MiniZinc `network_flow_cost` broken about or-tools

[lperron]

I see, fully ground model.

[matsc-at-sics-se]

$ cat /tmp/latest.fzn
predicate ortools_network_flow_cost(array [int] of int: arc,array [int] of int: balance,array [int] of int: weight,array [int] of var int: flow,var int: cost);
array [1..6] of int: X_INTRODUCED_17_ = [0,2,0,2,1,1];
array [1..6] of int: X_INTRODUCED_18_ = [2,1,1,0,0,1];
array [1..3] of int: X_INTRODUCED_20_ = [1,1,-2];
array [1..12] of int: X_INTRODUCED_21_ = [0,1,0,2,1,0,1,2,2,0,2,1];
constraint ortools_network_flow_cost(X_INTRODUCED_21_,X_INTRODUCED_20_,X_INTRODUCED_18_,X_INTRODUCED_17_,3);
solve  satisfy;

[lperron]

So, nodes were 1 based in the old spec. Now I need it to support 0 based and 1 based.

[lperron]

And fixed. Thanks

[lperron]

Thanks Mats

[lperron]

I have an issue with your model.

The balance array is indexed by 1..3

but your nodes are valued from 0 to 2.

IMO, the model is invalid.

[matsc-at-sics-se]

No, Laurent. The given constraint was:

constraint
  network_flow_cost(array2d(1..6,1..2,[0,1,0,2,1,0,1,2,2,0,2,1]),
                    array1d(0..2,[1,1,-2]),
                    [2,1,1,0,0,1],
                    [0,2,0,2,1,1],
                    3);

with balance array indexed by 0..2. Since the index sets disappear in the FlatZinc, you need to pass a node base parameter to ortools_network_flow_cost, like for circuit.

[lperron]

Interesting. Let me fix this. Many thanks Laurent Perron | Operations Research | ***@***.*** | (33) 1 42 68 53 00 Le jeu. 25 janv. 2024 à 12:11, Mats Carlsson ***@***.***> a écrit :

…

No, Laurent. The given constraint was: constraint network_flow_cost(array2d(1..6,1..2,[0,1,0,2,1,0,1,2,2,0,2,1]), array1d(0..2,[1,1,-2]), [2,1,1,0,0,1], [0,2,0,2,1,1], 3); with balance array indexed by 0..2. Since the index sets disappear in the FlatZinc, you need to pass a node base parameter to ortools_network_flow_cost, like for circuit. — Reply to this email directly, view it on GitHub <#4067 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACUPL3MKSLQOKBAJVS6UR3LYQI4XVAVCNFSM6AAAAABCI5TJU2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSMBZHEZTSOJTGU> . You are receiving this because you were assigned.Message ID: ***@***.***>

[lperron]

Ok, I changed the solver specific rewrite rules and the code.

Thanks again