I'm trying to procedurally generate replacement rules of the following form
X[{a,a}] -> X1
X[{a,b}]X[{b,a}] -> X2
X[{a,b}]X[{b,c}]X[{c,a}] -> X3
X[{a,b}]X[{b,c}]X[{c,d}]X[{d,a}] -> X4
Also, I know the number of maximum required replacement rules in advance.
Implementing {a1___, a2___, a3___, ... } instead of {a,b,c, ... }, my pseudocode reads
X[{a[1],a[2]}] X[{a[2],a[3]}]... X[{a[n-1],a[n]}] X[{a[n],a[1]}] -> Xn
Product[ X[{a[i],a[i+1]}], {i,1,n-1} ] X[{a[n],a[1]}] -> Xn
which translated into actual Mathematica code gives:
MyRule[n_] :=
a___ Product[
Subscript[X, {Symbol["μ"<>ToString[i]<>"___"], Symbol["μ"<>ToString[i+1]<>"___"]}],
{i,1,n-1}] Subscript[X,
{Symbol["μ"<>ToString[n]<>"___"], Symbol["μ"<>ToString[1]<>"__"]}
] :> a Subscript[X, n]
However,
Subscript[X, {a, b}] Subscript[X, {b, a}] /. MyRule[2]
shows that the rule definition is not working properly, allegedly because of a conflict in the way the dummy indices are written and some issues with their 'Symbol' character but I don't really get it. how could I fix this?
Subscript[X, {b, a}]
is not the same asX[{b, a}]
. You just have to decide on a single way of indexing. $\endgroup$