This paper seems to point out that Bouziane’s solution is incorrect.
In the conclusion:
This false claim served for deriving a bound on the number of states of the constructed automaton – which in reality can have infinitely many states.
I think I understand this point as it relates to the general reachibility problem, However The Petri-Net used as an illustration contains 'loops' at transitions: t3 and t1 - transitions that have both incoming and outgoing arcs.
If Petri-Net being examined were composed of only elementary paths, the resulting witness sequence / state machine language would not be susceptible to the pumping lemma.
Does this seem accurate? why or why not?