python小咖爬虫基础

迷南。 2024-04-18 23:44 7阅读 0赞

### python小咖 爬虫基础 ###

安装爬虫相关库,一般是进入python所在目录的scripts目录下,在文件里按住shift点右键,选择![20190915065949247.png][].然后执行pip install 库名

**html元素  
标签  
选择器  
属性**

**css**

**script**

## **安装requests** ##

PS D:\\Program Files\\Python37\\Scripts> **pip list**  
Package    Version  
\---------- -------  
numpy      1.17.2  
pip        19.2.3  
setuptools 41.2.0  
wheel      0.33.6  
PS D:\\Program Files\\Python37\\Scripts> **pip install requests**  
Collecting requests  
  Downloading

………

Installing collected packages: **urllib3, chardet, idna, certifi, requests**  
**Successfully installed certifi-2019.9.11 chardet-3.0.4 idna-2.8 requests-2.22.0 urllib3-1.25.3**  
PS D:\\Program Files\\Python37\\Scripts> **pip lis**t  
Package    Version  
\---------- ---------  
certifi    2019.9.11  
chardet    3.0.4  
idna       2.8  
numpy      1.17.2  
pip        19.2.3  
requests   2.22.0  
setuptools 41.2.0  
urllib3    1.25.3  
wheel      0.33.6  
PS D:\\Program Files\\Python37\\Scripts>  
显示自动增加了5个文件**urllib3, chardet, idna, certifi, requests**。

urlib3没有requests方便?

### 爬虫利器requests介绍 ###

http://docs.python-requests.org/en/master/  
一个简单的例子:  
**import requests   
req = requests. get('http://docs.python-requests.org/en/master/')  
print(type(req))  
print(req.status\_code)  
print(req.encoding)  
print(req.cookies)**

**status code  状态码  
encoding      编码方式  
cookies        Cookies**  
状态码           含义  
200               请求成功  
301               资源(网页等)被永久转移到其它URL  
404               请求的资源(网页等)不存在  
500              内部服务器错误

### **Cookies的用途** ###

**1.会话状态管理**(如用户登录状态、购物车、游戏分数或其它需要记录的信息)  
**2.个性化设置**(如用户自定义设置、主题等)  
**3.浏览器行为跟踪**(如跟踪分析用户行为等)

### 基本请求 ###

**requests库提供http的所有基本请求方式**

**import requests **  
req= requests.**get**('http://docs.python-requests.org/en/master/')     \#200  
req= requests.**post**('http://docs.python-requests.org/en/master/')   \#200  
req= requests.**put**('http://docs.python-requests.org/en/master/')     \#405  
req= requests.**delete**('http://docs.python-requests.org/en/master/')  \#405  
req= requests.**head**('http://docs.python-requests.org/en/master/')   \#301  
req= requests.**options**(' http://docs.python-requests.org/en/master/')  \#405

**print(req.elapsed.total\_seconds())  \#本句可以记录发出请求到返回响应时长**

### GET请求 ###

**可利用params参数**  
import requests  
payload=\{'key1':'value1','key2':'value2'\}    \#定义一个字典,作为参数  
r=requests.**get**("http://docs.python-requests.org/en/master/", **params**=payload)    \#用params传递参数,  
print(r.url)  
输出结果  
\#老师结果http:/ /docs . python- requests . org/ en/master/ ?key2=value2&key1=value1

实际结果[https://2.python-requests.org//en/master/?key1=value1&key2=value2][https_2.python-requests.org_en_master_key1_value1_key2_value2]

(可参看官方说明举例:[https://2.python-requests.org//en/master/user/quickstart/\#make-a-request][https_2.python-requests.org_en_master_user_quickstart_make-a-request])

***用get的方式,如果传输密码账号则太明显,不安全***

### **POST请求** ###

利用**data**参数为**POST**添加参数  
data\_form = \{'key1': ' value1', 'key2' : ' value2 ' \}  
req = requests. **post**("http://httpbin.org/post", data=data\_form)  
print(req.text)

**上传文件**  
第一种方法

files = \{'file': open( 'test. txt' ,'rb')\}  
req = requests. **post**("http://httpbin. org/post", file=files)

**第二种方法(较优,安全一点)**  
with open( 'test.txt' ) as f:  
    requests.**post**("http://httpbin.org/post",data=f)

###    获取/发送cookies ###

**获取www.baidu.com的cookies                                   **

import requests
    url ='http://httpbin.org/ cookies/'
    url = 'https://www.baidu.com/'
    req = requests.get(url)
    coo = req.cookies
    print(coo)
    for key in coo.keys():
    	print(key)
    	print(coo[key])

**发送cookies**

import requests
    url = 'http://httpbin.org/cookies'
    cookies = dict(my_cookie='test')        #自己建一个cookies字典
    req=requests.get(url,cookies=cookies)
    print(req.text)
    
    {
      "cookies": {
        "my_cookie": "test"
      }
    }

### 超时配置 ###

利用**timeout**变量来配置最大请求时间  
import requests  
requests.get( 'http:/ /www. baidu. com' , **timeout**=0. 001)    \#只是连接过程的时长,至于连接之后下载什么的所用时长不在此计算.

### **会话对象** ###

**Session() 会话**

import requests  
s = requests.Session()   \#表示后面的登录及操作是同一会话中的,不是另外的身份进行的.  
s. get('http://www.baidu.com/S?wd=pycharm')  
print(s.cookies)  
s.get('http://www.baidu.com/s?wd=python')  
print(s.cookies)    \#上面两行,s用了Session(),print出的cookies代码ID数值一样

req = requests.get('http://www.baidu.com/s?wd=pycharm')  
print(req.cookies)  
req = requests.get('http://www.baidu.com/s?wd=python')  
print(req.cookies,end="cookies")  \#后面的这两行的req是没有用Session的,结果ID不一样.

<RequestsCookieJar\[<Cookie BAIDUID=**22854AAD8AE2CFE857F18EF2AFE287F7**:FG=1 for .baidu.com/>, <Cookie  
<RequestsCookieJar\[<Cookie BAIDUID=**22854AAD8AE2CFE857F18EF2AFE287F7**:FG=1 for .baidu.com/>, <Cookie  
<RequestsCookieJar\[<Cookie BAIDUID=**9ED208CCDEAC45DC61AD9F6F82A589DF**:FG=1 for .baidu.com/>, <Cookie  
<RequestsCookieJar\[<Cookie BAIDUID=**9ED208CCDEAC45DC676ED68A05DA6496**:FG=1 for .baidu.com/>, <Cookie

### **设置代理** ###

**利用proxies**参数使用代理  
import requests  
proxies = \{  
"https": "http:/ /101.6.162.19:22"  
)\}  
r = requests. post( "http:/ /httpbin. org/post", **proxies**=proxies)  
print(r. text )  
**或者设置全局变量来使用代理**

windows下通过网络设置

**右击**win10右下网络符号,"**打开"网络和Internet"设置**

在"设置"页面,往**下拉**,"代理",-->再下拉,"手动设置代理"

**手动设置代理**  
将代理服务器用于以太网或Wi Fi连接。这些设置不适用于VPN连  
接。

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70][]  
Mac或者Linux下在终端输入  
export HTTP PROXY= ""http://101.6.162.19""  
export HTTPS PROXY= "http://101.6.162.19"

\-------------------------------------------------------------------

## Beautiful Soup介绍 ##

[https://www.crummy.com/software/BeautifulSoup/bs4/doc/][https_www.crummy.com_software_BeautifulSoup_bs4_doc]

爬虫利器，出色的解析工具

## 安装beautifulsoup4 ##

两个安装命令:[https://www.crummy.com/software/BeautifulSoup/bs4/doc/\#installing-beautiful-soup][https_www.crummy.com_software_BeautifulSoup_bs4_doc_installing-beautiful-soup]

安装方式，在命令行中输入  
pip install lxml  
pip install beautifulsoup4

Mac用户输入  
pip3 install lxml

pip3 install beautifulsoup4

### **使用beautifulsoup** ###

注意如何导入时的模块名称  
**from bs4 import BeautifulSoup**   
(我们需要的是bs4里的BeautifulSoup模块)

import requests
    from bs4 import BeautifulSoup
    data_form = { 'key1' : 'value1', 'key2' : 'value2' }
    r = requests. post('http://httpbin.org/post', data = data_form )
    print(r.text)

一个例子  
**prettify()格式化输出**

**把输出的内容归整,使用缩进等.便于阅读.**

import requests
    from bs4 import BeautifulSoup
    r = requests.get("https://www.crummy.com/software/BeautifulSoup/bs4/doc/")
    html = r.text
    # print (html )
    soup=BeautifulSoup(html, "lxml") #使用lxml工具来分析
    print (soup.prettify())

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 1][]

## 四大对象种类1一 Tag ##

Tag就是HTML中的一个个标签  
注意:返回的是第一个符合要求的标签 (即使HTML中有多个符合要求的标签)

print(soup.head)

print(soup.title)

print(soup.a)

import requests
 from bs4 import BeautifulSoup
 r = requests.get("https://www.Crummy.com/software/BeautifulSoup/bs4/doc/")
 html = r.text
 # print (html )
 soup = BeautifulSoup(html, "lxml")
 # print( soup.prettify())
 print(' ### title ###')
 print(soup.title)
 print('### head ###')
 # print ( soup. head )
 # print('### a ###')
 # print( soup.a)
 
 ### title ###
 <title>Beautiful Soup Documentation — Beautiful Soup 4.4.0 documentation</title>
 ### head ###
 [Finished in 3.4s]

上面例子中改换两行:
 r = requests.get("https://www.Crummy.com/software/BeautifulSoup/bs4/doc.zh/")	#中文文档
 print(soup.title)
 print(soup.h2)
 
 
 ### title ###
 <title>Beautiful Soup 4.2.0 文档 — Beautiful Soup 4.2.0 documentation</title>
 <h2>寻求帮助<a class="headerlink" href="#id1" title="Permalink to this headline">¶</a></h2>
 ### head ###
 [Finished in 3.5s]

**attrs**  获取标签的元素属性  
**get()方法  **获取标签的某个属性值  
可以通过修改字典的方式对这些属性和内容等等进行修改,删除等操作  
print( soup.a.attrs)  
print(soup.a\['title'\])  \#f元素的属性相当于一个字典,通过key来获取值.  
print( soup.a.get( 'title'))  \#效果与上句等同  
soup.a\['title'\] = "a new title'   \#修改  
![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 2][]

### 四大对象种类2一NavigableString ###

**获取某个标签里面的内容**  
print( ' a.string: \{\}' .format( soup.a. string) )  
print('a. string type: \{\}'. format (type( soup.a. string)**))**  
![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 3][]

import requests  
from bs4 import BeautifulSoup  
\# r = requests.get("https://www.Crummy.com/software/BeautifulSoup/bs4/doc/")  
r = requests.get("http://www.mcyz.net.cn//")    \#蒙城一中网址  
html = r.text  
soup = BeautifulSoup(html, "lxml")  
\# \# print( soup.prettify())  
\# print(soup.div)  
print(soup.head.meta)  
print(soup.title)  
print(soup.title.**string**)

### 四大对象种类3一Comment ###

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 4][]

\# HTML中可以用来添加一段暂不通过网页渲染出来的内容

要获取网页中的注释部分comment,可以用.string

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 5][]

观察上图中html代码

![20190917102633411.png][]

观察type(soup.a.string))和type(soup.p.string))输出的内容差异:都是用.string获取,但类型不同.

.a.string的是**注释内容**,类型为bs4.element.**Comment**

.p.string的是**标签中的正文**Test,类型为bs4.element**.NavigableString**

![20190917103304962.png][]

### 四大对象种类4一BeautifulSoup ###

BeautifulSoup对象表示的是一个文档的全部内容

import requests  
from bs4 import BeautifulSoup  
r = requests.get("https://www.Crummy.com/software/BeautifulSoup/bs4/doc/")  
html = r.text

soup = BeautifulSoup(html,'lxml')   
  
print(type(soup))  
print( soup.name)  
print(soup.attrs)

<class 'bs4.BeautifulSoup'> 
\[document\] 
\{\}

### 文档树一直接子节点 ###

from bs4 import BeautifulSoup  
import requests

r = requests.get("https://www.crummy.com/software/BeautifulSoup/bs4/doc/")  
html = r.text  
soup=BeautifulSoup(html,'lxml')  
\#直接子节点  
print (soup.head.contents)  
print (soup.head.children)   
for child in soup.head.children:  
    print (child)

### 文档树-所有子孙节点 ###

**.descendants**把某个标签内的所有子孙节点都列出来  
可以通过**for循环**来进行处理

for child in soup.descendants:  \#所有节点  
    print(child)

for child in soup.head.descendants:  \#标签<head></head>m内的子孙节点 
    print(child)

### 文档树-节点内容 ###

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 6][]

注意:当与<a>之间有**换行或空行**时,

空行或换行都作为一个节点.

soup.**a**.string与soup.**p**.string的输出是**不一样**的,空行换行返回None

![20190918135729990.png][]

![2019091813580056.png][]

注意:  
**如果tag包含有多个(能够调用.string的)节点, .string方法会返回:None**

print(soup.head.string)  
print(soup.title.string)  
print(soup.html.string)

**直接用本地temp.html文件作测试,不去网站,:** 
<!DOCTYPE html> 
<html lang="en"> 
<head> 
    <meta charset="UTF-8"> 
    <title>Title</title> 
    <a>Test</a> 
</head> 
<body>

</body> 
</html>

**方法一是,内容作字符串  
方法二是,读取这个文件  f = open(r'文件路径', 'r').read()**  
(感谢网友q\*\*\*7553"满怀心"帮助)

<table> 
 <tbody> 
 <tr> 
 <td colspan="3"> from bs4 import BeautifulSoup import requests r = requests.get("https://www.crummy.com/software/BeautifulSoup/bs4/doc/") html=r.text soup=BeautifulSoup(html,'lxml') print(soup.head.string) print(soup.title.string) print(soup.html.string) </td> 
 </tr> 
 <tr> 
 <td colspan="1" style="width:296px;"> &lt;!DOCTYPE html&gt; &lt;html lang="en"&gt; &lt;head&gt; &nbsp;&nbsp; &nbsp;&lt;meta charset="UTF-8"&gt; &nbsp;&nbsp; &nbsp;&lt;title&gt;Title&lt;/title&gt; &nbsp;&nbsp; &nbsp;&lt;a&gt;&lt;p&gt;Test&lt;/p&gt;&lt;/a&gt; &lt;/head&gt; &lt;body&gt; &lt;/body&gt; &lt;/html&gt; </td> 
 <td style="vertical-align:top;width:242px;"> html='''&lt;!DOCTYPE html&gt; &lt;html lang="en"&gt; &lt;head&gt; &nbsp;&nbsp; &nbsp;&lt;meta charset="UTF-8"&gt; &nbsp;&nbsp; &nbsp;&lt;title&gt;Title&lt;/title&gt; &nbsp;&nbsp; &nbsp;&lt;a&gt;&lt;p&gt;Test&lt;/p&gt;&lt;/a&gt; &lt;/head&gt; &lt;body&gt; &lt;/body&gt; &lt;/html&gt;''' </td> 
 <td style="vertical-align:top;width:411px;"> 设为字符串,分别用三个单引号作首尾,多行引用. rom bs4 import BeautifulSoup import requests # r = requests.get("https://www./")&nbsp; #不用网络请求库 # html=r.text html='''&nbsp; &lt;!DOCTYPE.............'''&nbsp; 内容作字符串,参看左侧. soup=BeautifulSoup(html,'lxml') print(soup.head.string) print(soup.title.string) print(soup.html.string) </td> 
 </tr> 
 <tr> 
 <td colspan="2" style="width:552px;"> rom bs4 import BeautifulSoup import requests # r = requests.get("https://www...../") # html=r.text html=open(r'temp.html', 'r').read()&nbsp; &nbsp;#加此句读入文件 soup=BeautifulSoup(html,'lxml') print(soup.head.string) print(soup.title.string) print(soup.html.string) &nbsp; </td> 
 </tr> 
 </tbody> 
</table>

# #

### 文档树-节点内容 ###

如果想要**获得一个tag 下面的多个内容**，  
我们该如何操作?  
**.strings**或者**.stripped\_strings**

for string in soup.**o**l. strings:  
print(string)  
for string in soup. **ol.** stripped\_ strings :  
print( string )  
**.strings与.stripped strings 区别?  
.stripped\_ strings可以去除多余空白内容**

html=''' <!DOCTYPE html> 
<html lang="en"> 
<head> 
    <meta charset="UTF-8"> 
    <title>Title</title> 
    <a> 
    Test 
    </a> 
</head> 
<body>

</body> 
</html>

'''

from bs4 import BeautifulSoup  
import requests

soup=BeautifulSoup(html,'lxml')

print(soup.a.string)  \#因后有换行,所以输出为None  
print(soup.p.string)  
\# for child in soup.head.**descendants:**  
\#     print(child)  
print(soup.head.strings)  
print(5555)  
**for string in soup.head.strings:  \#l输出包括因换行而输出的空白  
    print(string)  
print(5555)  
for string in soup.head.stripped\_strings :  \#输出去除空白  
    print( string )**

None 
Test 
<generator object Tag.\_all\_strings at 0x000002649AD59EC8> 
5555

Title   \#用strings,前后有空白

5555  
Title   \#用stripped\_strings,前后无空白  
\[Finished in 0.5s\]

### 文档树-父节点   .parent 获得当前tag的父节点   .parents可以获得当前元素的所有父辈节点  (这个待查) ###

print( soup.p.parent.name)

content=soup.head.title.string  
print(content.parent.name)

<a>MakerBean</ a>

### 文档树-兄弟节点 ###

**兄弟节点:和当前节点在同一级的节点 
< !DOCTYPE html> 
<html lang="en"> 
<head> 
<meta charset="UTF-8"> 
<title>Title</title> 
<a class="test"></a> 
Test 
< /head> 
<body> 
< /body>  
< /html> 
上面的<a>和是兄弟节点**

**.next\_sibling获取下一个兄弟节点  
.prev\_sibling获取上一个兄弟节点**

**触类旁通**:  
如何找出当前节点的所有  
之前的兄弟节点和之后的兄弟节  
print(soup.p.next\_sibling)  
print (soup.p.prev\_sibling)  
print(soup.p.next\_sibling.prev\_sibling)

## 安装scrapy ##

PS D:\\Program Files\\Python37\\Scripts> **pip install scrapy**

**这样安装,因为有许多相关小文件需要安装,以致可能会比较慢,有时不易成功,可以反复运行这个命令.早上安装速度稍快.**

**另一种分步安装:**

PS D:\Program Files\Python37\Scripts> pip list
    Package          Version
    ---------------- ---------
    asn1crypto       0.24.0
    attrs            19.1.0
    Automat          0.7.0
    beautifulsoup4   4.8.0
    certifi          2019.9.11
    cffi             1.12.3
    chardet          3.0.4
    constantly       15.1.0
    cryptography     2.7
    cssselect        1.1.0
    hyperlink        19.0.0
    idna             2.8
    incremental      17.5.0
    lxml             4.4.1
    numpy            1.17.2
    parsel           1.5.2
    pip              19.2.3
    pyasn1           0.4.7
    pyasn1-modules   0.2.6
    pycparser        2.19
    PyDispatcher     2.0.5
    PyHamcrest       1.9.0
    pyOpenSSL        19.0.0
    queuelib         1.5.0
    requests         2.22.0
    Scrapy           1.7.3
    service-identity 18.1.0
    setuptools       41.2.0
    six              1.12.0
    soupsieve        1.9.3
    Twisted          19.7.0
    urllib3          1.25.3
    w3lib            1.21.0
    wheel            0.33.6
    zope.interface   4.6.0
    PS D:\Program Files\Python37\Scripts>

[20190915065949247.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/b9e552366b2b4e01b6a54862d2c90b51.png
[https_2.python-requests.org_en_master_key1_value1_key2_value2]: https://2.python-requests.org//en/master/?key1=value1&key2=value2
[https_2.python-requests.org_en_master_user_quickstart_make-a-request]: https://2.python-requests.org//en/master/user/quickstart/#make-a-request
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/4a1903055378478bbdddc896cb36282a.png
[https_www.crummy.com_software_BeautifulSoup_bs4_doc]: https://www.crummy.com/software/BeautifulSoup/bs4/doc/
[https_www.crummy.com_software_BeautifulSoup_bs4_doc_installing-beautiful-soup]: https://www.crummy.com/software/BeautifulSoup/bs4/doc/#installing-beautiful-soup
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 1]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/3482426504f04062b4d0f42d89a4cc61.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 2]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/32993d076f8b4b9fa63db0272e4a7aac.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 3]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/c4e807e725f9453d8d519720de04b0d1.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 4]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/6e40c00e4cbd4a969e4a9cd27385b07f.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 5]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/aa12190f975b4c08a6fa554530183eea.png
[20190917102633411.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/cdb8a297e9ee4eb79dc20fb5930d2cd5.png
[20190917103304962.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/6fe0221534bf43448411560be65bd0f9.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L2FobWN3dA_size_16_color_FFFFFF_t_70 6]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/2f7e42ea222945679a85dc2cfaf1e054.png
[20190918135729990.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/5f211f021d1f42a3930c89d79e8a5bda.png
[2019091813580056.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/04/18/dcb0f2192adb4bf3af1bcd999f5240b7.png