[Gretl-devel] Re: The 2021c release candidate is available (with apologies for cross-posting)

On Thu, 19 Aug 2021, Sven Schreiber wrote: It should be enough to test on 64-bit Windows. Well, it would be nice to know if the ("secret") lpsolve function works. Here's an example (Kantorovich's "Problem A"): <hansl> clear set verbose off n = 8 # number of variables m = 6 # number of constraints # objective matrix Obj = zeros(1, n) Obj[1] = 1 cnameset(Obj, "A m1A m2A m3A B m1B m2B m3B") matrix R = zeros(m, n+1) # A - B = 0 R[1,] = {1,0,0,0, -1,0,0,0, 0} # total output of A R[2,] = {1,-10,-20,-30, 0,0,0,0, 0} # total output of B R[3,] = {0,0,0,0, 1,-20,-30,-80, 0} # m1 = m1A + m1B <= 3 R[4,] = {0,1,0,0, 0,1,0,0, 3} # m2 = m2A + m2B <= 3 R[5,] = {0,0,1,0, 0,0,1,0, 3} # m3 = m3A + m3B <= 1 R[6,] = {0,0,0,1, 0,0,0,1, 1} strings S = defarray("=", "=", "=", "<=", "<=", "<=") b = _(objective=Obj, constraints=R, ctypes=S, intvars={1,5}) ret = lpsolve(b) print ret.variable_values </hansl> Running it should produce: <output> A 86 m1A 2.6 m2A 3 m3A 0 B 86 m1B 0.4 m2B 0 m3B 0.975 </output> Allin