性能工具之Ngrinder源码部署
作者:互联网
转载:https://cloud.tencent.com/developer/article/1526398

背景
为了更好了解nGrinder怎么工作或者为下次二次开发便开启使用源码部署。
下载源码:
https://github.com/naver/ngrinder/releases

也可以直接通过:https://github.com/naver/ngrinder.git这种方式下部署

本次使用下载zip进行安装:

点击目录再点击:

等待执行成功便把如下安装到本地创库:

打开idea:

点击文件导入:

点击Open as Project:

打开新窗口:

等待maven加载相应的jar。
修改代码

具体代码如下:
package org.ngrinder.perftest.service; import org.ngrinder.infra.config.Config; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.context.ApplicationContext; import org.springframework.context.ApplicationContextAware; import org.springframework.context.annotation.Bean; import org.springframework.context.annotation.Configuration; import org.springframework.context.annotation.EnableAspectJAutoProxy; import org.springframework.context.annotation.Profile; import org.springframework.scheduling.annotation.EnableScheduling; import org.springframework.transaction.annotation.EnableTransactionManagement; /** * Dynamic creation of {@link PerfTestService} depending on the cluster enable or disable. * * @author JunHo Yoon * @since 3.1 */ @Configuration @Profile("production") @EnableScheduling @EnableTransactionManagement @EnableAspectJAutoProxy public class PerfTestServiceConfig implements ApplicationContextAware { @Autowired private Config config; private ApplicationContext applicationContext; /** * Create PerTest service depending on cluster mode. * * @return {@link PerfTestService} */ @Bean(name = "perfTestService") public PerfTestService perfTestService() { if (config.isClustered()) { return applicationContext.getAutowireCapableBeanFactory().createBean(ClusteredPerfTestService.class); } else { return applicationContext.getAutowireCapableBeanFactory().createBean(PerfTestService.class); } // return applicationContext.getAutowireCapableBeanFactory().createBean( // config.isClustered() ? ClusteredPerfTestService.class : PerfTestService.class); } @Override public void setApplicationContext(ApplicationContext applicationContext) { this.applicationContext = applicationContext; } }
再次配置tomcat

选择tomcat所在目录:

选择:

添加运行war:


选择时时更新运行:


注意最好是这样:
-Xms1024m -Xmx1024m -XX:MaxPermSize=200m防止内存出现异常显现

点击确定:

启动:

打开浏览器验证是否成功:
http://localhost:8081/ngrinder/login

登录成功:

使用源码调试简单脚本
在script-sample工程下的pom.xml文件种增加:

代码如下
<!-- https://mvnrepository.com/artifact/junit/junit --> <dependency> <groupId>junit</groupId> <artifactId>junit</artifactId> <version>4.12</version> <scope>test</scope> </dependency>
再次在idea中全局搜索:
groovy-all

查看版本号,统一修改为:
<version>2.4.16</version>
模仿编写脚本
通过平台生成脚本:

点击:

查看脚本
importstatic net.grinder.script.Grinder.grinder importstatic org.junit.Assert.* importstatic org.hamcrest.Matchers.* import net.grinder.plugin.http.HTTPRequest import net.grinder.plugin.http.HTTPPluginControl import net.grinder.script.GTest import net.grinder.script.Grinder import net.grinder.scriptengine.groovy.junit.GrinderRunner import net.grinder.scriptengine.groovy.junit.annotation.BeforeProcess import net.grinder.scriptengine.groovy.junit.annotation.BeforeThread // import static net.grinder.util.GrinderUtils.* // You can use this if you're using nGrinder after 3.2.3 import org.junit.Before import org.junit.BeforeClass import org.junit.Test import org.junit.runner.RunWith import java.util.Date import java.util.List import java.util.ArrayList importHTTPClient.Cookie importHTTPClient.CookieModule importHTTPClient.HTTPResponse importHTTPClient.NVPair /** * A simple example using the HTTP plugin that shows the retrieval of a * single page via HTTP. * * This script is automatically generated by ngrinder. * * @author admin */ @RunWith(GrinderRunner) classTestRunner{ publicstaticGTest test publicstaticHTTPRequest request publicstaticNVPair[] headers = [] publicstaticNVPair[] params= [] publicstaticCookie[] cookies = [] @BeforeProcess publicstaticvoid beforeProcess() { HTTPPluginControl.getConnectionDefaults().timeout = 6000 test = newGTest(1, "www.baidu.com") request = newHTTPRequest() grinder.logger.info("before process."); } @BeforeThread publicvoid beforeThread() { test.record(this, "test") grinder.statistics.delayReports=true; grinder.logger.info("before thread."); } @Before publicvoid before() { request.setHeaders(headers) cookies.each { CookieModule.addCookie(it, HTTPPluginControl.getThreadHTTPClientContext()) } grinder.logger.info("before thread. init headers and cookies"); } @Test publicvoid test(){ HTTPResponse result = request.GET("https://www.baidu.com/", params) if(result.statusCode == 301|| result.statusCode == 302) { grinder.logger.warn("Warning. The response may not be correct. The response code was {}.", result.statusCode); } else{ assertThat(result.statusCode, is(200)); } } }
复制脚本:
在idea中新建脚本脚本

选择groovy脚本

输入名字点击保存即可

新建完毕把刚才脚本复制过来修改下方法名称
如下:

点击运行:

提示:

解决方法:
在Idea菜单栏->Run->Edit Configurations->Default->Junit->在VM options填写自定义配置,点击Apply按钮保存配置即生效

再次点击:

运行结果如下:

本机脚本调试成功:
下次再次分享本地参数化与post请求
送大家一句话:
学习不等于学会!知道不等于做到!唯有严格的反复操练!才是灵活应用的保证!学习向学会走近一步,知道向做到拉近了一些,反复的练习向应用靠近了距离,我们只要走,就会见到别人的微笑。
标签:grinder,部署,springframework,源码,Ngrinder,org,import,net,junit 来源: https://www.cnblogs.com/ceshi2016/p/13845285.html