Can a UFL statement be evaluated at each iteration?
I'm trying to implement a time-dependent DG scheme that looks similar to here. The following bit of code is causing a bit of trouble for me, I think, because it appears that with each iteration, the expression for g gets longer and longer. Is it possible to condense the expression for g0 during the "g = g0" step where this expression would be evaluated rather than adding all of the UFL expressions together with each iteration? u0 is the solution from a previous iteration, if that helps. Thanks!
g0 = 1.0 for t in time: g = exp(1e-12*jump(u0)) + g0 ... g0 = g
I was questioning the sense of the statement g=g0 ;)