BTW, if i1...in are the indices of the array element, and sk is the size of dimension k in the array, than the mapping is: a[i1, i2, ... in] =
b[i1 + E{j=2..n}(ij * T{k=1..j-1}(sk))]
The E and T are symbolic of summation and multiplication
summation (I forgot the term, but it's the pi summation as opposed
to the sigma summation), respectively. Damn these non-MathML-supporting
browsers.
Now I need to use this to make my ArrayElementNodes capable of figuring
out where in the actual (one-dimensional) array they reference. Then,
they will become self-aware and take over the world! AH HA HA HA HA!
Or maybe they will just help me get my compiler done.
Mon Jun 05 2000 09:15:
Why does it {hurt when I pee, take me so long to derive the mapping
of an element of an n-dimensional array to an element of a one-dimensional
array}? It's not difficult, yet I must needs fill up a whole page with
poorly drawn diagrams and equations to figure it out.