Data-driven testing (DDT) is a very important aspect of test automation. In short, the goal is to run a given test or set of tests multiple times with different sets of input data and expected results. The input sets typically contain values which correspond to boundary or partition values.
QF-Test has various means to store data or load external data for use in data-driven tests.
- Database
- Excel
- CSV
- RESTful WebServices
- Property Files
- Language Packs
- Anything via JavaScript, Groovy or Jython
The big number of Test-Case iterations makes DDT cheap for automation with QF‑Test but expensive for manual testing, see test automation and ROI.
Data-driven Testing with QF‑Test – Manual
Datadriven Testing in use at GEBIT Solutions GmbH
A key aspect of agile software development is the demand for rapid feedback. It is not for nothing that explorative test approaches are gaining in importance. But automated tests can also help to get timely feedback regarding quality and not only in the form of regression tests.
This presentation by Ralf Somplatzki (GEBIT Solutions GmbH) “Wenn das Daten Chamäleon der User Story auf den Zahn fühlt”, which he gave at ASQF Net Week, uses a concrete example to show how data-driven end-to-end testing with QF-Test can be an alternative to the often linear structure of test suites.
The challenge is a suitable, modular cut of individual test cases that can be run on the basis of structured test data. Through data-driven automation, the test object can be examined in a variant-rich manner without redundant automation units. At the same time, the actual test suites remain small and maintainable. Furthermore, they can be adapted within a sprint without great effort. This means that comprehensive test variants can be used to obtain reliable quality feedback even for modified program parts.
