TestEngine + Jenkins Plugin: Execute list of Project/Suite/Case SoapUI tests
In their current implementation 1.) the TestEngine needs to be provided a zip with all required files by the calling client 2.) the TestEngine Jenkins plugin will execute one Project/Suite/Case per Jenkins job Problem: This setup does not scale to execute many Project/Suite/Case SoapUI tests in a row Suggestion: ad 1.) a.) Provide TestEngine the ability to access/checkout GIT repos containing SoapUI Projects b.) Add an API to TestEngine that will take the Project/Suite/Case/... parameters and pick all required files from above local checkout area --> Like this no files need to be created/transfered to the TestEngine server What still needs to be done is the specification of the list of Project/Suite/Case to be executed. This could be provided by one or more config files maintained in the GIT repo(s) ad 2.) In case the TestEngine runs on the Jenkins slave(s), the Jenkins job could - checkout the GIT repo(s) - locate the config file(s) containing the list of /Project/Suite/Case SoapUI Tests in the checkout area (e.g. via parameter) - loop the config file(s) calling the TestEngine API with Project/Suite/Case/... parameters --> like this - The TestEngine is running as a server (contrary to testruner.sh) - The invocation of many Project/Suite/Case's will run with minimal overhead (presumably msecs instead of secs) - all required files are taken from a repo -> any updates propagate to TestEngine automatically Remark: This type of automated execution of SoapUI Test from different Project/Suite/Case needs to complemented with a suitable test report (e.g. csv file with Project/Suite/Case/Step, timeStamp, Status, ExecTime, ErrorMsg, ...
ReadyAPI and TestEngine with Zephyr Scale
In this post, we are going to talk about using some of the SmartBear API testing tools and writing the results in an automated fashion to Zephyr Scale. But first let's take a step back and discuss a general approach of using testing tools with test management tools. At a manual level, when creating simple Multi-Protocol API tests, we might use a lighter weight tool like SoapUI Open Source, a great tool for quick and basic functional and performance API tests. After we trigger the test manually, we as humans would parse the report for any details on the test run, and to track it would update a spreadsheet, testcase in Jira, but somewhere. This sequence is important to note for when we start to advance to an automated workflow. Now let's talk about automating this process. We execute our API tests in an ephemeral environment, publish the results to a relative location, and then pass those results from that relative location to a test management platform. It is quite simple. In this workflow it tends to be easier to use a tool that has native integrations into CI platforms like Jenkins and Azure Dev Ops, that make managing our executions in these ephemeral environments much easier. Here, ReadyAPI, or TestEngine is the clear choice as they both seamlessly integrate into any CI-CD system and have native integrations into Jenkins, Azure, and many more. Now that sequence I mentioned earlier is very important. We will execute our API tests, and then pass the results into Zephyr Scale as step 2 in the pipeline. Here is a look at a scripted pipeline approach, note that there is freestyle job options available in Jenkins too. ReadyAPI Jenkins Plugin: https://support.smartbear.com/readyapi/docs/integrations/jenkins.html TestEngine Jenkins Plugin: https://support.smartbear.com/testengine/docs/admin/jenkins.html node { stage('Run API Tests') { // Run the API Tests using ReadyAPI or TestEngine stage('Pass Results') { //Pass Results to Zephyr Scale } } Now we need to be a little more specific with the 'Run API Tests' Stage. Above I mention when we run our API Tests in an automated fashion, we need to write the results to a relative location so that we can then send those results to Zephyr Scale from that location. Both ReadyAPI and TestEngine allow us to write results to locations as part of the command-line or native integrations. I will show command- line options for ReadyAPI and UI native integration for TestEngine but both options are available for both tools. Starting with ReadyAPI, testrunner CLI, the -f and -F flags represent the directory we are writing to, and the report format, respectively. ReadyAPI offers reports in PDF, XLS, HTML, RTF, CSV, TXT and XML, but the automation recommendation would be to pass results in the Junit-XML option. At a basic level we need this: testrunner.bat [optional-arguments] <test-project> And we need to specify -f and -F testrunner.bat -f<directory> -F<args> <test-project> and with -f requiring a relative directory, that can change based on the CI system. I will use Azure Dev Ops for both my examples here. In Azure I pull my test cases from the $(System.DefaultWorkingDirectory), which contains my git repo. In Azure I publish results to the $(Common.TestResultsDirectory) An example full command would look like this: "C:\Program Files\SmartBear\ReadyAPI-3.40.1\bin\testrunner.bat" -r -a -j -f$(Common.TestResultsDirectory) "-RJUnit-Style HTML Report" -FXML "-EDefault environment" "$(System.DefaultWorkingDirectory)/" With TestEngine it's very similar, but I am highlighting it through the native integration, note the publish test results and save Junit report option enabled below: Now lastly, we need to send the results to Zephyr Scale from the pipeline, before our release is over. I think it's easiest with the Zephyr Scale API: https://support.smartbear.com/zephyr-squad-cloud/docs/api/index.html along with the Auto-Create Test Case option to true. The command below is a basic example, and replica of the one seen in the Azure Pipeline screenshot. curl -H "Authorization: Bearer Zephyr-Scale-Token-Here" -F file= Relative-Location-of-Report-Here\report.xml;type=application/xml "https://api.zephyrscale.smartbear.com/v2/automations/executions/junit?projectKey=Project-Key-Here&autoCreateTestCases=true" After you modify the API token, Relative location, and project key you are good to run the pipeline. When we jump to Jira, we can see inside Zephyr Scale that the results are populating. Even with transactional Data on the failed test stepsReadyAPI and TestEngine with Zephyr Scale
In this post, we are going to talk about using some of the SmartBear API testing tools and writing the results in an automated fashion to Zephyr Scale. But first let's take a step back and discuss a general approach of using testing tools with test management tools. At a manual level, when creating simple Multi-Protocol API tests, we might use a lighter weight tool like SoapUI Open Source, a great tool for quick and basic functional and performance API tests. After we trigger the test manually, we as humans would parse the report for any details on the test run, and to track it would update a spreadsheet, testcase in Jira, but somewhere. This sequence is important to note for when we start to advance to an automated workflow. Now let's talk about automating this process. We execute our API tests in an ephemeral environment, publish the results to a relative location, and then pass those results from that relative location to a test management platform. It is quite simple. In this workflow it tends to be easier to use a tool that has native integrations into CI platforms like Jenkins and Azure Dev Ops, that make managing our executions in these ephemeral environments much easier. Here, ReadyAPI, or TestEngine is the clear choice as they both seamlessly integrate into any CI-CD system and have native integrations into Jenkins, Azure, and many more. Now that sequence I mentioned earlier is very important. We will execute our API tests, and then pass the results into Zephyr Scale as step 2 in the pipeline. Here is a look at a scripted pipeline approach, note that there is freestyle job options available in Jenkins too. ReadyAPI Jenkins Plugin: https://support.smartbear.com/readyapi/docs/integrations/jenkins.html TestEngine Jenkins Plugin: https://support.smartbear.com/testengine/docs/admin/jenkins.html node { stage('Run API Tests') { // Run the API Tests using ReadyAPI or TestEngine stage('Pass Results') { //Pass Results to Zephyr Scale } } Now we need to be a little more specific with the 'Run API Tests' Stage. Above I mention when we run our API Tests in an automated fashion, we need to write the results to a relative location so that we can then send those results to Zephyr Scale from that location. Both ReadyAPI and TestEngine allow us to write results to locations as part of the command-line or native integrations. I will show command- line options for ReadyAPI and UI native integration for TestEngine but both options are available for both tools. Starting with ReadyAPI, testrunner CLI, the -f and -F flags represent the directory we are writing to, and the report format, respectively. ReadyAPI offers reports in PDF, XLS, HTML, RTF, CSV, TXT and XML, but the automation recommendation would be to pass results in the Junit-XML option. At a basic level we need this: testrunner.bat [optional-arguments] <test-project> And we need to specify -f and -F testrunner.bat -f<directory> -F<args> <test-project> and with -f requiring a relative directory, that can change based on the CI system. I will use Azure Dev Ops for both my examples here. In Azure I pull my test cases from the $(System.DefaultWorkingDirectory), which contains my git repo. In Azure I publish results to the $(Common.TestResultsDirectory) An example full command would look like this: "C:\Program Files\SmartBear\ReadyAPI-3.40.1\bin\testrunner.bat" -r -a -j -f$(Common.TestResultsDirectory) "-RJUnit-Style HTML Report" -FXML "-EDefault environment" "$(System.DefaultWorkingDirectory)/" With TestEngine it's very similar, but I am highlighting it through the native integration, note the publish test results and save Junit report option enabled below: Now lastly, we need to send the results to Zephyr Scale from the pipeline, before our release is over. I think it's easiest with the Zephyr Scale API: https://support.smartbear.com/zephyr-squad-cloud/docs/api/index.html along with the Auto-Create Test Case option to true. The command below is a basic example, and replica of the one seen in the Azure Pipeline screenshot. curl -H "Authorization: Bearer Zephyr-Scale-Token-Here" -F file= Relative-Location-of-Report-Here\report.xml;type=application/xml "https://api.zephyrscale.smartbear.com/v2/automations/executions/junit?projectKey=Project-Key-Here&autoCreateTestCases=true" After you modify the API token, Relative location, and project key you are good to run the pipeline. When we jump to Jira, we can see inside Zephyr Scale that the results are populating. Even with transactional Data on the failed test steps.
Difference result between TestEngine and ReadyAPI
Hi Hi build my test with ready API, my test is OK, my test contains a groovy scripts however When I execute this test in TestEngine this same test is Failed. I add in this parametre : By default, Groovy scripting has limitations on writing data to properties and files. If you need fully functional Groovy scripts in your tests, you can turn off these restrictions. Turning off the restrictions can cause security vulnerabilities. To turn off the limitations, you need to set the groovy.allow.all system property to true. The way you do this depends on how you run TestEngine: but in my Test on TestEngine I have this error on my grouvy script., why the behaviour is different between TestEngine and ReadyAPI IDE ? Error. java.net.UnknownHostException: https Error. error at line: 20
