Section: New Results

Tools for understanding applications: IDEs and Visualization

Performance Evolution Blueprint: Understanding the Impact of Software Evolution on Performance. Understanding the root of a performance drop or improvement requires analyzing different program executions at a fine grain level. Such an analysis involves dedicated profiling and representation techniques. JProfiler and YourKit, two recognized code profilers fail, on both providing adequate metrics and visual representations, conveying a false sense of the performance variation root. We propose performance evolution blueprint, a visual support to precisely compare multiple software executions. Our blueprint is offered by Rizel, a code profiler to efficiently explore performance of a set of benchmarks against multiple software revisions. [31]

Seamless Composition and Reuse of Customizable User Interfaces with Spec Implementing UIs is often a tedious task. To address this, UI Builders have been proposed to support the description of widgets, their location, and their logic. A missing aspect of UI Builders is however the ability to reuse and compose widget logic. In our experience, this leads to a significant amount of duplication in UI code. To address this issue, we built Spec: a UIBuilder for Pharo with a focus on reuse. With Spec, widget properties are defined declaratively and attached to specific classes known as composable classes. A composable class defines its own widget description as well as the model-widget bridge and widget interaction logic. Spec enables seamless reuse of widgets, its use in Pharo 2.0 has cut in half the amount of lines of code of six of its tools, mostly through reuse. This shows that Spec meets its goals of allowing reuse and composition of widget logic. [17]

Pragmatic Visualizations for Roassal: a Florilegium Software analysis and in particular reverse engineering often involves a large amount of structured data. This data should be presented in a meaningful form so that it can be used to improve software artefacts. The software analysis community has produced numerous visual tools to help understand different software elements. However, most of the visualization techniques, when applied to software elements, produce results that are difficult to interpret and comprehend. We present five graph layouts that are both expressive for polymetric views and agnostic to the visualization engine. These layouts favor spatial space reduction while emphasizing on clarity. Our layouts have been implemented in the Roassal visualization engine and are available under the MIT License. [23]