Data and tools of the paper "Demystifying and Assessing Code Understandability in Java Decompilation"

The dataset accompanying the paper "Demystifying and Assessing Code Understandability in Java Decompilation" is structured to facilitate a comprehensive analysis of code understandability in Java decompilation. The data is organized into two main directories: data/original and data/testset, representing the original dataset and the test set, respectively. Each directory encompasses three components. First, the code directory contains experimental data, including source code and corresponding decompiled code produced by three decompilers: CFR, Fernflower, and Jadx. Second, the results directory holds the outcomes of various metrics calculations pertinent to the study. Lastly, the annotated dataset files provide a comparative analysis of the relative understandability of decompiled files against the original source files. This dataset is integral for evaluating and comparing the effectiveness of different metrics in assessing the code understandability in the context of decompilation.