Relationship between eig(A,B) and eig(A)

Lets say we have two matrices

A=[1 2 3;4 5 6;7 8 9];

B=[1 0 0;0 3 0;0 0 0];

It should be noted that matrix B will always be singular.

Now i want to perform the generalized eigen value analysis i.e Av=Bvd

i.e [EigenVectors,EigenValues]=eig(A,B)

Now what i want is that there should be single matrix from matrix A and B lets say matrix D so that [EigenVectos,EigenValues]=eig(D) must be

equal to [EigenVectors,EigenValues]=eig(A,B).

Now the issue is how this D matrix will be calculated from matrix A and B.

Thank you in advance