爬蟲-模擬登入

阿新 • • 發佈：2020-07-20

模擬登入的程式碼實現：

#socket服務端
import socket
import json
import threading
server = socket.socket()
#繫結到0.0.0.0:8000埠上
server.bind(('0.0.0.0', 8002))
server.listen()

#服務在使用者登入成功之後，給使用者返回一段字串sessionid(夠複雜，生成演算法別人偽造不了)
user_info = {
    "sessionid":"bobby"
}
#瀏覽器每一次請求(所有的url)都自動帶上這個sessionid
#1.如何告知瀏覽器這個sessionid
#2.如何確保瀏覽器每一次請求都帶上這個sessionid

#session和cookie的區別
# 
1. session是由伺服器維護的，並由伺服器解釋，通過set-cookie交給瀏覽器
#2. cookie是瀏覽器的工具，並在後續的每一次請求中都帶上這些值


def handle_sock(sock, addr):
    while True:
        # recv方法是阻塞的
        tmp_data = sock.recv(1024)
        print(tmp_data.decode("utf8"))
        response_template = '''HTTP/1.0 200 OK  
Content-type: text/html  
Set-Cookie: name=bobby
Set 
-Cookie: course_id=78
Set-Cookie: sessionid=abc123; Expires=Wed, 09 Jun 2021 10:18:14 GMT

{}

'''
        data = [
            {
                "name":"django打造線上教育",
                "teacher":"bobby",
                "url":"https://coding.imooc.com/class/78.html"
            },
            {
                 
"name": "python高階程式設計",
                "teacher": "bobby",
                "url": "https://coding.imooc.com/class/200.html"
            },
            {
                "name": "scrapy分散式爬蟲",
                "teacher": "bobby",
                "url": "https://coding.imooc.com/class/92.html"
            },
            {
                "name": "django rest framework打造生鮮電商",
                "teacher": "bobby",
                "url": "https://coding.imooc.com/class/131.html"
            },
            {
                "name": "tornado從入門到精通",
                "teacher": "bobby",
                "url": "https://coding.imooc.com/class/290.html"
            },
        ]
        sock.send(response_template.format(json.dumps(data)).encode("utf8"))
        sock.close()
        break

#獲取客戶端連線並啟動執行緒去處理
while True:
    # 阻塞等待連線
    sock, addr = server.accept()

    #啟動一個執行緒去處理新的使用者連線
    client_thread = threading.Thread(target=handle_sock, args=(sock, addr))
    client_thread.start()

requests+session模擬登入豆瓣：

import json
import pickle

import requests


def login():
    session = requests.session()
    username = "18782902568"
    password = "admin123"
    url = "https://accounts.douban.com/j/mobile/login/basic"
    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"
    }

    post_data = {
        "ck": "",
        "name": username,
        "password": password,
        "remember": "true",
        "ticket": ""
    }

    res = session.post(url, data=post_data, headers=headers)
    res_json = json.loads(res.text)
    if res_json["status"] == "success":
        print("登入成功")
        with open("douban.cookie", "wb") as f:
            pickle.dump(res.cookies, f)
    else:
        print("登入失敗")

    with open("douban.cookie", "rb") as f:
        cookies = pickle.load(f)
        html = requests.get("https://www.douban.com/", cookies=cookies).text
        if "bobby_liyao" in html:
            print("已經登入")
        else:
            print("未登入")


if __name__ == "__main__":
    login()

注意：

1】使用requests.session()而不是requests。是為了使得登入的cookie能夠實現共享。

2】使用pickle，是為了儲存與獲取序列化資料更加的便捷。

3】當然資料在無需儲存到檔案裡面，可以直接使用res.cookies或者序列化為字典res.cookie.get_dic()。獲取cookie並傳遞。

selenium模擬登入：

import time

import requests
from selenium import webdriver

url = "https://www.douban.com/"
browser = webdriver.Chrome(executable_path="E:/in32/chromedriver.exe")


def login():
    #通過selenium模擬登入都豆瓣
    username = "18782902568"
    password = "admin123"
    browser.get(url)
    time.sleep(3)
　　#切換到frame
    browser.switch_to.frame(browser.find_element_by_tag_name("iframe"))
    login_ele = browser.find_element_by_xpath("//li[@class='account-tab-account']")
    login_ele.click()

    username_ele = browser.find_element_by_xpath("//input[@id='username']")
    password_ele = browser.find_element_by_xpath("//input[@id='password']")
    username_ele.send_keys(username)
    password_ele.send_keys(password)
#解決按鈕屬性變動
    submit_btn = browser.find_element_by_xpath("//a[@class='btn btn-account btn-active']")
    submit_btn.click()

    time.sleep(10)
#得到[{}],轉化為字典進行傳送
    cookies = browser.get_cookies()
    cookie_dict = {}
    for item in cookies:
        cookie_dict[item["name"]] = item["value"]

    res = requests.get(url, cookies=cookie_dict)
    if "bobby_liyao" in res.text:
        print("已經登入")


if __name__ == "__main__":
    login()