一、缘 起
笔者之前一直在Windows环境下编写UI自动化测试脚本，近日在看《京东系统质量保障技术实战》一书中，萌生出在jenkins下构建UI自动化测试的想法
二、思 路
首先，在Linux环境搭建jenkins平台
然后，将脚本上传至Linux服务器
最后，在jenkins平台构建自动化任务
三、说 明
- Linux环境，Centos7 64位 虚拟机
- JDK1.8 + jenkins 2.121.2
- Python3.6 + Selenium 2.5.3 + Phantomjs 2.1.1
- 注意：
1. Centos7下默认的Python版本为2.7.5，安装Python3的方法：Centos7：5分钟安装python3.5 并存python2.7，兼容yum及Gnome
2. Centos7下默认不带pip命令，安装Selenium需要用到pip命令，安装方法：Centos7：3分钟安装pip3
3. Linux服务器通常为无界面状态，安装浏览器跑UI自动化脚本是不太可能的，解决方法就是用无头浏览器（phantomjs）来代替，何为无头浏览器请读者自行百度，值得注意的是Selenium3.0之后不再支持Phantomjs，故为了在不安装浏览器情况下运行UI自动化脚本，就得安装Selenium3.0之前的版本
selenium与phantomjs分离
四、搭 建
- 搭建jenkins平台，详细教程请自行百度，主要步骤如下
# 下载jenkins.war包
wget http://mirrors.jenkins.io/war-stable/latest/jenkins.war
# 直接用java起war包即可
(nohup)java -jar jenkins.war (port=8090)
# nohup 后台运行，port为指定jenkins端口，默认为8080
# 运行起来后，在浏览器输入:http://localhost:808/即可访问jenkins平台
jenkins平台
- Selenium + Phantomjs
# 安装selenium
pip install selenium==2.53.6
# 检查
pip show selenium
>> Name: selenium
>> Version: 2.53.6
>> Summary: Python bindings for Selenium
>> Home-page: https://github.com/SeleniumHQ/selenium/
# 下载phantomjs
mkdir /usr/local/phantomjs/
wget https://bitbucket.org/ariya/phantomjs/downloads/phantomjs-2.1.1-linux-x86_64.tar.bz2
# 解压
tar -jxvf phantomjs-2.1.1-linux-x86_64.tar.bz2
# 链接
ln -s /usr/local/phantomjs/phantomjs-2.1.1-linux-x86_64/bin/phantomjs /user/bin/phantosjs
# 检查
phantomjs -v
>> 2.1.1
- 脚本说明
- testBaiduProj
testBaiduProj 脚本结构 - baidu.py： 打开百度，输入Python搜索，并截图
#!/usr/bin/python
# coding=utf-8
from selenium import webdriver
from time import sleep
import time, os
cur_path = os.path.dirname(__file__)
url = 'http://www.baidu.com'
driver = webdriver.PhantomJS(service_args=['--ignore-ssl-errors=true', '--ssl-protocol=TLSv1'])
# driver = webdriver.Firefox()
driver.get(url)
driver.find_element_by_name("wd").send_keys("Python")
driver.find_element_by_xpath("//*[@id='su']").submit()
sleep(2)
assert driver.title == 'Python_百度搜索', 'title错误'
cur_time = time.strftime('%Y-%m-%-d %H_%M_%S')
fp = cur_path + '/pic/'
driver.get_screenshot_as_file(r'./pic/' + cur_time + ' image.png')
sleep(1)
driver.quit()
print('完成！')
- sendAttEmail.py： 查找最新的一个png截图，自动发邮件
#!/usr/bin/python
#coding=utf-8
import smtplib
from email.mime.text import MIMEText
from email.mime.multipart import MIMEMultipart
from email.header import Header
import datetime
import os
import time
def send_att_mail(username, passwd, recv, title, content, fp, mail_host='mail.xxx.com'):
"""
自动发邮件
:param username:
:param passwd:
:param recv:
:param title:
:param content:
:parm fp:
:param mail_host:
:return:
"""
msg = MIMEMultipart()
msg['Subject'] = Header(title, 'utf-8')
msg['From'] = username + "@xxx.com"
msg['To'] = ';'.join(recv)
# msg['Cc'] = ','.join(cc_recv) # 抄送
# 邮件正文
msg.attach(MIMEText(content, 'html', 'utf-8')) # 文本'plain', HTML为'html'
# 构造附件
att = MIMEText(open(fp, 'rb').read(), 'base64', 'utf-8')
att['Content-Type'] = 'application/octet-stream'
att.add_header('Content-Disposition', 'attachment', filename=('%s' % fp.split('/')[-1]))
# encoders.encode_base64(att)
msg.attach(att)
smtp = smtplib.SMTP()
smtp.connect(mail_host)
smtp.starttls()
smtp.login(username, passwd)
smtp.sendmail(username + "@xxx.com", recv , msg.as_string())
smtp.quit()
# 邮件内容（文本/HTML）
content = """
<p>
<font size="4" face="SimSun">
<p>你好，请查收邮件~</p>
</font>
</p>
""".format(start, end, tomorrow)
# 邮件标题
title = '百度测试报告'
email_user = '账号' # 发送者账号
email_pwd = '密码' # 发送者密码
# 接收者邮件列表
recv_lst = ['xx@xxx.com']
len(cc_recv_lst)))
time.sleep(1)
# 将pic文件夹里面的png文件名称生成list，对列表排序，选择最后一个即为最新一个png截图
png_lst = os.listdir('./pic')
png_lst.sort()
filename = './pic/' + png_lst[-1]
send_att_mail(email_user, email_pwd, recv_lst, title, content, filename)
print('邮件发送成功！')
time.sleep(1)
五、构 建 任 务
- 构建自由风格项目
- 填写项目名和说明，其他均为默认选项
- 构建部分选择 Execute shell，具体脚本如下
#!/usr/bin/bash
cd /home/lid/pyproj/testBaiduProj/
python baidu.py
sleep 1
python sendAttEmail.py
- 保存项目，并构建
构建部分shell脚本
保存项目，构建中
- 初步OK，然而 ~ ~ ~
六、问 题
- 当phantomjs情况下，用.click()方法报错，但是用.submit()方法却正常
- 当phantomjs情况下，find_element_by_id(‘xx’).send_keys(‘xx’)，报同样错误，但是find_element_by_name(‘xx’).send_keys(‘xx’)，却不报错
Traceback (most recent call last):
File "baidu.py", line 17, in <module>
driver.find_element_by_id("su").click()
File "/usr/local/python3/lib/python3.6/site-packages/selenium/webdriver/remote/webelement.py", line 72, in click
self._execute(Command.CLICK_ELEMENT)
File "/usr/local/python3/lib/python3.6/site-packages/selenium/webdriver/remote/webelement.py", line 461, in _execute
return self._parent.execute(command, params)
File "/usr/local/python3/lib/python3.6/site-packages/selenium/webdriver/remote/webdriver.py", line 236, in execute
self.error_handler.check_response(response)
File "/usr/local/python3/lib/python3.6/site-packages/selenium/webdriver/remote/errorhandler.py", line 192, in check_response
raise exception_class(message, screen, stacktrace)
selenium.common.exceptions.ElementNotVisibleException: Message: {"errorMessage":"Element is not currently visible and may not be manipulated","request":{"headers":{"Accept":"application/json","Accept-Encoding":"identity","Connection":"close","Content-Length":"81","Content-Type":"application/json;charset=UTF-8","Host":"127.0.0.1:36212","User-Agent":"Python-urllib/3.6"},"httpVersion":"1.1","method":"POST","post":"{\"id\": \":wdc:1532881699579\", \"sessionId\": \"670f3e90-934c-11e8-accd-f1bcf1bf551c\"}","url":"/click","urlParsed":{"anchor":"","query":"","file":"click","directory":"/","path":"/click","relative":"/click","port":"","host":"","password":"","user":"","userInfo":"","authority":"","protocol":"","source":"/click","queryKey":{},"chunks":["click"]},"urlOriginal":"/session/670f3e90-934c-11e8-accd-f1bcf1bf551c/element/:wdc:1532881699579/click"}}
Screenshot: available via screen
七、总 结
Jenkins下构建UI自动化测试，看似简单，实则里面无数坑，笔者遇到的如：selenium3.0以上不支持phantomjs问题、Linux环境下安装Python3但是输出中文报错问题、在Linux环境下安装pip3命令无效问题、自动发邮件/添加附件问题，等等，只有你亲自尝试一遍，才能体会所有的不起眼都不容被忽视~
~
~
~
不积跬步，无以至千里
