Interpreter¶

The interpreter is a special mode where Kind 2 reads inputs from a file and prints the outputs of the system at each step.

To use the interpreter, run

kind2 --enable interpreter <lustre_file> --interpreter_input_file <input_file>


You can specify the number of steps to run with the option --interpreter_steps <int>. By default, the number of steps is determined by the input file.

Structure of the input file¶

The inputs must be specified in a JSON file.

The overall structure is as follows:

[
{
"var1": "42",
"var2": true,
"var3": "0.5"
},
{
"var1": "24",
"var2": false,
"var3": "0.5"
}
]


The top-level JSON array corresponds to the successive time steps. Each time step is described by a JSON object associating to each input variable its value for this time step.

NOTE: Kind2 also accepts the CSV format for backward compatibility reasons. However, it does not support records, arrays and tuples. Please give your input file the adequate extension (*.json or *.csv) in order to indicate to Kind2 which format you are using.

Integers and reals¶

As in the above example, integers and reals should be written as strings in order to avoid a potential loss of precision or an integer overflow while parsing the file. Nevertheless, small integers can be written as native JSON integers without problem.

Records¶

Record values can be expressed using a JSON object.

For instance, a variable c of type { re: real; im: real } can be assigned as follows:

[
{
"c": { "re": "-1.0", "im": "0.25" }
}
]


Arrays¶

Array values can be expressed using a JSON array.

For instance, a variable a of type bool^3^2 can be assigned as follows:

[
{
"a": [[true, true, false], [false, true, true]]
}
]


Tuples¶

The JSON format does not support tuples by default. However, Kind2 extends the JSON syntax so that tuples can be easily expressed.

For instance, a variable t of type [int, bool, real] can be assigned as follows:

[
{
"t": ("36", false, "5.0")
}
]


An alternative syntax using a JSON object is allowed in case you want to produce a valid JSON file:

[
{
"t": { "0":"36", "1": false, "2":"5.0" }
}
]