Jump to content

Rational Functional Tester

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Bbryson (talk | contribs) at 19:10, 25 November 2009 (moved User:Bbryson/Rational Functional Tester to IBM Rational Functional Tester). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

IBM Rational Functional Tester
Developer(s)IBM / Rational Software
Stable release
8.1 / Fall 2009
Operating systemMicrosoft Windows Linux
Typetest automation tools
LicenseProprietary
WebsiteIBM Rational Functional Tester

IBM Rational Functional Tester is a tool for automated testing of software applications from the Rational Software division of IBM. It allows users to create tests that mimic the actions and assessments of a human tester.[1]. It is primarily used by Software Quality Assurance teams to perform automated regression testing.


Overview

IBM Rational Functional Tester is a software test automation tool used by quality assurance teams to perform automated regression testing. Testers create scripts by using a test recorder which captures a user's actions against their application under test. The recording mechanism creates a test script from the actions. The test script is produced as either a Java or Visual Basic.net application, and with the release of version 8.1, is represented as a series of screen shots that form a visual storyboard. Testers can edit the script using standard commands and syntax of these languages, or by acting against the screen shots in the storyboard.. Test scripts can then be executed by Rational Functional Tester to validate application functionality. Typically, test scripts are run in a batch mode where several scripts are grouped together and run unattended.

During the recording phase, the user must introduce verification points. Verification points capture an expected system state, such as a specific value in a field, or a given property of an object, such as enabled or disabled. During playback, any discrepancies between the baseline captured during recording and the actual result achieved during playback are noted in the Rational Functional Tester log. The tester can then review the log to determine if an actual software bug was discovered.


Key Technologies

Storyboard Testing

Introduced in version 8.1 of Rational Functional Tester, this technology enables testers to edit test scripts by acting against screen shots of the application. Prior to this feature, testers edited their scripts by modifying the test script code. Enabling test editing against screenshots significantly reduces test script editing time, and greatly simplifies the process of understanding maintaining automated tests.

Object Map

The Rational Functional Tester Object Map is the underlying technology used by Rational Functional Tester to find and act against the objects within an application. The Object Map is automatically created by the test recorder when tests are created and contains a list of properties used to identify objects during playback.

ScriptAssure

During playback, Rational Functional Tester uses the Object Map to find and act against the application interface. However, during development it is often the case that objects change between the time a script was recorded and when a script was executed. ScriptAssure technology enables Rational Functional Tester to ignore discrepancies between object definitions captured during recording and playback to ensure that test script execution runs uninterrupted. ScriptAssure sensitivity, which determines how big an object map discrepancy is acceptable, is set by the user.

Data Driven Testing

It is common for a single functional regression test to be executed multiple times with different data. To facilitate this, the test recorder can automatically parametrize data entry values, and store the data in a spreadsheet like data pool. This enables tester to add additional test data cases to the test data pool without having to modify any test code. This strategy increases test coverage and the value of a given functional test.

Object Proxy Mechanism

In order for Rational Functional Tester to interact with a given object in an application it must be able to understand that object's interfaces. This is typically not a problem for standard objects such as edit fields, buttons and listboxes. However, in some cases, application developers create their own custom objects to meet a certain application requirement. In those cases, developers can create an object proxy where they can identify in code the interfaces to a custom object. This mechanism enables Rational Functional Tester to be able to interact with objects with which it would otherwise not be able to communicate.


Release History

The following is a release history of the product initially released as RobotJ, and today known as IBM Rational Functional Tester

  • RobotJ 1.0 Released Summer 2002. Initial release. Part of Rational Test Studio 2002.05.02 release.
  • XDE Tester 1.0 Relased in Summer, 2003. Renamed as part of Ratioanl XDE (eXtended Development Environment), in line with Rational Test Studio version 2003.06.00.
  • IBM® Rational Functional Tester for Java and Web. Released 2004. Renamed after IBM acquisition of Rational Software.
  • v6.0 Released September, 2005. Renamed to IBM Rational Functional Tester. Internally known as Atlantic Release.
  • v6.1 Released January, 2005. Renamed to IBM Rational Functional Tester. Internally known as Atlantic Release.
  • v6.1.1 Released March, 2005. Internally known as Atlantic Release.
  • v6.1.2 Released December, 2005.
  • v7.0 Released December, 2006. Internally known as Caspian Release.
  • v7.0.1 Released December, 2007.
  • v7.0.2 Released April, 2008.
  • v8.0 Released October, 2008.
  • v8.1 Released July, 2009.

Criticisms

Test automation tools such as Rational Functional Tester can only find software defects they have been programmed to find. Test test scripts merely mimic human interactions, and only note discrepancies to expected baseline data. If no baseline data is captured, no validation is done and no discrepancy, should there by one, is noted. Test automation is also criticized for the effort required to maintain an automated test. Which technologies like ScriptAssure significantly reduce the burden, test scripts typically need to be validated against new versions of an application under test to ensure that any defects reported are as a result of application problems and not problems with the underlying test script.

References

  1. ^ Software Test Engineering with IBM Rational Functional Tester. The Definitive Resource