ViSTA (Visualization of Solving Trace for ASP)

ViSTA is a toolset for tracing and visualizing the solving process of Answer Set Programming (ASP) programs.

ASP is a declarative programming paradigm used for knowledge representation and combinatorial problem solving. Modern ASP solvers use a CDNL (Conflict-Driven Nogood Learning) algorithm — similar to CDCL in SAT solving — to search for answer sets. ViSTA lets you step through that internal solving process visually, making it useful for understanding, debugging, and educational purposes.

Overview

The pipeline consists of three stages:

1. Grounding — gringo converts a high-level .lp program into a flat .aspif format.
2. Conversion — converter.py normalizes weighted/cardinality rules into plain rules (.con format).
3. Tracing — cdnl-tracer.py executes the CDNL search and records each solving event as a JSON trace.
4. Visualization — The web-based visualizer renders the trace interactively.

Visualizer

The online visualizer is available at: https://tomoyay1622.github.io/ViSTA/

Getting Started

Requirements

• Python 3.8+
• gringo / clingo (for grounding .lp files)
• A modern web browser

Installation

Clone this repository:

$ git clone https://github.com/tomoyay1622/ViSTA.git
$ cd ViSTA

No additional Python packages are required.

How to Use

Step 1. Ground the ASP program with gringo:

$ gringo example/test.lp > example/test.aspif

Step 2. Normalize rules with converter.py (converts weighted/cardinality rules into plain rules):

$ python3 converter.py example/test.aspif > example/test.con

Step 3. Run the CDNL tracer to obtain a solving trace in JSON format:

$ python3 cdnl-tracer.py example/test.con > example/test.json

Step 4. Open the online visualizer and paste the contents of example/test.json to explore the trace interactively.