Building test cases and oracles to automate the testing of web database applications

Many organizations rely on web applications that use back-end databases to store important data. Testing such applications requires significant effort. Manual testing alone is often impractical, so testers also rely on automated testing techniques. However, current automated testing techniques may p...

Full description

Saved in:
Bibliographic Details
Published in:Information and software technology 2009-02, Vol.51 (2), p.460-477
Main Authors: Ran, Lihua, Dyreson, Curtis, Andrews, Anneliese, Bryce, Renée, Mallery, Christopher
Format: Article
Language:English
Subjects:
Automated
Automation
Database modification
Functional testing
Guards
Mathematical models
Prototypes
Queries
Software testing
Specifications
Stores
Studies
Web applications
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Many organizations rely on web applications that use back-end databases to store important data. Testing such applications requires significant effort. Manual testing alone is often impractical, so testers also rely on automated testing techniques. However, current automated testing techniques may produce false positives (or false negatives) even in a perfectly working system because the outcome of a test case depends on the state of the database which changes over time as data is inserted and deleted. The Automatic Database Tester (AutoDBT) generates functional test cases that account for database updates. AutoDBT takes as input a model of the application and a set of testing criteria. The model consists of a state transition diagram that shows how users navigate pages, a data specification that captures how data flows, and an update specification that shows how the database is updated. AutoDBT generates guard queries to determine whether the database is in a state conducive to performing and evaluating tests. AutoDBT also generates partial oracles to help validate whether a back-end database is updated correctly during testing. This paper describes the design of AutoDBT, a prototype implementation, several experiments with the prototype, and four case studies.
ISSN:0950-5849
1873-6025
DOI:10.1016/j.infsof.2008.05.016