我们可以借助插件来做
打开插件，找到自己需要的验证码
筛选有用的路径
把对应的视图函数也拿过来，注意还需要一个geetest.py的文件
具体实
urls
#滑动验证码
url(r'^pc-geetest/register', pcgetcaptcha, name='pcgetcaptcha'),
url(r'^pc-geetest/ajax_validate', pcajax_validate, name='pcajax_validate'),
views
from app01.geetest import GeetestLib
pc_geetest_id = "b46d1900d0a894591916ea94ea91bd2c"
pc_geetest_key = "36fc3fe98530eea08dfc6ce76e3d24c4"
mobile_geetest_id = "7c25da6fe21944cfe507d2f9876775a9"
mobile_geetest_key = "f5883f4ee3bd4fa8caec67941de1b903"
# 滑动验证码
def pcgetcaptcha(request):
user_id = 'test'
gt = GeetestLib(pc_geetest_id, pc_geetest_key)
status = gt.pre_process(user_id)
request.session[gt.GT_STATUS_SESSION_KEY] = status
request.session["user_id"] = user_id
response_str = gt.get_response_str()
return HttpResponse(response_str)
# 滑动验证码
def pcajax_validate(request):
if request.method == "POST":
# 验证的验证码
ret = {"flag": False, "error_msg": None}
gt = GeetestLib(pc_geetest_id, pc_geetest_key)
challenge = request.POST.get(gt.FN_CHALLENGE, '')
validate = request.POST.get(gt.FN_VALIDATE, '')
seccode = request.POST.get(gt.FN_SECCODE, '')
status = request.session[gt.GT_STATUS_SESSION_KEY]
user_id = request.session["user_id"]
print("status",status)
if status:
result = gt.success_validate(challenge, validate, seccode, user_id)
else:
result = gt.failback_validate(challenge, validate, seccode)
if result: #如果验证验证码正确，就验证用户名是否正确
username = request.POST.get("username")
password = request.POST.get("password")
# 验证用户名和密码
user = auth.authenticate(username=username, password=password)
if user:
# 如果验证成功就让登录
ret["flag"] = True
auth.login(request, user)
else:
ret["error_msg"] = "用户名和密码错误"
else:
ret["error_msg"] = "验证码错误"
return HttpResponse(json.dumps(ret))
else:
return render(request, "login.html")
views
login.html
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width">
<title>Title</title>
<link rel="stylesheet" href="/static/bootstrap-3.3.7-dist/css/bootstrap.min.css">
<link rel="stylesheet" href="/static/css/login.css">
<script src="/static/jquery-3.2.1.min.js"></script>
滑动验证码的时候导入
<script src="http://static.geetest.com/static/tools/gt.js"></script>
<script src="/static/bootstrap-3.3.7-dist/js/bootstrap.min.js"></script>
<script src="https://cdn.bootcss.com/jquery-cookie/1.4.1/jquery.cookie.js"></script>
</head>
<body>
<div class="container">
<div class="row">
<div class="col-md-1=10">
<form class="form-horizontal" id="form_data" action="/login/" method="post">
{% csrf_token %}
<div class="form-group">
<label for="username" class="col-sm-2 control-label">用户名</label>
<div class="col-sm-5">
<input type="text" class="form-control" id="username" placeholder="username" name="username">
</div>
</div>
<div class="form-group">
<label for="password" class="col-sm-2 control-label">密码</label>
<div class="col-sm-5">
<input type="password" class="form-control" id="password" placeholder="password" name="password">
</div>
</div>
<div class="form-group">
<div class="row">
<div class="col-md-6 col-md-offset-1">
{# 文字部分#}
<label for="vialdCode" class="col-sm-2 control-label">验证码</label>
<div class="col-sm-5">
<input type="text" class="form-control vialdCode_text" id="vialdCode" placeholder="验证码" name="vialdCode">
</div>
{# 图片部分#}
<div class="col-md-5">
<img class="vialdCode_img" src="/get_vaildCode_img/" alt="" width="200px" height="100px">
{# <a href=""></a> #}
</div>
</div>
</div>
</div>
<div class="form-group">
<div class="col-sm-offset-2 col-sm-10">
<div class="checkbox">
<label>
<input type="checkbox"> 下次自动登录
</label>
</div>
</div>
</div>
<div class="form-group">
<div class="col-sm-offset-2 col-sm-10">
<p>
<button type="button" class="btn btn-success login" id="submit">登录</button>
<span class="error has-error"></span></p>
<p>
<button type="button" class="btn btn-primary register">注册</button>
</p>
</div>
<div id="popup-captcha"></div>
</div>
</form>
</div>
</div>
</div>
{#滑动验证码#}
<script>
var handlerPopup = function (captchaObj) {
$("#submit").click(function () {
captchaObj.show();
});
//定时函数
$(".login").click(function () {
function foo() {
$(".error").html("")
}
// 成功的回调
captchaObj.onSuccess(function () {
var validate = captchaObj.getValidate();
$.ajax({
url: "/pc-geetest/ajax_validate", // 进行二次验证
type: "post",
dataType: "json",
headers: {"X-CSRFToken": $.cookie('csrftoken')},
data: {
username: $('#username').val(),
password: $('#password').val(),
geetest_challenge: validate.geetest_challenge,
geetest_validate: validate.geetest_validate,
geetest_seccode: validate.geetest_seccode
},
success: function (data) {
console.log(data);
if (data["flag"]) {
{# alert(location.search);#}
{# alert(location.search.slice(6));#}
{# 方式一#}
{# if (location.search.slice(6)) {#}
{# 如果用户没有登录点赞的时候，当用户后来又登录了，就直接让跳转到当前点赞的那个路径#}
{# location.href = location.search.slice(6)#}
{# }#}
{# else {#}
{# window.location.href = '/index/'#}
{# }#}
{# 方式二：#}
alert($.cookie("next_path"));
if ($.cookie("next_path")){
location.href = $.cookie("next_path")
}
else{
location.href = "/index/"
}
}
else {
$(".error").html(data["error_msg"]);
setTimeout(foo, 3000)
}
}
});
});
});
// 将验证码加到id为captcha的元素里
captchaObj.appendTo("#popup-captcha");
// 更多接口参考：http://www.geetest.com/install/sections/idx-client-sdk.html
};
// 验证开始需要向网站主后台获取id，challenge，success（是否启用failback）
$.ajax({
url: "/pc-geetest/register?t=" + (new Date()).getTime(), // 加随机数防止缓存
type: "get",
dataType: "json",
success: function (data) {
// 使用initGeetest接口
// 参数1：配置参数
// 参数2：回调，回调的第一个参数验证码对象，之后可以使用它做appendTo之类的事件
initGeetest({
gt: data.gt,
challenge: data.challenge,
product: "popup", // 产品形式，包括：float，embed，popup。注意只对PC版验证码有效
offline: !data.success // 表示用户后台检测极验服务器是否宕机，一般不需要关注
// 更多配置参数请参见：http://www.geetest.com/install/sections/idx-client-sdk.html#config
}, handlerPopup);
}
});
</script>
login.html
爬虫
破解极验滑动验证码
一些网站会在正常运行的正常的账号密码认证之外加上一些验证
码，以此来明确地区分人行为，从一定程度上达到反爬的效果，对于简单的验证码tesserocr就可以搞定如下
图片.png
但一些网站加入了滑动验证码，
图片.png
对于这类验证,如果我们直接模拟表单请求，繁琐的认证参数与认证流程会特别的麻烦我们可以用selenium驱动浏览器来解决这个问题，大致分为
#1、输入账号、密码，然后点击登陆
#2、点击按钮，弹出没有缺口的图
#3、针对没有缺口的图片进行截图
#4、点击滑动按钮，弹出有缺口的图
#5、针对有缺口的图片进行截图
#6、对比两张图片，找出缺口，即滑动的位移
#7、按照人的行为行为习惯，把总位移切成一段段小的位移
#8、按照位移移动
#9、完成登录
实现
安装：selenium+chrome/phantomjs
#安装：Pillow
Pillow:基于PIL，处理python 3.x的图形图像库.因为PIL只能处理到python 2.x，而这个模块能处理Python3.x，目前用它做图形的很多.
http://www.cnblogs.com/apexchu/p/4231041.html
C:\Users\Administrator>pip3 install pillow
C:\Users\Administrator>python3
Python 3.6.1 (v3.6.1:69c0db5, Mar 21 2017, 18:41:36) [MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> from PIL import Image
>>>
view.code
from selenium import webdriver
from selenium.webdriver import ActionChains
from selenium.webdriver.common.by import By
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.wait import WebDriverWait
from PIL import Image
import time
def get_snap():
'''
对整个网页截图，保存成图片，然后用PIL.Image拿到图片对象
:return: 图片对象
'''
driver.save_screenshot('snap.png')
page_snap_obj=Image.open('snap.png')
return page_snap_obj
def get_image():
'''
从网页的网站截图中，截取验证码图片
:return: 验证码图片
'''
img=wait.until(EC.presence_of_element_located((By.CLASS_NAME,'geetest_canvas_img')))
time.sleep(2) #保证图片刷新出来
localtion=img.location
size=img.size
top=localtion['y']
bottom=localtion['y']+size['height']
left=localtion['x']
right=localtion['x']+size['width']
page_snap_obj=get_snap()
crop_imag_obj=page_snap_obj.crop((left,top,right,bottom))
return crop_imag_obj
def get_distance(image1,image2):
'''
拿到滑动验证码需要移动的距离
:param image1:没有缺口的图片对象
:param image2:带缺口的图片对象
:return:需要移动的距离
'''
threshold=60
left=57
for i in range(left,image1.size[0]):
for j in range(image1.size[1]):
rgb1=image1.load()[i,j]
rgb2=image2.load()[i,j]
res1=abs(rgb1[0]-rgb2[0])
res2=abs(rgb1[1]-rgb2[1])
res3=abs(rgb1[2]-rgb2[2])
if not (res1 < threshold and res2 < threshold and res3 < threshold):
return i-7 #经过测试，误差为大概为7
return i-7 #经过测试，误差为大概为7
def get_tracks(distance):
'''
拿到移动轨迹，模仿人的滑动行为，先匀加速后匀减速
匀变速运动基本公式：
①v=v0+at
②s=v0t+½at²
③v²-v0²=2as
:param distance: 需要移动的距离
:return: 存放每0.3秒移动的距离
'''
#初速度
v=0
#单位时间为0.2s来统计轨迹，轨迹即0.2内的位移
t=0.3
#位移/轨迹列表，列表内的一个元素代表0.2s的位移
tracks=[]
#当前的位移
current=0
#到达mid值开始减速
mid=distance*4/5
while current < distance:
if current < mid:
# 加速度越小，单位时间的位移越小,模拟的轨迹就越多越详细
a= 2
else:
a=-3
#初速度
v0=v
#0.2秒时间内的位移
s=v0*t+0.5*a*(t**2)
#当前的位置
current+=s
#添加到轨迹列表
tracks.append(round(s))
#速度已经达到v,该速度作为下次的初速度
v=v0+a*t
return tracks
try:
driver=webdriver.Chrome()
driver.get('https://account.geetest.com/login')
wait=WebDriverWait(driver,10)
#步骤一：先点击按钮，弹出没有缺口的图片
button=wait.until(EC.presence_of_element_located((By.CLASS_NAME,'geetest_radar_tip')))
button.click()
#步骤二：拿到没有缺口的图片
image1=get_image()
#步骤三：点击拖动按钮，弹出有缺口的图片
button=wait.until(EC.presence_of_element_located((By.CLASS_NAME,'geetest_slider_button')))
button.click()
#步骤四：拿到有缺口的图片
image2=get_image()
# print(image1,image1.size)
# print(image2,image2.size)
#步骤五：对比两张图片的所有RBG像素点，得到不一样像素点的x值，即要移动的距离
distance=get_distance(image1,image2)
#步骤六：模拟人的行为习惯（先匀加速拖动后匀减速拖动），把需要拖动的总距离分成一段一段小的轨迹
tracks=get_tracks(distance)
print(tracks)
print(image1.size)
print(distance,sum(tracks))
#步骤七：按照轨迹拖动，完全验证
button=wait.until(EC.presence_of_element_located((By.CLASS_NAME,'geetest_slider_button')))
ActionChains(driver).click_and_hold(button).perform()
for track in tracks:
ActionChains(driver).move_by_offset(xoffset=track,yoffset=0).perform()
else:
ActionChains(driver).move_by_offset(xoffset=3,yoffset=0).perform() #先移过一点
ActionChains(driver).move_by_offset(xoffset=-3,yoffset=0).perform() #再退回来，是不是更像人了
time.sleep(0.5) #0.5秒后释放鼠标
ActionChains(driver).release().perform()
#步骤八：完成登录
input_email=driver.find_element_by_id('email')
input_password=driver.find_element_by_id('password')
button=wait.until(EC.element_to_be_clickable((By.CLASS_NAME,'login-btn')))
input_email.send_keys('18611453110@163.com')
input_password.send_keys('linhaifeng123')
# button.send_keys(Keys.ENTER)
button.click()
import time
time.sleep(200)
finally:
driver.close()
案列：
破解博客园后台登录
from selenium import webdriver
from selenium.webdriver import ActionChains
from selenium.webdriver.common.by import By
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.wait import WebDriverWait
from PIL import Image
import time
def get_snap():
driver.save_screenshot('full_snap.png')
page_snap_obj=Image.open('full_snap.png')
return page_snap_obj
def get_image():
img=driver.find_element_by_class_name('geetest_canvas_img')
time.sleep(2)
location=img.location
size=img.size
left=location['x']
top=location['y']
right=left+size['width']
bottom=top+size['height']
page_snap_obj=get_snap()
image_obj=page_snap_obj.crop((left,top,right,bottom))
# image_obj.show()
return image_obj
def get_distance(image1,image2):
start=57
threhold=60
for i in range(start,image1.size[0]):
for j in range(image1.size[1]):
rgb1=image1.load()[i,j]
rgb2=image2.load()[i,j]
res1=abs(rgb1[0]-rgb2[0])
res2=abs(rgb1[1]-rgb2[1])
res3=abs(rgb1[2]-rgb2[2])
# print(res1,res2,res3)
if not (res1 < threhold and res2 < threhold and res3 < threhold):
return i-7
return i-7
def get_tracks(distance):
distance+=20 #先滑过一点，最后再反着滑动回来
v=0
t=0.2
forward_tracks=[]
current=0
mid=distance*3/5
while current < distance:
if current < mid:
a=2
else:
a=-3
s=v*t+0.5*a*(t**2)
v=v+a*t
current+=s
forward_tracks.append(round(s))
#反着滑动到准确位置
back_tracks=[-3,-3,-2,-2,-2,-2,-2,-1,-1,-1] #总共等于-20
return {'forward_tracks':forward_tracks,'back_tracks':back_tracks}
try:
# 1、输入账号密码回车
driver = webdriver.Chrome()
driver.implicitly_wait(3)
driver.get('https://passport.cnblogs.com/user/signin')
username = driver.find_element_by_id('input1')
pwd = driver.find_element_by_id('input2')
signin = driver.find_element_by_id('signin')
username.send_keys('linhaifeng')
pwd.send_keys('xxxxx')
signin.click()
# 2、点击按钮，得到没有缺口的图片
button = driver.find_element_by_class_name('geetest_radar_tip')
button.click()
# 3、获取没有缺口的图片
image1 = get_image()
# 4、点击滑动按钮，得到有缺口的图片
button = driver.find_element_by_class_name('geetest_slider_button')
button.click()
# 5、获取有缺口的图片
image2 = get_image()
# 6、对比两种图片的像素点，找出位移
distance = get_distance(image1, image2)
# 7、模拟人的行为习惯，根据总位移得到行为轨迹
tracks = get_tracks(distance)
print(tracks)
# 8、按照行动轨迹先正向滑动，后反滑动
button = driver.find_element_by_class_name('geetest_slider_button')
ActionChains(driver).click_and_hold(button).perform()
# 正常人类总是自信满满地开始正向滑动，自信地表现是疯狂加速
for track in tracks['forward_tracks']:
ActionChains(driver).move_by_offset(xoffset=track, yoffset=0).perform()
# 结果傻逼了，正常的人类停顿了一下，回过神来发现，卧槽，滑过了,然后开始反向滑动
time.sleep(0.5)
for back_track in tracks['back_tracks']:
ActionChains(driver).move_by_offset(xoffset=back_track, yoffset=0).perform()
# 小范围震荡一下，进一步迷惑极验后台，这一步可以极大地提高成功率
ActionChains(driver).move_by_offset(xoffset=-3, yoffset=0).perform()
ActionChains(driver).move_by_offset(xoffset=3, yoffset=0).perform()
# 成功后，骚包人类总喜欢默默地欣赏一下自己拼图的成果，然后恋恋不舍地松开那只脏手
time.sleep(0.5)
ActionChains(driver).release().perform()
time.sleep(10) # 睡时间长一点，确定登录成功
finally:
driver.close()
修订版本
from selenium import webdriver
from selenium.webdriver import ActionChains
from selenium.webdriver.common.by import By
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.wait import WebDriverWait
from PIL import Image
import time
def get_snap(driver):
driver.save_screenshot('full_snap.png')
page_snap_obj=Image.open('full_snap.png')
return page_snap_obj
def get_image(driver):
img=driver.find_element_by_class_name('geetest_canvas_img')
time.sleep(2)
location=img.location
size=img.size
left=location['x']
top=location['y']
right=left+size['width']
bottom=top+size['height']
page_snap_obj=get_snap(driver)
image_obj=page_snap_obj.crop((left,top,right,bottom))
# image_obj.show()
return image_obj
def get_distance(image1,image2):
start=57
threhold=60
for i in range(start,image1.size[0]):
for j in range(image1.size[1]):
rgb1=image1.load()[i,j]
rgb2=image2.load()[i,j]
res1=abs(rgb1[0]-rgb2[0])
res2=abs(rgb1[1]-rgb2[1])
res3=abs(rgb1[2]-rgb2[2])
# print(res1,res2,res3)
if not (res1 < threhold and res2 < threhold and res3 < threhold):
return i-7
return i-7
def get_tracks(distance):
distance+=20 #先滑过一点，最后再反着滑动回来
v=0
t=0.2
forward_tracks=[]
current=0
mid=distance*3/5
while current < distance:
if current < mid:
a=2
else:
a=-3
s=v*t+0.5*a*(t**2)
v=v+a*t
current+=s
forward_tracks.append(round(s))
#反着滑动到准确位置
back_tracks=[-3,-3,-2,-2,-2,-2,-2,-1,-1,-1] #总共等于-20
return {'forward_tracks':forward_tracks,'back_tracks':back_tracks}
def crack(driver): #破解滑动认证
# 1、点击按钮，得到没有缺口的图片
button = driver.find_element_by_class_name('geetest_radar_tip')
button.click()
# 2、获取没有缺口的图片
image1 = get_image(driver)
# 3、点击滑动按钮，得到有缺口的图片
button = driver.find_element_by_class_name('geetest_slider_button')
button.click()
# 4、获取有缺口的图片
image2 = get_image(driver)
# 5、对比两种图片的像素点，找出位移
distance = get_distance(image1, image2)
# 6、模拟人的行为习惯，根据总位移得到行为轨迹
tracks = get_tracks(distance)
print(tracks)
# 7、按照行动轨迹先正向滑动，后反滑动
button = driver.find_element_by_class_name('geetest_slider_button')
ActionChains(driver).click_and_hold(button).perform()
# 正常人类总是自信满满地开始正向滑动，自信地表现是疯狂加速
for track in tracks['forward_tracks']:
ActionChains(driver).move_by_offset(xoffset=track, yoffset=0).perform()
# 结果傻逼了，正常的人类停顿了一下，回过神来发现，卧槽，滑过了,然后开始反向滑动
time.sleep(0.5)
for back_track in tracks['back_tracks']:
ActionChains(driver).move_by_offset(xoffset=back_track, yoffset=0).perform()
# 小范围震荡一下，进一步迷惑极验后台，这一步可以极大地提高成功率
ActionChains(driver).move_by_offset(xoffset=-3, yoffset=0).perform()
ActionChains(driver).move_by_offset(xoffset=3, yoffset=0).perform()
# 成功后，骚包人类总喜欢默默地欣赏一下自己拼图的成果，然后恋恋不舍地松开那只脏手
time.sleep(0.5)
ActionChains(driver).release().perform()
def login_cnblogs(username,password):
driver = webdriver.Chrome()
try:
# 1、输入账号密码回车
driver.implicitly_wait(3)
driver.get('https://passport.cnblogs.com/user/signin')
input_username = driver.find_element_by_id('input1')
input_pwd = driver.find_element_by_id('input2')
signin = driver.find_element_by_id('signin')
input_username.send_keys(username)
input_pwd.send_keys(password)
signin.click()
# 2、破解滑动认证
crack(driver)
time.sleep(10) # 睡时间长一点，确定登录成功
finally:
driver.close()
if __name__ == '__main__':
login_cnblogs(username='linhaifeng',password='xxxx')
