A Dynamic Logic Program (DLP) Interpreter

moviola is an interactive interpreter for dynamic logic programs. It implements the following causal rejection-based update semantics on DLPs.

• Update answer set semantics (AS)
• Justified update semantics (JU)
• Dynamic stable model semantics (DS)
• Refined dynamic stable model semantics (RD)

moviola implements these semantics by encoding them in online Answer Set Programming (ASP). It uses clingo from Potassco.

moviola utilizes online solving capacity of clingo to compute answer sets of DLPs in a multi-shot fashion. In this way there is no need to start the solving process from scratch whenever a new update program comes.

moviola needs clingo with Python 3 scripting support. Check Potassco page on getting clingo.

Additionally, it needs Python 3 to be installed on your system. In case the cmd2 package has not been installed on your system, please check package repositories for your operating system. In Debian/Ubuntu based systems, you can run apt-get install python3-cmd2 to install cmd2 for Python.

moviola is in early development stage. There is no official release yet. You can download the repository as a zip or fetch it using git.

The main mode of use for moviola is via interactive command-line shell. It can be run as follows. The output shows the active update semantics (refined update semantics - RD - in this case) and the >> prompt denotes moviola is waiting your commands.

$ clingo --acyc-prop=0 moviola-py.lp
clingo version 5.1.0
Reading from moviola-py.lp
Changing semantics: RD
>> 

The active update semantics can be inspected and changed. The options are as, ju, ds, and rd, update answer set semantics, justified update semantics, dynamic stable model semantics, and refined dynamic stable model semantics, respectively.

>> semantics
Update semantics: RD
>> semantics ds
Changing semantics: DS

Using update command you can enter an update logic program. It opens your default editor (in Linux via EDITOR environment variable).

>> update

After you enter and save the propositional normal logic program using your editor, moviola processes it and updates the current DLP. Let's enter the logic program with only one fact p. as an update. Here is what moviola outputs:

p.

Reifying u.1.lp ...
Updating ...

We can compute the models of the current DLP by solve command. Additionally, the models command expects an integer that shows the maximum number of models to compute (1 is the default value and 0 means all models). Note that we are using the RD semantics and the only model we get is {p}.

>> models 0
>> solve
Solving...
Answer: 1
p

Let's update the DLP with the program not p :- not p.. {p} is still the only model we get according to the RD semantics.

>> update
not p :- not p.

Reifying u.2.lp ...
Updating ...
>> solve
Solving...
Answer: 1
p

You can change the underlying semantics anytime and investigate their differences.

>> semantics ju
Changing semantics: JU
>> solve
Solving...
Answer: 1
p
Answer: 2

>> semantics ds
Changing semantics: DS
>> solve
Solving...
Answer: 1
p

Note that according to the JU semantics, our DLP has an extra model {}, which is unintended considering the tautological second update.