Extracellular potential is a function of time and spatial location V(x,y,z,t). In your particular case, is this function separable into the product of two terms
V(x,y,z,t) = a(x,y,z)*f(t)
where f describes the basic "shape" of V as a function of time
and a(x,y,z) specifies how V varies with position
If yes, then it will be easiest to implement extracellular stimulation using xtra (or a modification of xtra if you want to avoid having to think about transfer resistance and stimulus current). That will avoid the trivial but tedious details required to read (tens? hundreds? thousands? of) thousands of values that differ only by constant multiplicative factors from files, filling up thousands of vectors with those values, and then using vector play to drive e_extracellular.
Anything that doesn't fit elsewhere.