In the original formulation of intersection homology, the strata need to satisfy $X{n-1} = X{n-2}$. I am not yet checking for this condition to be satisfied. It could depend on the definition of the perversity values that are used in the algorithm.
I just realized that this is not fully fixed yet, as I have so far only checked whether the original indexing of Goresky--MacPherson needs to be used or not.
In the original formulation of intersection homology, the strata need to satisfy $X{n-1} = X{n-2}$. I am not yet checking for this condition to be satisfied. It could depend on the definition of the perversity values that are used in the algorithm.