Slight Refactoring of a few models

realistic/TTP_PV/TTP_PV.py

@@ -39,9 +39,9 @@
 
 def automaton():
     qi, q01, q02, q03, q11, q12, q13 = states = "q", "q01", "q02", "q03", "q11", "q12", "q13"
-    tr2 = [(qi, 0, q01), (qi, 1, q11), (q01, 0, q02), (q01, 1, q11), (q11, 0, q01), (q11, 1, q12), (q02, 1, q11), (q12, 0, q01)]
-    tr3 = [(q02, 0, q03), (q12, 1, q13), (q03, 1, q11), (q13, 0, q01)]
-    return Automaton(start=qi, final={q for q in states if q != qi}, transitions=tr2 + tr3)
+    t2 = [(qi, 0, q01), (qi, 1, q11), (q01, 0, q02), (q01, 1, q11), (q11, 0, q01), (q11, 1, q12), (q02, 1, q11), (q12, 0, q01)]
+    t3 = [(q02, 0, q03), (q12, 1, q13), (q03, 1, q11), (q13, 0, q01)]
+    return Automaton(start=qi, final={q for q in states if q != qi}, transitions=t2 + t3)
 
 
 A = automaton()
@@ -80,31 +80,51 @@ def automaton():
     # computing travelled distances wrt venues of current and next-round games
     [
         [
-            (
-                If(h[i][0] == 1, Then=t[i][0] == 0),
-                If(h[i][0] != 1, Then=t[i][0] == distances[i][o[i][0]])
+
+            If(
+                h[i][0] == 1,
+                Then=t[i][0] == 0,
+                Else=t[i][0] == distances[i][o[i][0]]
             ) for i in range(nTeams)
         ],
 
         [
-            (
-                If(h[i][k] == 1, h[i][k + 1] == 1, Then=t[i][k + 1] == 0),
-                If(h[i][k] != 1, h[i][k + 1] == 1, Then=t[i][k + 1] == distances[o[i][k]][i]),
-                If(h[i][k] == 1, h[i][k + 1] != 1, Then=t[i][k + 1] == distances[i][o[i][k + 1]]),
-                If(h[i][k] != 1, h[i][k + 1] != 1, Then=t[i][k + 1] == distances[o[i][k]][o[i][k + 1]])
+            Match(
+                (h[i][k], h[i][k + 1]),
+                Cases={
+                    (1, 1): t[i][k + 1] == 0,
+                    (0, 1): t[i][k + 1] == distances[o[i][k]][i],
+                    (1, 0): t[i][k + 1] == distances[i][o[i][k + 1]],
+                    (0, 0): t[i][k + 1] == distances[o[i][k]][o[i][k + 1]]
+                }
             ) for i in range(nTeams) for k in range(nRounds - 1)
         ],
 
         [
-            (
-                If(h[i][-1] == 1, Then=t[i][-1] == 0),
-                If(h[i][-1] != 1, Then=t[i][-1] == distances[o[i][-1]][i])
+
+            If(
+                h[i][-1] == 1,
+                Then=t[i][-1] == 0,
+                Else=t[i][-1] == distances[o[i][-1]][i]
             ) for i in range(nTeams)
         ]
-    ]
+    ],
+
 )
 
 minimize(
     # minimizing summed up travelled distance
     Sum(t)
 )
+
+"""
+1) Note how the Match structure is equivalent to:
+  (
+        If(h[i][k] == 1, h[i][k + 1] == 1, Then=t[i][k + 1] == 0),
+        If(h[i][k] != 1, h[i][k + 1] == 1, Then=t[i][k + 1] == distances[o[i][k]][i]),
+        If(h[i][k] == 1, h[i][k + 1] != 1, Then=t[i][k + 1] == distances[i][o[i][k + 1]]),
+        If(h[i][k] != 1, h[i][k + 1] != 1, Then=t[i][k + 1] == distances[o[i][k]][o[i][k + 1]])
+    ) for i in range(nTeams) for k in range(nRounds - 1)
+
+
+"""

realistic/Tower/Tower.py

@@ -53,7 +53,12 @@ def min_transmit_power(t, h):
     [tr[h] == MaximumArg([attenuation_i[h][t] * effective_power[pr[t]] for t in range(nTowers)], rank=rnk) for h in range(nHandsets)],
 
     # ensuring minimum signal strength
-    [If(tr[h] == t, Then=pr[t] >= min_transmit_power(t, h)) for h in range(nHandsets) for t in range(nTowers)],
+    [
+        If(
+            tr[h] == t,
+            Then=pr[t] >= min_transmit_power(t, h)
+        ) for h in range(nHandsets) for t in range(nTowers)
+    ],
 
     # determining which towers are overloaded
     [od[t] == (Sum(demands[h] * (tr[h] == t) for h in range(nHandsets)) > capacities[t]) for t in range(nTowers)],

realistic/UnitCommitment/UnitCommitment.py

@@ -23,7 +23,7 @@
 
 gen_max, dispatch_cost, gen_min, init_commitment, startup_cost, shutdown_cost, max_ramp_rate, demand, shed_cost, min_down, max_num_start = data
 horizon, nGenerators, nLoads = len(gen_max[0]), len(gen_max), len(demand)
-maxg = max(v for row in gen_max for v in row)
+maxg = max(max(row) for row in gen_max)
 
 # x[i][t] is 1 if the ith generator is committed at time t
 x = VarArray(size=[nGenerators, horizon], dom={0, 1})
@@ -32,7 +32,7 @@
 gl = VarArray(size=[nGenerators, horizon], dom=range(maxg // 2 + 1))
 
 # ll[j][t] is the loss of the jth load at time t
-ll = VarArray(size=[nLoads, horizon], dom=lambda l, t: range(demand[l][t] + 1))
+ll = VarArray(size=[nLoads, horizon], dom=lambda j, t: range(demand[j][t] + 1))
 
 # up[i][t] is 1 if the ith generator is up at time t
 up = VarArray(size=[nGenerators, horizon], dom={0, 1})

realistic/VRP_LC/VRP_LC.py

@@ -27,12 +27,12 @@
 from pycsp3 import *
 
 horizon, nVehicles, vehicleCapacity, nLocations, locationCapacity, nPickups, times, requests = data
-l, a, b, s, q = zip(*requests)
+rl, ra, rb, rs, rq = zip(*requests)
 nNodes, n = len(times), len(requests)  # n is the number of requests (pickups and deliveries)
 
 MainNodes = range(n + nVehicles)  # request and start nodes
 Depots = range(n, nNodes)  # start and end nodes
-qq = cp_array(list(q) + [0 for i in Depots])
+load_changes = cp_array(list(rq) + [0 for _ in Depots])
 
 # veh[i] is the vehicle visiting the ith node
 veh = VarArray(size=nNodes, dom=range(nVehicles))
@@ -73,9 +73,9 @@
     (
         (
             arr[i] <= ser[i],
-            ser[i] + s[i] <= dep[i],
-            a[i] <= ser[i],
-            ser[i] <= b[i]
+            ser[i] + rs[i] <= dep[i],
+            ra[i] <= ser[i],
+            ser[i] <= rb[i]
         ) for i in range(n)
     ),
 
@@ -89,7 +89,7 @@
     [dep[i] + times[i][succ[i]] == arr[succ[i]] for i in MainNodes],
 
     # accumulating load along route
-    [load[i] + qq[succ[i]] == load[succ[i]] for i in MainNodes],
+    [load[i] + load_changes[succ[i]] == load[succ[i]] for i in MainNodes],
 
     # setting load at start and end nodes
     [load[i] == 0 for i in Depots],
@@ -103,8 +103,8 @@
     # handling service resources
     [
         Cumulative(
-            tasks=[Task(origin=ser[i], length=s[i], height=1) for i in range(n) if l[i] == ll]
-        ) <= locationCapacity for ll in range(nLocations)
+            tasks=[Task(origin=ser[i], length=rs[i], height=1) for i in range(n) if rl[i] == p]
+        ) <= locationCapacity for p in range(nLocations)
     ],
 
     # tag(symmetry-breaking)

realistic/Vaccine/Vaccine.py

@@ -48,7 +48,12 @@
     [Sum(x[i]) <= sizes[i] // min_size for i in range(nGroups)],
 
     # imposing limits wrt vaccines and ages
-    [Cardinality([x[i][j] * ages[i] for i in range(nGroups)], occurrences={a + 1: bounds for a, bounds in enumerate(ageBounds)}) for j in range(nVaccines)],
+    [
+        Cardinality(
+            [x[i][j] * ages[i] for i in range(nGroups)],
+            occurrences={a + 1: bounds for a, bounds in enumerate(ageBounds)}
+        ) for j in range(nVaccines)
+    ],
 
     # computing the total number of vaccinated persons
     [y[v] == Sum(x[g][v] * (sizes[g] // (1 + Sum(x[g][j] for j in range(nVaccines) if j != v))) for g in range(nGroups)) for v in range(nVaccines)],

realistic/WWTPP/WWTPP.py

@@ -46,13 +46,13 @@ def table_spanning(i, start, stop):
 
 satisfy(
     # not exceeding the Wastewater Treatment Plant
-    [Sum(c[:, j] + d[:, j]) <= plantCapacity for j in range(nPeriods)],
+    [Sum(c[:, j], d[:, j]) <= plantCapacity for j in range(nPeriods)],
 
     # managing scheduled discharge flows at period 0
     [Sum(b[i][0], c[i][0]) == sd[i][0] for i in range(nIndustries) if sd[i][0] != 0],
 
     # managing scheduled discharge flows at all periods except 0
-    [[b[i][j], b[i][j - 1], d[i][j], c[i][j]] * [1, -1, 1, 1 if c[i][j] else None] == sd[i][j] for i in range(nIndustries) for j in range(1, nPeriods)],
+    [Sum(b[i][j], -b[i][j - 1], d[i][j], c[i][j]) == sd[i][j] for i in range(nIndustries) for j in range(1, nPeriods)],
 
     # ensuring compatibility between stored and discharge flows
     [(d[i][j], b[i][j - 1]) in table_compatibility(i, j) for i in range(nIndustries) for j in range(1, nPeriods)],
@@ -64,4 +64,6 @@ def table_spanning(i, start, stop):
 """ Comments
 1) when managing scheduled discharge flows, we have a list with four cells for variables and integers.
  However, when there is the special value None (at the fourth position in both lists), the two lists will be automatically reduced to three cells.
+2) one could also write the more complex expression:
+  [[b[i][j], b[i][j - 1], d[i][j], c[i][j]] * [1, -1, 1, 1 if c[i][j] else None] == sd[i][j] for i in range(nIndustries) for j in range(1, nPeriods)],
 """

realistic/Zephyrus/Zephyrus.py

@@ -4,7 +4,6 @@
 
 The model, below, is close to (can be seen as the close translation of) the one submitted to the 2016/2019 challenges.
 The MZN model was proposed by Jacopo Mauro (under the terms of the ISC License)
-No Licence was explicitly mentioned (MIT Licence assumed).
 
 ## Data Example
   12-06-8-3.json

recreational/Dominoes/Dominoes.py

@@ -55,7 +55,10 @@
         # adjacency constraints
         If(
             d != nCols,  # if not adjacent in the same column
-            Then=both(d == 1, x[i][j] // nCols == y[i][j] // nCols)  # then adjacent in the same line
+            Then=both(  # then adjacent in the same line
+                d == 1,
+                x[i][j] // nCols == y[i][j] // nCols
+            )
         ) for i, j in dominoes if (d := abs(x[i][j] - y[i][j]),)
     )
 

recreational/Fillomino/Fillomino_z.py

@@ -36,7 +36,7 @@
 
 def join(r, c):
     return [
-        AllHold(  # TODO AllHold vs conjunction
+        AllHold(  # TODO test AllHold vs conjunction
             when[r][c] == 1 + when[rr][cc],
             area[r][c] == area[rr][cc],
             what[r][c] == what[rr][cc]