There are several examples where the pest parser is extremely slow on mCRL2 files

[mlaveaux]

It is not yet clear why this happens, but I suspect it might be an issue with the grammar.

[mlaveaux]

It seems that it does terminate, but the parsing time increased (seemingly exponentially) with the number of alternative compositions used in the process declaration. It is not yet clear if this is a Pest parser limitation or whether structuring the grammar different can resolve this. According to wikipedia PEG parsing with packrat takes linear time worst-case, and more when supporting (indirect) left recursive rules, but this does not seem to hold for Pest.

[mlaveaux]

Pest does not use memoization or anything like a packrat parser, so the exponential behaviour is to be expected.