To study the effects of using a different source language and to determine the performance behavior for larger input
files, we performed a second benchmark using the mobl language [17], shown in Figure 7.
In this benchmark, we generated artificial input files ranging between 0 and 16.000 lines of code.
Syntactically, the mobl language is slightly more sophisticated, as it has a total of 704 production rules versus
the 507 of the Java grammar. Still, the results are similar: parse times linearly increase with larger files, and become
prohibitively long for responsiveness somewhere between 2000 and 4000 lines of code.