This repository contains auxiliary data pertaining to the "Fuzzing With Grammar Variants" master's thesis written by Mark Schuegraf.
A total of 1380 scatter plots depict the coverage progression differences of fuzzing before and after grammar transformation.
Each plot represents 500,000 data points generated by JaCoCo-instrumented test subjects. To be precise, the measurements over the course of 50 runs at each of the 5000 inputs are displayed for each of the two grammar variants.
This incurs considerable overlap. To this end, opacity is used to represent multiple occurrences of the same coverage value across different runs at the same file number. The more transparent a data point is, the less runs produced this coverage value at that stage.
The plots are organized in order of the input language whose grammar model was transformed, followed by the type of transformation that was applied. Next comes the fuzzing strategy used to produce inputs from the grammars of the given language and finally the test subject itself.