小甲鱼056节轮一只爬虫的自我修养4：OOXX” 最终更新2021.6.27日，更新后可用！！！

本文链接：https://blog.csdn.net/Crystal_remember/article/details/109278548

2020.10.25日更新

需要改动有两个方面：

第一方面，网址编码有变化，如http://jandan.net/ooxx/MjAyMDEwMjUtODY=#comments，尾部的为base64编码，在网址构造时对应处使用base64编码.即可。

第二方面，图片地址发生了小改变，只需要将图片地址加入列表时，增加 http： 即可。

代码：

import urllib.request
import os
import base64


def url_open(url):
    opener = urllib.request.build_opener()
    # 向opener传入请求头信息
    opener.addheaders = ([
        ('User-Agent',
         'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36')
    ])
    # 将创建好的opener对象装入request
    urllib.request.install_opener(opener)
    req = urllib.request.Request(url)
    response = urllib.request.urlopen(url)
    html = response.read()
    return html


def get_page(url):
    html = url_open(url).decode('utf-8')
    a = html.find('current-comment-page') + 23
    b = html.find(']', a)
    return html[a:b]


def find_imgs(url):
    html = url_open(url).decode('utf-8')
    img_addrs = []
    a = html.find('img src=')
    while a != -1:
        b = html.find('.jpg', a, a + 255)
        if b != -1:
            temp = html[a + 9:b + 4]
            address = 'http:' + temp
            img_addrs.append(address )
        else:
            b = a + 9
        a = html.find('img src=', b)
    return img_addrs


def save_imgs(folder, img_addrs):
    for each in img_addrs:
        filename = each.split('/')[-1]
        with open(filename, 'wb') as f:
            img = url_open(each)
            f.write(img)


def download_mm(folder='ooxx', pages=10):
    os.mkdir(folder)
    os.chdir(folder)

    url = 'https://jandan.net/ooxx/'
    page_num = int(get_page(url))

    for i in range(pages):
        page_num -= i
        s_page = '20201025-' + str(page_num)
        base_page = base64.b64encode(s_page.encode('utf-8'))
        str_page = str(base_page, 'utf-8')
        page_url = url + str_page + '#comments'
        img_addrs = find_imgs(page_url)
        save_imgs(folder, img_addrs)


if __name__ == '__main__':
    download_mm()

值得注意的是，照骗的网址http://jandan.net/ooxx/MjAyMDEwMjUtODY=#comments后边的base64码解码后为 20201025-86，也就是今天的日期+‘-’+照骗页码。猜测网址会随着日期而变化，如果成立，使用时可以将代码段的图示位置按照图中提示改一下即可。也可以编写一个日期读取代码，动态调整该位置的代码。

测试图如下

2020.11.10日更新

上述假设成立，代码更新，自动读取当前日期，因为图片地址随着日期变化。将此处代码升级：

def get_time():
    now_time=datetime.datetime.now().strftime('%Y-%m-%d')
    list_time = list(now_time)
    list_time.pop(4)
    list_time.pop(6)
    time_now = ''.join(list_time)
    return time_now
time_now = get_time()

注意导入datetime包。

代码

import urllib.request
import os
import base64
import datetime

def url_open(url):
    opener = urllib.request.build_opener()
    # 向opener传入请求头信息
    opener.addheaders = ([
        ('User-Agent',
         'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36')
    ])
    # 将创建好的opener对象装入request
    urllib.request.install_opener(opener)
    req = urllib.request.Request(url)
    response = urllib.request.urlopen(url)
    html = response.read()
    return html


def get_page(url):
    html = url_open(url).decode('utf-8')
    a = html.find('current-comment-page') + 23
    b = html.find(']', a)
    return html[a:b]

def get_time():
    now_time=datetime.datetime.now().strftime('%Y-%m-%d')
    list_time = list(now_time)
    list_time.pop(4)
    list_time.pop(6)
    time_now = ''.join(list_time)
    return time_now

time_now = get_time()

def find_imgs(url):
    html = url_open(url).decode('utf-8')
    img_addrs = []
    a = html.find('img src=')
    while a != -1:
        b = html.find('.jpg', a, a + 255)
        if b != -1:
            temp = html[a + 9:b + 4]
            address = 'http:' + temp
            img_addrs.append(address )
        else:
            b = a + 9
        a = html.find('img src=', b)
    return img_addrs


def save_imgs(folder, img_addrs):
    for each in img_addrs:
        filename = each.split('/')[-1]
        with open(filename, 'wb') as f:
            img = url_open(each)
            f.write(img)


def download_mm(folder='ooxx', pages=10):
    os.mkdir(folder)
    os.chdir(folder)

    url = 'https://jandan.net/ooxx/'
    page_num = int(get_page(url))
    time_now = get_time()
    for i in range(pages):
        page_num -= i
        s_page = time_now + str(page_num)
        base_page = base64.b64encode(s_page.encode('utf-8'))
        str_page = str(base_page, 'utf-8')
        page_url = url + str_page + '#comments'
        img_addrs = find_imgs(page_url)
        save_imgs(folder, img_addrs)


if __name__ == '__main__':
    download_mm()

2021.6.27日更新，更新后可用

近日看了些爬虫，更新一下。

添加注释，修改bug，网站网址子目录有所更改xxoo=>girl，变为http://jandan.net/girl/，。当前时间该目录下共计1855张妹子图片。 enjoy yourself！

import urllib.request
import os
import base64
import datetime


def url_open(url):
    """
    打开网址，并返回网址内容
    :param url: 输入含妹子的网页
    :return: html：所输入的网页的内容/源码
    """
    opener = urllib.request.build_opener()
    # 向opener传入请求头信息
    opener.addheaders = ([
        ('User-Agent',
         'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36')
    ])
    # 将创建好的opener对象装入request
    urllib.request.install_opener(opener)
    req = urllib.request.Request(url)
    response = urllib.request.urlopen(url)
    html = response.read()
    return html


def get_page(url):
    """
    获取妹子网站最新页码数目，便于遍历
    :param url:
    :return: 返回最新更新的妹子的页面数字
    """
    html = url_open(url).decode('utf-8')
    a = html.find('current-comment-page') + 23
    b = html.find(']', a)
    #print(html[a:b])
    return html[a:b]


def get_time():
    """
    获取目标格式的当前时间
    :return: 返回时间，格式如 20210627
    """
    now_time = datetime.datetime.now().strftime('%Y-%m-%d')
    list_time = list(now_time)
    list_time.pop(4)
    list_time.pop(6)
    time_now = ''.join(list_time)
    return time_now


def find_imgs(url):
    """
    查找url内的图片地址
    :param url: 待查找url
    :return: 返回所查找url内的图片地址
    """
    html = url_open(url).decode('utf-8')
    img_addrs = []
    a = html.find('img src=')
    while a != -1:
        b = html.find('.jpg', a, a + 255)
        if b != -1:
            temp = html[a + 9:b + 4]
            address = 'http:' + temp
            img_addrs.append(address)
        else:
            b = a + 9
        a = html.find('img src=', b)
    return img_addrs


def save_imgs(folder, img_addrs):
    """
    下载保存图片
    :param folder: 保存图片的文件夹名字
    :param img_addrs: 图片地址
    :return: 无
    """
    print("此页面共计" + str(len(img_addrs)) +"张妹子图片...\n")
    i = len(img_addrs)
    for each in img_addrs:
        i = i + 1
        filename = each.split('/')[-1]
        with open(filename, 'wb') as f:
            img = url_open(each)
            f.write(img)
    print("此页面的" + str(len(img_addrs)) +"张妹子图片下载成功...\n")


def download_mm(folder='ooxx'):
    """
    主函数
    :param folder: 保存图片的文件夹名字
    :param pages:
    :return: 无
    """
    path = os.path.join(os.getcwd(), folder)
    if not os.path.exists(path):
        os.mkdir(path)  # 如果不存在这个pro文件夹，就自动创建一个
    else:
        print("文件夹已经存在...\n")
    os.chdir(folder)

    url = 'https://jandan.net/girl/'  ##2021.6.27妹子网址被更改
    print("正在查找含有妹子的网页...\n")
    page_num = int(get_page(url)) #含有妹子网页的最大页码数目
    print("查找成功，总共含有妹子的页面数目为: "+str(page_num))
    time_now = get_time()
    #print(time_now)
    num_all = 0 #统计已经下载的图片数目
    for i in range(1,page_num):
        print("=====================================================")
        print("正在对页面" + str(i) + "进行操作...\n" )
        s_page = time_now + '-' + str(i)
        base_page = base64.b64encode(s_page.encode('utf-8'))
        str_page = str(base_page, 'utf-8')
        page_url = url + str_page + '#comments'
        #print(page_url)
        #print(page_url)
        print("正在查找妹子图片地址...\n")
        img_addrs = find_imgs(page_url)
        print("正在下载页面"+ str(i) + "内的妹子图片...\n")
        num_all = num_all + len(img_addrs)
        save_imgs(folder, img_addrs)
        print("已经下载" + str(num_all) + "张图片\n")


if __name__ == '__main__':
    #修改下面待存储的文件夹名字，默认为 ooxx
    folder = 'ooxx'
    download_mm(folder)
    print("下载结束\n")

下载完成共计1855张图片，如下图