This is an open-source maven plugin for interacting with Xray (server/datacenter and cloud), used in CI/CD for assisting in test automation flows, such as reporting test results back to Xray and, by consequence, Jira.

This plugin is a totally new implementation, made from scratch, and should not be mistaken with the proprietary maven plugin originaly provided by Xray. It makes use of the public APIs provided by Xray (see references at bottom of this page), which we recommend for you to have a look as these may have some constraints that also apply to this plugin.

The Xray Maven Plugin is maintained by and in the open-source spirit.

The general flow for importing test automation results is as simple as submiting them back to Xray. For Gherkin-based frameworls (e.g., Cucumber, Behave, SpecFlow), the flow is slightly different and require additional steps (more on these flows here).

This plugin provides these tasks:

• xray:import-results
• xray:import-features
• xray:export-features

More info about these ahead.

This plugin is available on (Maven) Central Repository, which is configured by default in your Maven instalation.

Add the following dependency to your pom.xml, where the <configuration> is optional and should be adapted to your use case.

        <plugin>
            <groupId>app.getxray</groupId>
            <artifactId>xray-maven-plugin</artifactId>
            <version>0.8.0</version>
            <scope>test</scope>
            <configuration>
                <clientId>215FFD69FE46447280000000000</clientId>
                <clientSecret>1c00f8f22f56a8684d7c18cd61470000000</clientSecret>
                <cloud>true</cloud>
                <projectKey>CALC</projectKey>
                <fixVersion>1.0</fixVersion>
                <reportFormat>junit</reportFormat>
                <reportFile>target/junit.xml</reportFile>
            </configuration>
        </plugin>

Configuration parameters (e.g., "reportFormat") can also be specified from the command line using -D (e.g., "-Dxray.reportFormat=junit"). In this case, the parameters have the xray. prefix.

There are a set of common configurations related to the Xray details and its authentication. Besides, each task has its own configuration parameters as shown ahead.

This repository provides a bunch of usage examples, both for Xray cloud and Xray server/datacenter.

For Xray server/datacenter users (i.e., using Xray on Jira server/datacenter):

For Xray cloud users (i.e., using Xray on Jira cloud):

In order to obtain the API key (client id + client secret pair) please ask you Jira admin (see reference at bottom). There are also task specific configurations. More info ahead, on the respective task section.

In order to import test results, we need to use the xray:import-results task.

 mvn clean compile test xray:import-results

Note that in the previous example the test task may fail aborting maven and thus not importing the test results to Xray. To overcome this, we can pass an argument to instruct Surefire to ignore the test failures; this can also be done at pom.xml level.

 mvn clean compile test xray:import-results -Dmaven.test.failure.ignore=true 

The pom.xml needs to be configured properly (see available configurations). As an alternative to hardcode the configurations, it's also possible to pass them right from the command line as mentioned earlier, or even have some on the pom.xml and another specificed through command line parameters.

 mvn clean compile test xray:import-results -Dxray.reportFormat=junit -Dxray.reportFile=results/junit.xml

There are two ways of importing results. The first one is more simple and also known as "standard" (due to how it is called in terms of REST API) where we provide all or a mix of predefined and common parameters (projectKey, version, revision, testPlanKey, testExecKey, testEnvironment) that are enough for most usage scenarios. There is another way importing results though, that allow us to customize any field on the Test Execution issue or even on the Test issues that may be created; in this case, we need to pass the testExecInfoJson and/or the testInfoJson fields. This approach, also known has "multipart" due to the endpoint it uses, even though more flexible will require us to specify common fields such as the project key, version, and other, within the respective JSON field.

In sum, if we decide to use the first approach we can use these parameters: projectKey, version, revision, testPlanKey, testExecKey, testEnvironment. If we decide to use the second approach to either customize the target Test Execution or Test issues, then we can only use the testInfoJson and testExecInfoJson parameters. **For this reason, it's not possible to use the parameters projectKey, version, revision, testPlanKey, testExecKey, testEnvironment together with testInfoJson and testExecInfoJson at the same time.

Xray server/DC and Xray cloud support mostly the same formats, but not exactly for legacy reasons. Besides, not all formats support the same parameters; please check the respective product documentation. The following table sums this info.

This task makes use of different REST APIs for importing test automation results (see reference at the bottom). Related to each report format (e.g., junit, testng, etc) there are usually 2 endpoints for submiting test results: the so called "standard" and the "multipart" one. Whenever importing results, for a given report format, the plugin will use one of these endpoints based on whether testInfoJson or testExecInfoJson are provided; if none of them are provided, the "standard" endpoint, otherwise the "multipart" endpoint will be used.

As detailed before, and not like happens with other formats, whenever importing cucumber/behave results it's not possible to use the parameters projectKey, version, revision, testPlanKey, testExecKey, testEnvironment parameters. This is due to the way the Cucumber endpoints work (as of July 2024). The "standard" Cucumber/Behave REST API endpoints dont' support these parameters; as this plugin is mostly an wrapper around the available REST API endpoints, it also inherits their limitations. However, Xray provides slightly different endpoints, also known as the "multipart" endpoints; these endpoints are more flexible and allow you to customize the fields you wish on the target Test Execution that will be created (yes, using these endpoints will always lead to the creation of new Test Executions).

To make this maven plugin use the underlying multipart endpoint and therefore be able to customize the target Test Execution, we just have to define the testExecInfoJson parameter with a filename containing the customization JSON. We'll show 2 examples: one for Xray Cloud and an equivalent for Xray Datacenter (the syntax is slightly different). So, don't specify any of the parameters projectKey, version, revision, testPlanKey, testExecKey, testEnvironment as they will be simply ignored.

Example follows ahead of testExecInfoJson content for Xray Cloud. In this case we're going to create the Test Execution with the results on the Jira project having the key CALC, on the version v3.0, and associate it with the Test Plan having key CALC-15. The Test Execution will also be associated with the Test Environment named chrome.

{
    "fields": {
        "project": {
            "key": "CALC"
        },
        "summary": "Brand new Test execution",
        "issuetype": {
            "name": "Test Execution"
        },
        "fixVersions" : 
                [
                 {
                        "name": "v3.0"
                 }
                ]

    },
    "xrayFields": {
        "testPlanKey": "CALC-15",
        "environments": ["chrome"]
    }
}

Example follows of testExecInfo content for Xray Datacenter. In this case we're going to create the Test Execution with the results on the Jira project having the key CALC, on the version v3.0, and associate it with the Test Plan having key CALC-15. The Test Execution will also be associated with the Test Environment named chrome. Note that the syntax is different from the one for Xray Cloud; in this case, to be able to set the Test Plan or the Test Environment(s), you'll need to find out the corresponding custom field ids; to do that, check with your Jira administrator (info can be obtained from Issues > Custom fields). Also, a more simple way would be to invoke the REST API to figure it out (e.g., https://yourjiraserver/rest/api/2/field as mentioned here).. and that info is visible on the response. Anwyway, the following example assumes Test Plan custom field has an id "11807" and the Test Environments has an id "11805"; adjust it for your case.

{
    "fields": {
        "project": {
            "key": "CALC"
        },
        "summary": "Brand new Test execution",
        "issuetype": {
            "name": "Test Execution"
        },
        "customfield_11805" : [
            "chrome"
        ],
        "customfield_11807" : [
            "CALC-15"
        ],
        "fixVersions" : 
                [
                 {
                        "name": "v3.0"
                 }
                ]
    }
}

For more information, please read the following Cucumber related documentation (or the corresponding one for Behave):

• Xray Cloud documentation on the Cucumber multipart REST API endpoint
• Xray Datacenter documentation on the Cucumber multipart REST API endpoint

One of the possible workflows for using Gherkin-based frameworks is to use git (or other versioning control system) as the master to store the corresponding .feature files (more info here). In order to provide visibility of test results for these tests (i.e. gherkin Scenarios), these need to exist in Xray beforehand. Therefore, we need to import/synchronize them to Xray. Note that there is no direct integration; the integration is adhoc, i.e., the following task is run on a local copy of the respository where the .features are stored in.

 mvn clean compile test xray:import-features -Dxray.inputFeatures=features/

Note: how Xray relates the Scenarios/Background to the corresponding Test or Precondition issues is described in Xray technical documentation (e.g., Xray cloud docs).

This task makes use of a specific REST API avaiable for importing .feature files (for legacy reasons, also known as "importing Cucumber tests"). The .feature files are zipped and sent in one REST API call made to Xray. Link to the REST API endpoint details at the bottom (both for server/datacenter and cloud).

No matter which worflow you decide to use for dealing with Gherkin-based tests (e.g., Cucumber .feature files and corresponding Scenarios), as part of that workflow comes the need to extract/generate the .feature files based on Scenarios or Backgrounds detailed in Xray using Test or Precondition issues. This plugin provides a task for this purpose. It will download/generate .feature files to a local folder from existing information in Xray. The Feature, Scenario elements will be properly tagged with info from Xray.

 mvn xray:export-features -Dxray.issueKeys=CALC-1,CALC-2 -Dxray.outputDir=features/

Files on the destination folder will be overwritten; however, if this directory contains other information (including other .feature files) you may need to remove them before generating the .feature files into this directory.

Note: how Xray generates the .feature files with the Scenarios/Background from existing Test or Precondition issues is described in Xray technical documentation (e.g., Xray cloud docs).

This task makes use of a specific REST API avaiable for exporting/generating .feature files (for legacy reasons, also known as "exporting Cucumber tests"). Link to the REST API endpoint details at the bottom (both for Xray on Jira server/datacenter and cloud).

1. Is this the same maven plugin as the original one made by Xray team?

No. This is a totally new one, made from scratch and open-source.

2. If we have questions/support issues, where should those be addressed?

It's an open-source project, so it should be handled in this GitHub project and supported by the community. If you want to use the previous, proprietary plugin, you can do so and that has commercial support, if you have a valid license.

3. Are the underlying APIs the same for Xray server/datacenter and Xray Cloud? Are the available options the same? Are the supported test automation report formats the same?

Not exactly. Xray server/datacenter and Xray cloud, even though similar, are actually distinct products; besides Jira server/datacenter and Jira cloud are different between themselves and have different capabilities. This plugin makes use of the available REST APIs for Xray server/datancer and Xray cloud, so you should check them to see exactly what is supported for your environment.

You may find me on Twitter. Any questions related with this code, please raise issues in this GitHub project. Feel free to contribute and submit PR's. For Xray specific questions, please contact Xray's support team.

This project is in early stage; the setting names and other are subject to change.

• Thanks to Eugen, and other authors, along with the many tutorials provided at baeldung.com.
• Thanks to Hugo Braz for all the valuable feedback and previous work done in the scope of the previous, proprietary Maven plugin for Xray.

• review pom.xml and make it cleaner (help neeeded and appreciated!)
• add code coverage information with jacoco, as I couldnt get it working (help needed and appreciated!)
• add unit tests
• option to remove temporary .zip file created whenever importing Gherkin/cucumber .feature files
• ability to attach files/evidence to test runs (needs some thought)

• Importing test results (Xray server/datacenter)
• Importing test results (Xray cloud)
• Import Cucumber tests (Xray server/datacenter)
• Import Cucumber tests (Xray cloud)
• Export Cucumber tests (Xray server/datacenter)
• Export Cucumber tests (Xray cloud)
• Using Personal Access Tokens (Xray datacenter)
• API keys (client id + client secret pair) on Xray cloud

BSD 3-Clause