基本环境配置
1. Selenium- 建议使用低版本的Python-Selenium库，因为高版本在Chrome中不支持。
pip3 install selenium==2.48.0
2. chromedriver.exe- 下载地址：/mirrors/chromedriver/, 版本要匹配。将 .exe 程序放在"..Python\Python36\Scripts"目录下。
3. pymysql
pip3 install pymysql
1.登录界面iframe
iframe的id值添加了时间戳,直接获取相对麻烦。可通过XPATH或CSS选择器获取该节点。如图：
2. "写信"节点
写信节点的元素定位li节点, 不要定位span子节点，否则获取不到。另外，如果是获取APP节点，可以选择小一级的。
3. 邮件主题
主题节点不可交互，无法输入文字，这里选择不设置。
4. 邮件内容
邮件内容的文本输入框处于iframe中，输入文本前需要切换frame，可直接通过class获取并切换。如图：
5. "发送"节点
由于输入邮件内容时切换至子页面，在点击发送前需要切换到父级Frame。
6. 登录限制
不要频繁使用Selenium, 否则会出现点触式验证。当然，完全可以破解。但是，网易相对友好，短时间过后便可恢复正常访问，也不会ban IP。
from selenium import webdriverfrom mon.by import Byfrom selenium.webdriver.support.ui import WebDriverWaitfrom selenium.webdriver.support import expected_conditions as ECfrom mon.exceptions import NoSuchElementExceptionfrom utils.config import *import pymysqlimport reimport time'''想要学习Python？Python学习交流群：984632579满足你的需求，资料都已经上传群文件，可以自行下载！'''class Mailbox(object):def __init__(self):"""初始化"""self.url = URLself.browser = webdriver.Chrome()self.wait = WebDriverWait(self.browser, TIME_OUT) # 显式等待时间10秒# 邮箱账号密码self.mail_user = MAIL_USERself.mail_pass = MAIL_PASS# MySql用户名密码# self.user = MYSQL_USER# self.pass_ = MYSQL_PASSdef __del__(self):self.browser.close()def login_email(self):"""登陆网页版网易163邮箱:return: None"""self.browser.get(self.url)# 获取iframe节点frame_1 = self.wait.until(EC.presence_of_element_located((By.XPATH,'//*[@id="loginDiv"]/iframe')))# 切换Frameself.browser.switch_to.frame(frame_1)time.sleep(3)# 账号输入框user = self.wait.until(EC.presence_of_element_located((By.NAME, 'email')))# 密码输入框password = self.wait.until(EC.presence_of_element_located((By.NAME, 'password')))# 登录按钮button = self.wait.until(EC.element_to_be_clickable((By.ID, 'dologin')))# 输入账号user.clear()user.send_keys(self.mail_user)time.sleep(3)# 输入密码password.clear()password.send_keys(self.mail_pass)time.sleep(3)# 登录button.click()time.sleep(10)# 获取登录成功后界面左上角的Logo,验证登录成功.logo = self.wait.until(EC.presence_of_element_located((By.ID, 'h1Logo')))if logo:print('Successful login!')def send_email(self):"""发送邮件:return: None"""try:# "写信"节点send_ = self.wait.until(EC.presence_of_element_located((By.XPATH, '//*[contains(@class, "mD0")]')))send_.click()time.sleep(5)# 收件人recipient = self.wait.until(EC.presence_of_element_located((By.CLASS_NAME, 'nui-editableAddr-ipt')))recipient.clear()recipient.send_keys(RECIPIENT)time.sleep(3)# 切换frameframe = self.wait.until(EC.presence_of_element_located((By.XPATH, '//*[@class="APP-editor-iframe"]')))self.browser.switch_to.frame(frame)# 邮件内容content = self.wait.until(EC.presence_of_element_located((By.XPATH, '//*[@class="nui-scroll"]')))content.clear()content.send_keys(CONTENT)time.sleep(3)# 切换至父级frameself.browser.switch_to.parent_frame()# 发送send_button = self.wait.until(EC.element_to_be_clickable((By.XPATH, '//*[@class="jp0"]/div[1]/span[1]')))send_button.click()time.sleep(3)# 确定不添加主题sure = self.wait.until(EC.presence_of_element_located((By.XPATH, '//*[@class="nui-msgbox-ft-btns"]/div[1]/span')))sure.click()time.sleep(3)print('Send successfully！')except NoSuchElementException as e:print('Failed!')'''def get_email_info(self):"""获取邮件的相关信息:return: None"""# 收件箱节点Inbox = self.wait.until(EC.presence_of_element_located((By.XPATH, '//*[@id="dvNavTree"]/ul/li[1]')))# 切换至"收件箱"页面Inbox.click()# 延时等待time.sleep(10)# messagemessages = list()try:# 获取包含页码的节点page_element = self.wait.until(EC.presence_of_element_located((By.XPATH, '//*[@class="nui-select-text"]')))# 获取收件箱总页码page = page_element.text.split('/')[-1]time.sleep(3)if page:for i in range(int(page)):# 获取当前页面所有邮箱节点divs = self.browser.find_elements((By.XPATH, '//*[contains(@class, "rF0")]'))print(divs)# divs = self.wait.until(EC.presence_of_element_located((By.CSS_SELECTOR,'')))for div in range(len(divs)):# 获取每个div节点的aria-label属性(包含所有有效信息)summary = div.get_attritubute('aria-label')# 获取aria-label属性值result = summary.split()# result = re.findall(r'(.*?) 发件人 ：(.*?) 时间：(.*?$)', s) 正则匹配有效数据messages.append({'sender': result[3],'content': result[0],'time': '{} {} {}'.format(result[5], result[6], result[7])})# 下拉底部self.browser.execute_script('window.scrollTo(0, document.body.scrollHeight)')# 下一页time.sleep(3)print('正在爬取第{}页'.format(str(i)))next_page = self.wait.until(EC.presence_of_element_located((By.XPATH, '//*[@class="nui-btn-text"]')))next_page.click()return messagesexcept NoSuchElementException as e:print('Error acquiring information!')return Nonedef connect_db(self):"""创建数据库连接对象:return: db"""db = pymysql.connect(host=MYSQL_LOCALHOST, user=self.user, password=self.pass_,port=MYSQL_PORT, db=MYSQL_DB)# print('Successful connection!')return dbdef save_db(self, messages):"""数据入库:messages: 邮件信息:return: None"""db = self.connect_db()# 获取数据库操作指针cursor = db.cursor()# 数据库表格table = '163'for message in messages:keys = ', '.join(message.keys)values = ', '.join(['%s'] * len(message))sql = 'INSERT INTO {table}({keys}) VALUES ({values})'.format(table = table, keys=keys, values=values)try:if cursor.execute(sql, tuple(message.values())):# print('Successful')mit()except:# print('Failed')db.rollback()db.close()'''def main(self):"""主函数:return: None"""# 登录邮箱self.login_email()# 发送邮件self.send_email()if __name__ == '__main__':mail = Mailbox()mail.main()
