Home >Backend Development >Python Tutorial >Python learns requests to crawl web images

Python learns requests to crawl web images

coldplay.xixiforward: 2021-02-18 09:37:364067browse

免费学习推荐：python视频教程

前言

最近想做一个练练的小游戏给家里的小朋友玩儿，但是苦于选取素材，然后在一个巨佬的博客里找了灵感，就借用一下粉丝的头像试试爬取网页里的图片吧！（感谢各位啦！）

完成总目标：
爬取粉丝头像作为素材，完成一个连连看的小游戏
故本文分为两部分内容：
1、爬取素材部分；
2、利用素材完成连连看小游戏部分(链接)

（一）爬取粉丝头像

实现目标：

通过爬虫实现对粉丝头像的爬取并顺序排列存储，作为游戏设计的素材，其中爬取的头像一部分是使用了CSDN的默认头像，存在重复情况，所以还需要去重以得到完整且不重复的图像集

一、准备

1、python环境

2、涉及到的python库需要 pip install 包名 安装

二、代码编写

1.爬取内容

（1）所需要的库

import requestsimport json

（2）得到请求地址

url = 'https://blog.csdn.net//phoenix/web/v1/fans/list?page=1&pageSize=40&blogUsername=weixin_45386875'   #关注我的部分请求地址#url = 'https://blog.csdn.net//phoenix/web/v1/follow/list?page=1&pageSize=40&blogUsername=weixin_45386875'   #我关注的部分请求地址

请求地址获取方法：
右击所要爬取部分页面，点击审查元素，找到图中文件
Python learns requests to crawl web images

注：页面选到“TA的粉丝（13）”部分才能出现，如果点击Network什么也没有，则需要刷新页面就会出现页面内容了）
（3）带上请求头发送请求，做一些简单伪装

header = {
        'User-Agent' : 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36',
        'Cookie' : 'uuid_tt_dd=10_30826311340-1612520858912-361156; Hm_ct_6bcd52f51e9b3dce32bec4a3997715ac=6525*1*10_30826311340-1612520858912-361156!5744*1*weixin_45386875!1788*1*PC_VC; UN=weixin_45386875; p_uid=U010000; ssxmod_itna=Qui=DKiI3hkbG=DXDnD+r8h9eD53ecxPPit5bP1ODlOaYxA5D8D6DQeGTbcW1AoWGATqFYKmEWiH5/gbO4FpjQGcxLbbYx0aDbqGkqnU40rDzuQGuD0KGRDD4GEDCHHDIxGUBuDeKDR8qDg7gQcCM=DbSdDKqDHR+4FbG4oD8PdS0p=C7Gt3AuQ3DmtSije3r424rQ+iPqWzPDA3DK6jiD==; ssxmod_itna2=Qui=DKiI3hkbG=DXDnD+r8h9eD53ecxPPit5bP1D66Ii40vah303pFcXW0D6QALW==0tWLGtzKPUA76xoU10vpqD6AoOqs1R=Db=3olozYp0wVxUS0r/GeZCqzVezFQc8dZon7efYcr=1nhNr6nWKcTqqaDQYcwYSA+hNaqem=WWuDuDQ/+1PGEsN=atvS7WDp07vFuFDherg0AP0KFw0ea6kcTtK2rh/fy=/De0n1FNk+ONYxCXr=QrdTj6gxCuNNWXvp1IDdl2Ckjc=N/cqV6SmHZIZIuOEqml=dHMroMFDvdMVr8afnyQ+sbGPCbt3xdD07tAdRD7uDQ0gT=Bh7OYblbtYQFDDLxD2tGDD===; UserName=weixin_45386875; UserInfo=9863b829527c49a3ba1622396deaa7d9; UserToken=9863b829527c49a3ba1622396deaa7d9; UserNick=ryc875327878; AU=01F; BT=1612846374580; Hm_up_6bcd52f51e9b3dce32bec4a3997715ac=%7B%22uid_%22%3A%7B%22value%22%3A%22weixin_45386875%22%2C%22scope%22%3A1%7D%2C%22islogin%22%3A%7B%22value%22%3A%221%22%2C%22scope%22%3A1%7D%2C%22isonline%22%3A%7B%22value%22%3A%221%22%2C%22scope%22%3A1%7D%2C%22isvip%22%3A%7B%22value%22%3A%220%22%2C%22scope%22%3A1%7D%7D; __gads=ID=94978f740e79c9e5-22c918ed05c600ea:T=1613266189:RT=1613266189:S=ALNI_Mbwb8ad5kdYjogF7yImerVAzKaJuQ; dc_session_id=10_1613272889543.735028; announcement-new=%7B%22isLogin%22%3Atrue%2C%22announcementUrl%22%3A%22https%3A%2F%2Fblog.csdn.net%2Fblogdevteam%2Farticle%2Fdetails%2F112280974%3Futm_source%3Dgonggao_0107%22%2C%22announcementCount%22%3A0%2C%22announcementExpire%22%3A3600000%7D; dc_sid=3784575ebe1e9d08a29b0e3fc3621328; c_first_ref=default; c_first_page=https%3A//www.csdn.net/; c_segment=4; Hm_lvt_6bcd52f51e9b3dce32bec4a3997715ac=1613222907,1613266055,1613268241,1613273899; TY_SESSION_ID=82f0aa61-9b28-49b2-a854-b18414426735; c_pref=; c_ref=https%3A//www.csdn.net/; c_page_id=default; dc_tos=qoi2fq; log_Id_pv=925; Hm_lpvt_6bcd52f51e9b3dce32bec4a3997715ac=1613274327; log_Id_view=905; log_Id_click=658'
    }

获取请求头的方法：
Python learns requests to crawl web images

（4）向网页发送请求

try:
        data = requests.get(url,headers = header).text
        data_dist = json.loads(data)
    except:
        print('爬取失败')
        exit ()

2.保存所需图片并去重

定义一些函数备用

#保存文件def save_imag(file_name,img_url):
    request.urlretrieve(url=img_url,filename='D:\\ryc\python_learning\\10_linkup\\fan_avatar\\'+file_name)

#删除一个文件夹下的所有所有文件def del_file(path):
    ls = os.listdir(path)
    for i in ls:
        c_path = os.path.join(path, i)
        if os.path.isdir(c_path):#如果是文件夹那么递归调用一下
            del_file(c_path)
        else:                    #如果是一个文件那么直接删除
            os.remove(c_path)
    print ('文件已经清空完成')

#图像去重def compare_images(pic1,pic2):
    image1 = Image.open(pic1)
    image2 = Image.open(pic2)

    histogram1 = image1.histogram()
    histogram2 = image2.histogram()
    differ = math.sqrt(reduce(operator.add, list(map(lambda a,b: (a-b)**2,histogram1, histogram2)))/len(histogram1))
    print('differ:',differ)
    if differ == 0:
        return 'same'
    else:
        return 'diff'

#删除指定位置的图像def del_avatar(path):
    if os.path.exists(path):  # 如果文件存在
        os.remove(path)  
    else:
        print('no such file:%s'%(path))  # 则返回文件不存在

#先清空一下文件夹
    del_file('D:\\ryc\python_learning\\10_linkup\\fan_avatar')
    index = 0
    # i 是爬取列表的索引号； index 是保存的图片的索引号
    for i in range(0,len(fan_list)):
        fan_username = fan_list[i]['nickname']
        #print('fans_user%s:'%(i+1),fan_username)
        fan_avatar_url = fan_list[i]['userAvatar']
        #print('fans_avatar_url%s:'%(i+1),fan_avatar_url) 
        save_imag('fans_avatar%s.jpg'%(index+1),fan_avatar_url)
        #print('----------------save_image--fans_avatar%s.jpg'%(index+1))
        #图片去重
        for j in range(0,index):
            if index != j :
                comp_res = compare_images('./fan_avatar/fans_avatar%s.jpg'%(index+1),'./fan_avatar/fans_avatar%s.jpg'%(j+1))
                print('--------compare_images:--------'+'./fan_avatar/fans_avatar%s.jpg'%(index+1) + '------with---' + './fan_avatar/fans_avatar%s.jpg'%(j+1))
                print('comp_res:',comp_res)
                if comp_res == 'same':
                    del_avatar('D:\\ryc\python_learning\\10_linkup\\fan_avatar\\fans_avatar%s.jpg'%(index+1))
                    print('D:\\ryc\python_learning\\10_linkup\\fan_avatar\\fans_avatar%s.jpg'%(index+1))
                    index = index - 1
                    break
        
        index = index + 1

3、调用

if __name__ == "__main__":    spider_fanavatar()

三、完整代码

# 爬取网页图片import requestsfrom urllib import requestimport jsonfrom PIL import Imageimport osimport mathimport operatorfrom functools import reduce#保存文件def save_imag(file_name,img_url):
    request.urlretrieve(url=img_url,filename='D:\\ryc\python_learning\\10_linkup\\fan_avatar\\'+file_name)#爬取粉丝的头像def spider_fanavatar():
    url = 'https://blog.csdn.net//phoenix/web/v1/fans/list?page=1&pageSize=40&blogUsername=weixin_45386875'   #关注我的部分请求地址
    #url = 'https://blog.csdn.net//phoenix/web/v1/follow/list?page=1&pageSize=40&blogUsername=weixin_45386875'   #我关注的部分请求地址
    header = {
        'User-Agent' : 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36',
        'Cookie' : 'uuid_tt_dd=10_30826311340-1612520858912-361156; Hm_ct_6bcd52f51e9b3dce32bec4a3997715ac=6525*1*10_30826311340-1612520858912-361156!5744*1*weixin_45386875!1788*1*PC_VC; UN=weixin_45386875; p_uid=U010000; ssxmod_itna=Qui=DKiI3hkbG=DXDnD+r8h9eD53ecxPPit5bP1ODlOaYxA5D8D6DQeGTbcW1AoWGATqFYKmEWiH5/gbO4FpjQGcxLbbYx0aDbqGkqnU40rDzuQGuD0KGRDD4GEDCHHDIxGUBuDeKDR8qDg7gQcCM=DbSdDKqDHR+4FbG4oD8PdS0p=C7Gt3AuQ3DmtSije3r424rQ+iPqWzPDA3DK6jiD==; ssxmod_itna2=Qui=DKiI3hkbG=DXDnD+r8h9eD53ecxPPit5bP1D66Ii40vah303pFcXW0D6QALW==0tWLGtzKPUA76xoU10vpqD6AoOqs1R=Db=3olozYp0wVxUS0r/GeZCqzVezFQc8dZon7efYcr=1nhNr6nWKcTqqaDQYcwYSA+hNaqem=WWuDuDQ/+1PGEsN=atvS7WDp07vFuFDherg0AP0KFw0ea6kcTtK2rh/fy=/De0n1FNk+ONYxCXr=QrdTj6gxCuNNWXvp1IDdl2Ckjc=N/cqV6SmHZIZIuOEqml=dHMroMFDvdMVr8afnyQ+sbGPCbt3xdD07tAdRD7uDQ0gT=Bh7OYblbtYQFDDLxD2tGDD===; UserName=weixin_45386875; UserInfo=9863b829527c49a3ba1622396deaa7d9; UserToken=9863b829527c49a3ba1622396deaa7d9; UserNick=ryc875327878; AU=01F; BT=1612846374580; Hm_up_6bcd52f51e9b3dce32bec4a3997715ac=%7B%22uid_%22%3A%7B%22value%22%3A%22weixin_45386875%22%2C%22scope%22%3A1%7D%2C%22islogin%22%3A%7B%22value%22%3A%221%22%2C%22scope%22%3A1%7D%2C%22isonline%22%3A%7B%22value%22%3A%221%22%2C%22scope%22%3A1%7D%2C%22isvip%22%3A%7B%22value%22%3A%220%22%2C%22scope%22%3A1%7D%7D; __gads=ID=94978f740e79c9e5-22c918ed05c600ea:T=1613266189:RT=1613266189:S=ALNI_Mbwb8ad5kdYjogF7yImerVAzKaJuQ; dc_session_id=10_1613272889543.735028; announcement-new=%7B%22isLogin%22%3Atrue%2C%22announcementUrl%22%3A%22https%3A%2F%2Fblog.csdn.net%2Fblogdevteam%2Farticle%2Fdetails%2F112280974%3Futm_source%3Dgonggao_0107%22%2C%22announcementCount%22%3A0%2C%22announcementExpire%22%3A3600000%7D; dc_sid=3784575ebe1e9d08a29b0e3fc3621328; c_first_ref=default; c_first_page=https%3A//www.csdn.net/; c_segment=4; Hm_lvt_6bcd52f51e9b3dce32bec4a3997715ac=1613222907,1613266055,1613268241,1613273899; TY_SESSION_ID=82f0aa61-9b28-49b2-a854-b18414426735; c_pref=; c_ref=https%3A//www.csdn.net/; c_page_id=default; dc_tos=qoi2fq; log_Id_pv=925; Hm_lpvt_6bcd52f51e9b3dce32bec4a3997715ac=1613274327; log_Id_view=905; log_Id_click=658'
    }
    try:
        data = requests.get(url,headers = header).text   #得到返回的字符串
        data_dist = json.loads(data)                     #将字符串转为字典格式
    except:
        print('爬取失败')
        exit ()
    fan_list = data_dist['data']['list']                #提取所需内容
    #先清空一下文件夹
    del_file('D:\\ryc\python_learning\\10_linkup\\fan_avatar')
    index = 0
    # i 是爬取列表的索引号； index 是保存的图片的索引号
    for i in range(0,len(fan_list)):
        fan_username = fan_list[i]['nickname']
        #print('fans_user%s:'%(i+1),fan_username)
        fan_avatar_url = fan_list[i]['userAvatar']
        #print('fans_avatar_url%s:'%(i+1),fan_avatar_url) 
        save_imag('fans_avatar%s.jpg'%(index+1),fan_avatar_url)
        #print('----------------save_image--fans_avatar%s.jpg'%(index+1))
        #图片去重
        for j in range(0,index):
            if index != j :
                comp_res = compare_images('./fan_avatar/fans_avatar%s.jpg'%(index+1),'./fan_avatar/fans_avatar%s.jpg'%(j+1))
                print('--------compare_images:--------'+'./fan_avatar/fans_avatar%s.jpg'%(index+1) + '------with---' + './fan_avatar/fans_avatar%s.jpg'%(j+1))
                print('comp_res:',comp_res)
                if comp_res == 'same':
                    del_avatar('D:\\ryc\python_learning\\10_linkup\\fan_avatar\\fans_avatar%s.jpg'%(index+1))
                    print('D:\\ryc\python_learning\\10_linkup\\fan_avatar\\fans_avatar%s.jpg'%(index+1))
                    index = index - 1
                    break
        
        index = index + 1
                #图像去重def compare_images(pic1,pic2):
    image1 = Image.open(pic1)
    image2 = Image.open(pic2)
    histogram1 = image1.histogram()
    histogram2 = image2.histogram()
    differ = math.sqrt(reduce(operator.add, list(map(lambda a,b: (a-b)**2,histogram1, histogram2)))/len(histogram1))
    print('differ:',differ)
    if differ == 0:
        return 'same'
    else:
        return 'diff'
        #删除指定位置的图像def del_avatar(path):
    if os.path.exists(path):  # 如果文件存在
        os.remove(path)  
    else:
        print('no such file:%s'%(path))  # 则返回文件不存在#删除一个文件夹下的所有所有文件def del_file(path):
    ls = os.listdir(path)
    for i in ls:
        c_path = os.path.join(path, i)
        if os.path.isdir(c_path):#如果是文件夹那么递归调用一下
            del_file(c_path)
        else:                    #如果是一个文件那么直接删除
            os.remove(c_path)
    print ('文件已经清空完成')if __name__ == "__main__":
    spider_fanavatar()

最后

第一部分内容就到这里，第二部分内容将在下一篇文章完成，感兴趣的小伙伴可以关注我，然后去看下一片文章哦！

都读到这里了，各位亲爱的读者留下你们宝贵的赞和评论吧，这将是我继续前进的坚定动力！！！

大量免费学习推荐，敬请访问python教程(视频)

The above is the detailed content of Python learns requests to crawl web images. For more information, please follow other related articles on the PHP Chinese website!

Python pip

Statement：

This article is reproduced at:csdn.net. If there is any infringement, please contact admin@php.cn delete

Previous article：Introducing three methods of python data captureNext article：Introducing three methods of python data capture

See more

Python learns requests to crawl web images

前言

Related articles