BT is faced with a considerable estate of legacy software, either through its own creation or acquisition, which contains many valuable algorithms and components. Successful refactoring of these legacy systems to for example microservice architectures can be very challenging as it critical the new system functionally behaves the same as the original. A detailed test-harness for the original system generally provides the most insight, however most legacy code within BT lack a comprehensive test harness.
This project will focus on generating unit tests for legacy systems using functional specifications and historical data. This will allow BT developers to quickly fit legacy systems with a test-harness that can guide refactoring and guard against undesired regressions. This will also provide developers with significantly faster insight into the legacy system as well as improve refactoring time.
The PhD candidate will conduct at least one placement at BT (Adastral Park) in order to gain further insight into legacy code bases at BT and to explore their software architecture and historical data.
Carlos Diego Nascimento Damasceno, Mohammad Reza Mousavi, Adenilso da Silva Simão:
Learning to Reuse: Adaptive Model Learning for Evolving Systems. IFM 2019: 138-156
Sina Shamshiri, José Miguel Rojas, Juan Pablo Galeotti, Neil Walkinshaw, Gordon Fraser:
How Do Automatically Generated Unit Tests Influence Software Maintenance? ICST 2018: 250-261
Sina Shamshiri, José Miguel Rojas, Luca Gazzola, Gordon Fraser, Phil McMinn, Leonardo Mariani, Andrea Arcuri: Random or evolutionary search for object-oriented test suite generation? Softw. Test. Verification Reliab. 28(4) (2018)
José Miguel Rojas, Gordon Fraser, Andrea Arcuri: Automated unit test generation during software development: a controlled experiment and think-aloud observations. ISSTA 2015: 338-349
Sina Shamshiri, René Just, José Miguel Rojas, Gordon Fraser, Phil McMinn, Andrea Arcuri:
Do Automatically Generated Unit Tests Find Real Faults? An Empirical Study of Effectiveness and Challenges (T). ASE 2015: 201-211