Hi all,

I am working on creating an environment to write test cases to test
standalone features. With new GIT module components and features should be
released prior to product releases, so all features must be thoroughly
tested and should be production ready. Carbon runtime is needed to test the
standalone features, so our plan is to start carbon-core by installing all
features resided in a project repository.

For creating test execution environment, all features will be installed on
carbon-core using carbon-p2-plugin. Before installing we have to create
p2-repo with relevant features in it. To do that, need to have pom with all
feature dependencies required (including common dependencies) . P2-repo
will be created by installing that pom with relevant features specified.
When installing a feature you have to find out and install all the other
features this feature is depending on.

After creating carbon-core with all required features, test framework
dependency will be used with new extension classes to facilitate sever
management, code coverage generation and user/tenant creation.

I've done a POC to verify feature installation on carbon-core in
carbon-deployment project and everything seems to work fine. Just need
clarify whether this is the correct approach for feature level testing and
any suggestions for alternatives. Kindly expecting your ideas on this.

Attached diagram depict proposed architecture.


​​
​Thanks,
Iqbal

-- 
Irham Iqbal
Software Engineer - Test Automation
 WSO2, Inc.: http://wso2.com
lean. enterprise. middleware
phone: +94 777888452
_______________________________________________
Architecture mailing list
[email protected]
https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture

Reply via email to