Browse Prior Art Database

Method of computing software build sanity, percentage completion, attempted using a baseline for test results

IP.com Disclosure Number: IPCOM000167500D
Original Publication Date: 2008-Feb-16
Included in the Prior Art Database: 2008-Feb-16

Publishing Venue

IBM

Abstract

This invention is to make a claim on the process of performing software build verification test, calculating the quality of test results using a baseline and representing those results for analysis on a website.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 18% of the total text.

Page 1 of 12

Method of computing software build sanity , percentage completion, attempted using a baseline for test results

------------------------------------------------------------------------
1. Method of computing software build sanity, percentage completion, attempted using a baseline for test results ------------------------------------------------------------------------
ABSTRACT

    This invention makes a claim on the process of performing software build verification test, calculating the quality of test results using a baseline and representing those results for analysis on a website.

DRAWINGS
1. File System

1

[This page contains 1 picture or other non-text object]

Page 2 of 12

2. Feature to rpt mapping

3. TSTAT summary report

2

[This page contains 2 pictures or other non-text objects]

Page 3 of 12

4. TSTAT detail report

BACKGROUND OF THE INVENTION

    In a typical software development life cycle, there are number of steps that the product has to undergo before it is released for public use. One such phase is to test the product for any feature regression introduced during the development phase. While the software is developed in iterative cycle, periodic integration and build validation verify that newly added features did not break the existing features of the product. This assessment process is labor intensive as the product needs to be validated across several supported platforms. In a certain configuration, combinations of platform specific tests have to be performed to certify the product. There is no predefined and automated process that exists in the industry to fill this gap. This invention is an infrastructure and a process to automatically certify a software build and the quality of the product.

SUMMARY OF THE INVENTION

    The invention uses a baseline database to represent number of test scenarios, group scenarios into buckets and feature groups. The baseline database has count of test cases in each grouping, and when verified on several test platforms, compares the results with the baseline to compute the product quality. Based on the results, we can tell how many test scenarios were attempted, percentage passed on attempt and percentage passed on total number of scenarios in the system. Similar computations are done on multiple platforms.

3

[This page contains 1 picture or other non-text object]

Page 4 of 12

BRIEF DESCRIPTION OF THE DRAWINGS
1. File System

This diagram shows how the data is organized hierarchal in the file system which constitutes our test results repository. Each folder in the repository uses name derived from build name, platform type, hostname, date and time of the test execution. This offers a clean namespace for analyzing and archiving test results on a common repository. Repository features easy and reliable way to manage organic growth of data, supports implementation of data retention algo...