重点：对12306验证码的破解 -------仅供学习交流，请勿…
1.找到验证码的图片信息
2.点开headers查看（不难发现是经过base64加密的图片），但是没有请求的url
3.再查看分析发现（上面的一个js文件里也存储这图片信息）
4.点开headers查看 Requests Url
5.对地址发起请求下载验证码图片（别忘记 要把base64解密 哈哈 好坑）
Xq_url = 'https://kyfw.12306.cn/passport/captcha/captcha-image64'
#构造请求表单
Xq_parmas = {
"login_site": "E",
"module": "login",
"rand": "sjrand",
"1555036551227":"",
"callback": "jQuery19108754385247664451_1555036549517",
"_": "1555036549518",
}
#对图片页发送请求
response = sesson.get(url=Xq_url,params=Xq_parmas,headers=headers).text
#获取图片数据
image_bs64 = re.findall('"image":"(.*?)",',response)[0]
#解码数据
image = base64.b64decode(image_bs64)
#保存图片
with open('Yz_image.jpg','wb') as f:
f.write(image)
6.对验证码分析（经过几次实验 可以很容易确定传入的坐标是以像素位置传入的）
7.经过点击每个图片确定位置如下（弄得想吐呀），终于可以构造像素表单了
#椅子垫形式对应图片位置和像素
coord_data = {
"1":"40,40",
"2":"120,40",
"3":"180,40",
"4":"250,40",
"5":"40,100",
"6":"120,100",
"7":"180,100",
"8":"250,100",}
answerlist = []
input_answer = input('位置：')
#对输入的数字进行切割
answer_list = input_answer.split(' ')
#循环输入的值取出字典相应的坐标
for i in answer_list:
answerlist.append(coord_data.get(i))
print('answer：' + ','.join(answerlist))
answer = ','.join(answerlist)
8.验证码破解了，就可以模拟登录了 哈哈哈
分析可以发现在login里有我们要的东西 Fromdata 和 RequestURL
构造Formdata 发起post请求
Sy_url = 'https://kyfw.12306.cn/passport/web/login'
formdata = {
"username": 你的12306账号,
"password": 你的12306密码,
"appid": "otn",
"answer": answer,
}
#对比分析加入下面三个字段的cookies
sesson.cookies.update({
'RAIL_EXPIRATION':'1555310364529',
'RAIL_DEVICEID':'K71MCVMaU_fg6Lr5kLs9K5-HrmV-F-LdSuahWpFSW60X8GmWMZiS06V7InVpguAyYJOmW3cNWKx8Giau-aCZEqehQzwLYRMwjHxr1v1EkKjGTn_iX87fpiWNuGK_jPgg-1PgNgIpFMHeEEfDmXfwdmXX2nGNCcuC',
'route':'495c805987d0f5c8c84b14f60212447d',
})
response = sesson.post(Sy_url,data=formdata)
#返回编码后的数据
response.encoding = 'utf-8'
print(re.findall('"result_message":"(.*?)"',response.text))
9.附上源码
import requests
import re
import base64
#创建sesson对象 保持会话一至
sesson = requests.session()
#请求的url
Sy_url = 'https://kyfw.12306.cn/passport/web/login'
Xq_url = 'https://kyfw.12306.cn/passport/captcha/captcha-image64'
Yz_url = 'https://kyfw.12306.cn/passport/captcha/captcha-check'
#构造请求头
headers = {
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.86 Safari/537.36"
}
#构造请求表单
Xq_parmas = {
"login_site": "E",
"module": "login",
"rand": "sjrand",
"1555036551227":"",
"callback": "jQuery19108754385247664451_1555036549517",
"_": "1555036549518",
}
#对图片页发送请求
response = sesson.get(url=Xq_url,params=Xq_parmas,headers=headers).text
#获取图片数据
image_bs64 = re.findall('"image":"(.*?)",',response)[0]
#解码数据
image = base64.b64decode(image_bs64)
#保存图片
with open('Yz_image.jpg','wb') as f:
f.write(image)
#构建像素表单
coord_data = {
"1":"40,40",
"2":"120,40",
"3":"180,40",
"4":"250,40",
"5":"40,100",
"6":"120,100",
"7":"180,100",
"8":"250,100",}
answerlist = []
input_answer = input('位置：')
#对输入的数字进行切割
answer_list = input_answer.split(' ')
#循环输入的值取出字典相应的坐标
for i in answer_list:
answerlist.append(coord_data.get(i))
print('answer：' + ','.join(answerlist))
answer = ','.join(answerlist)
#构造data表单
formdata = {
"username": 你的12306账号,
"password": 你的12306密码,
"appid": "otn",
"answer": answer,
}
Yz_parmas = {
"callback": "jQuery19108754385247664451_1555036549517",
"answer": answer,
"rand": "sjrand",
"login_site": "E",
"_": "1555036549519",
}
#发送图片验证请求
response_new = sesson.get(url=Yz_url,params=Yz_parmas,headers=headers).text
#获得图片验证信息
print(re.findall('"result_message":"(.*?)"',response_new))
#增加cookies
sesson.cookies.update({
'RAIL_EXPIRATION':'1555310364529',
'RAIL_DEVICEID':'K71MCVMaU_fg6Lr5kLs9K5-HrmV-F-LdSuahWpFSW60X8GmWMZiS06V7InVpguAyYJOmW3cNWKx8Giau-aCZEqehQzwLYRMwjHxr1v1EkKjGTn_iX87fpiWNuGK_jPgg-1PgNgIpFMHeEEfDmXfwdmXX2nGNCcuC',
'route':'495c805987d0f5c8c84b14f60212447d',
})
response = sesson.post(Sy_url,data=formdata)
response.encoding = 'utf-8'
print(re.findall('"result_message":"(.*?)"',response.text))
10.最后的效果
