一、常见反爬机制及其破解方式
封禁IP,使用cookie等前面文章已经讲过
现在主要将下面的:
~ 验证码
—> 文字验证码 —> OCR(光学文字识别)—> 接口 / easyocr
程序自己解决不了的问题就可以考虑使用三方接口(付费/免费)
—> 行为验证码 —> 超级鹰
~ 手机号+短信验证码
—> 接码平台
~ 动态内容
—> JavaScript逆向 —> 找到提供数据的API接口
—> 手机抓接口 —> 抓包工具(Charles / Fiddler)
—> Selenium直接模拟浏览器操作获取动态内容
~ find_element_by_xxx / find_elements_by_xxx
~ page_source —> 获取包含动态内容的网页源代码
—> JavaScript加密和混淆技术 —> 读懂JavaScript是反反爬的前提
~ 字体反爬 / 内容来自于抠图
—> 例子
bytes —> 不变字节串 —> 二进制 —> BytesIO
str —> 不变字符串 —> 可阅读的字符 —> StringIO
二、调用三方API接口数据(天行数据)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
|
import requests
for page in range ( 1 , 6 ):
response = requests.get(
'http://api.tianapi.com/topnews/index' ,
params = {
'key' : 'd5eace66dccd771e36767ce3563efa09' ,
'page' : page,
'num' : 20 ,
'word' : '华为' ,
'src' : '人民日报'
}
)
result = response.json()
for news in result[ 'newslist' ]:
print (news[ 'title' ])
print (news[ 'url' ])
|
三、OCR(光学文字识别)库
python 自带的easyocr库
1
2
3
|
import easyocr
reader = easyocr.Reader([ 'ch_sim' , 'en' ], gpu = False )
print (reader.readtext( './files/captcha.jpg' , detail = 0 ))
|
例子:阿里云邮箱自动登陆
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
|
import io
import easyocr
from PIL import Image
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions
from selenium.webdriver.support.wait import WebDriverWait
browser = webdriver.Chrome()
browser.set_window_size( 1280 , 960 )
browser.get( 'http://mail.1000phone.com/' )
# 隐式等待(下面的方法在工作时如果取不到就等10秒)
browser.implicitly_wait( 10 )
# 显式等待
wait = WebDriverWait(browser, 10 )
wait.until(expected_conditions.presence_of_element_located((By.CSS_SELECTOR, '.login_panel_iframe' )))
iframe1 = browser.find_element_by_css_selector( '.login_panel_iframe' )
# 记录iframe1的位置(相对位置)
x1, y1 = iframe1.location[ 'x' ], iframe1.location[ 'y' ]
# Chrome对象的switch_to属性的frame方法,可以从页面切换到iframe中
browser.switch_to.frame(iframe1)
iframe2 = browser.find_element_by_css_selector( '#ding-login-iframe' )
x2, y2 = iframe2.location[ 'x' ], iframe2.location[ 'y' ]
browser.switch_to.frame(iframe2)
username_input = browser.find_element_by_css_selector( '#username' )
# 模拟用户输入
username_input.send_keys( 'xx@1000phone.com' )
password_input = browser.find_element_by_css_selector( '#password' )
password_input.send_keys( 'xxxxx!!' )
# 创建一个等待对象
wait = WebDriverWait(browser, 10 )
wait.until(expected_conditions.element_to_be_clickable((By.CSS_SELECTOR, '#login_checkcode_ico' )))
captcha_img = browser.find_element_by_css_selector( '#login_checkcode_ico' )
# WebElement对象的size属性代表元素宽度和高度,location属性代表元素在窗口中的位置
size, location = captcha_img.size, captcha_img.location
x3, y3, width, height = location[ 'x' ], location[ 'y' ], size[ 'width' ], size[ 'height' ]
# 截取整个浏览器窗口的图片获得图片的二进制数据
image_data = browser.get_screenshot_as_png()
# bytes(只读字节串) ----> io.BytesIO(可写字节串)---> getvalue() ---> bytes
# str(只读字符串) ----> io.StringIO(可写字符串)---> getvalue() ---> str
browser_image = Image. open (io.BytesIO(image_data))
# 从截图上剪裁出验证码的图片
x, y = x1 + x2 + x3, y1 + y2 + y3
# Windows系统的写法 ---> 如果截图有问题就把坐标写死
# print(x, y, width, height)
checkcode_image = browser_image.crop((x * 1.25 , y * 1.25 , (x + width) * 1.25 , (y + height) * 1.25 ))
# macOS系统的写法
# checkcode_image = browser_image.crop((x * 2, y * 2, (x + width) * 2, (y + height) * 2))
checkcode_image.save( 'result.png' )
# 通过easyocr做光学文字识别
reader = easyocr.Reader([ 'en' ], gpu = False )
code = reader.readtext( 'result.png' , detail = 0 )[ 0 ]
# 将识别出的验证码输入文本框
checkcode_input = browser.find_element_by_css_selector( '#login_checkcode' )
checkcode_input.send_keys(code)
login_button = browser.find_element_by_css_selector( '#login_submit_btn' )
# 模拟用户点击
login_button.click()
|
四、第三方打码平台(超级鹰打码平台)
补充:需要使用python 自带pillow库
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
|
"""
Pillow库 ---> PIL ---> Python Image Library
"""
from PIL import Image, ImageFilter
# 加载图像
guido_image = Image. open ( 'guido.jpg' )
# 剪裁
guido_image.crop(( 80 , 40 , 310 , 350 )).show()
# 滤镜
guido_image. filter (ImageFilter.CONTOUR).show()
# 缩略图
guido_image.thumbnail(( 125 , 185 ))
# 显示图像
guido_image.show()
|
编写超级鹰打码平台类
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
|
from hashlib import md5
import requests
class ChaojiyingClient:
def __init__( self , username, password, soft_id):
self .username = username
password = password.encode( 'utf8' )
self .password = md5(password).hexdigest()
self .soft_id = soft_id
self .base_params = {
'user' : self .username,
'pass2' : self .password,
'softid' : self .soft_id,
}
self .headers = {
'Connection' : 'Keep-Alive' ,
'User-Agent' : 'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0)' ,
}
def post_pic( self , image_data, code_type):
"""
image_data: 图片字节
code_type: 验证码类型 参考 http://www.chaojiying.com/price.html
"""
params = {
'codetype' : code_type,
}
params.update( self .base_params)
files = { 'userfile' : ( 'ccc.jpg' , image_data)}
response = requests.post(
url = 'http://upload.chaojiying.net/Upload/Processing.php' ,
data = params,
files = files,
headers = self .headers
)
return response.json()
# 超级鹰错误反馈函数(仅用于给超级鹰平台反馈)
def report_error( self , im_id):
"""
im_id:报错题目的图片ID
"""
params = {
'id' : im_id,
}
params.update( self .base_params)
r = requests.post( 'http://upload.chaojiying.net/Upload/ReportError.php' , data = params, headers = self .headers)
return r.json()
if __name__ = = '__main__' :
chaojiying = ChaojiyingClient( '账户' , '密码x' , 'ID' ) # 用户中心>>软件ID 生成一个替换 96001
with open ( 'img.png' , 'rb' ) as file :
image_data = file .read() # 本地图片文件路径 来替换 a.jpg 有时WIN系统须要//
print (chaojiying.post_pic(image_data, 1902 )) # 1902 验证码类型 官方网站>>价格体系 3.4+版 print 后要加()
|
例子:使用超级鹰进行阿里云邮箱自动登陆
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
|
import io
import easyocr
from PIL import Image
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions
from selenium.webdriver.support.wait import WebDriverWait
from chaojiying import ChaojiyingClient
browser = webdriver.Chrome()
browser.set_window_size( 1280 , 960 )
browser.get( 'http://mail.1000phone.com/' )
# 隐式等待(下面的方法在工作时如果取不到就等10秒)
browser.implicitly_wait( 10 )
# 显式等待
wait = WebDriverWait(browser, 10 )
wait.until(expected_conditions.presence_of_element_located((By.CSS_SELECTOR, '.login_panel_iframe' )))
iframe1 = browser.find_element_by_css_selector( '.login_panel_iframe' )
# 记录iframe1的位置(相对位置)
x1, y1 = iframe1.location[ 'x' ], iframe1.location[ 'y' ]
# Chrome对象的switch_to属性的frame方法,可以从页面切换到iframe中
browser.switch_to.frame(iframe1)
iframe2 = browser.find_element_by_css_selector( '#ding-login-iframe' )
x2, y2 = iframe2.location[ 'x' ], iframe2.location[ 'y' ]
browser.switch_to.frame(iframe2)
username_input = browser.find_element_by_css_selector( '#username' )
# 模拟用户输入
username_input.send_keys( 'xxxx.com' )
password_input = browser.find_element_by_css_selector( '#password' )
password_input.send_keys( 'xxxx!!' )
# 创建一个等待对象
wait = WebDriverWait(browser, 10 )
wait.until(expected_conditions.element_to_be_clickable((By.CSS_SELECTOR, '#login_checkcode_ico' )))
captcha_img = browser.find_element_by_css_selector( '#login_checkcode_ico' )
# WebElement对象的size属性代表元素宽度和高度,location属性代表元素在窗口中的位置
size, location = captcha_img.size, captcha_img.location
x3, y3, width, height = location[ 'x' ], location[ 'y' ], size[ 'width' ], size[ 'height' ]
# 截取整个浏览器窗口的图片获得图片的二进制数据
image_data = browser.get_screenshot_as_png()
# bytes(只读字节串) ----> io.BytesIO(可写字节串)---> getvalue() ---> bytes
# str(只读字符串) ----> io.StringIO(可写字符串)---> getvalue() ---> str
browser_image = Image. open (io.BytesIO(image_data))
# 从截图上剪裁出验证码的图片
x, y = x1 + x2 + x3, y1 + y2 + y3
# Windows系统的写法 ---> 如果截图有问题就把坐标写死
# print(x, y, width, height)
checkcode_image = browser_image.crop((x * 1.25 , y * 1.25 , (x + width) * 1.25 , (y + height) * 1.25 ))
# macOS系统的写法
# checkcode_image = browser_image.crop((x * 2, y * 2, (x + width) * 2, (y + height) * 2))
checkcode_image.save( 'result.png' )
# 通过超级鹰打码平台打码
chaojiying = ChaojiyingClient( '账户' , '密码' , 'ID' )
with open ( 'result.png' , 'rb' ) as file :
image_data = file .read()
result_dict = chaojiying.post_pic(image_data, 1902 )
# 将识别出的验证码输入文本框
checkcode_input = browser.find_element_by_css_selector( '#login_checkcode' )
checkcode_input.send_keys(result_dict[ 'pic_str' ])
login_button = browser.find_element_by_css_selector( '#login_submit_btn' )
# 模拟用户点击
login_button.click()
|
五、通过接码平台接收手机验证码(隐私短信平台)
通过隐私短信平台接收验证码(免费)
1
2
3
4
5
6
7
8
9
10
11
12
13
|
import re
import bs4
import requests
pattern = re. compile (r '\d{4,6}' )
resp = requests.get( 'https://www.yinsiduanxin.com/china-phone-number/verification-code-16521686439.html' )
soup = bs4.BeautifulSoup(resp.text, 'html.parser' )
# print(resp.text)
td = soup.select_one( 'body > div.container > div:nth-child(4) > div:nth-child(3) > div.main > div.layui-row > table > tbody > tr:nth-child(1) > td:nth-child(2)' )
results = pattern.findall(td.text)
print (results[ 0 ])
|
到此这篇关于Python中常见的反爬机制及其破解方法总结的文章就介绍到这了,更多相关Python反爬机制及其破解内容请搜索服务器之家以前的文章或继续浏览下面的相关文章希望大家以后多多支持服务器之家!
原文链接:https://blog.csdn.net/m0_54255329/article/details/117673635