Browse Prior Art Database

Method and Apparatus for Cognitive Test Advisor in Application DevOps

IP.com Disclosure Number: IPCOM000248235D
Publication Date: 2016-Nov-10
Document File: 2 page(s) / 40K

Publishing Venue

The IP.com Prior Art Database

Abstract

As a best practice, nowadays many development team build DevOps process to continuous test the new changes to project code. Any test failures will block the new change to be merged in master. The process can help the quality of master branch and aovid bad changes merged into master. But it also reduced efficiency. Many test failures are caused by other issues not related to the change submitted. The developer need take time to identify problem and find the root cause to fix the problem, even the failure is not caused by his change. In this disclosoure, we invent a system of cognitive test adivosr to give much more adisable test report in test failures by continous learning in testing logs and coverage report, it can improve the devops process efficiency greately.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 57% of the total text.

Page 01 of 2

Method and Apparatus for Cognitive Test Advisor in Application DevOps

Keep a quality and deployable master branch in software repository is very important in collaborative software development,, most development project have DevOps process to ensure it. A new feature or bug fix is recommended to developed in a private branch first in the developer local environment. After the development and test completed in local, the developer can submit the change for review and merge into the master. One or moreautomatic testing (Unit test, integration test, acceptance test etc) will run for the code quality validation, ensure the change will not break the master. Any testing failures will block the process, the developer need review and test results, amend his change and re-submit the change for review & merge again.

The process can help the quality of master branch and avoid bad changes merged into master. But it may means reduced efficiency. Many test failures are caused by others(e.g. know issues), not related to the change submitted. Some test are flaky. As the test failures will block the code merge process, the developer need to take time to investigate the failures, identify the root cause of the failure and fix the problem. Even the failure is not caused by his change, he also need to figure out the problem and prove the failure are not related to his change. It may take huge efforts especiallythe developer is not familiar to the whole system. It highly impact the...