Publications

Details

Adrian Lienhard, Tudor Gîrba, Orla Greevy, and Oscar Nierstrasz. Exposing Side Effects in Execution Traces. In Andy Zaidman, Abdelwahab Hamou-Lhadj, and Orla Greevy (Ed.), Proceedings of the 3rd International Workshop on Program Comprehension through Dynamic Analysis (PCODA'07), p. 11—17, Technische Universiteit Delft, 2007. PDF PDF →
Abstract
We need to understand the impact of side effects whenever changing complex object-oriented software systems. This can be difficult as side effects are at best implicit in static views of the software, and typically execution traces do not capture data flow between parts of the system. To solve this problem, we complement execution traces with dynamic object flow information. In our previous work we analyzed object flows between features and classes. In this paper, we use object flow information to analyze side effects in execution traces and to detect how future behavior in the trace is affected by it. Using a visualization, the developer can study how a selected part of the program accessed program state and what side effect its execution produced. Like this, the developer can investigate how a particular part of the program works without needing to understand the source code in detail. To illustrate our approach, we use a running example of writing unit tests for a legacy system.

You are here:

Most of my early papers are related to my PhD thesis and deal with evolution analysis.

The essence of my PhD can be found in a journal article published in JSME.

Since 2005 I got involved in various projects and I co-authored publications on topics mostly in the reengineering area.

We reported on the Story of the Moose and the role of Smalltalk and meta-modeling in its development.

Other works relate to semantic analysis or dynamic analysis. A particular topic that continues to capture my attention is identifying how teams work. Thus, I worked on correlating authorship information with code duplication or feature development

I also got involved in dynamic analysis. I worked on identifying features and their evolution, and I participated in the development of object-flow analysis, a novel dynamic analysis that led to a publication that received the distinguished paper award at ECOOP 2008.

Visualization and interaction are crosscutting concerns in my research. The experience resulted in the development of Mondrian, and of Glamour.