Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/runmingl/ltl-explainability
Linear Temporal Logic to Timeline
https://github.com/runmingl/ltl-explainability
Last synced: about 1 month ago
JSON representation
Linear Temporal Logic to Timeline
- Host: GitHub
- URL: https://github.com/runmingl/ltl-explainability
- Owner: runmingl
- License: gpl-3.0
- Created: 2022-11-04T18:02:46.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2023-08-29T03:31:05.000Z (over 1 year ago)
- Last Synced: 2024-09-08T04:23:25.660Z (4 months ago)
- Language: TeX
- Homepage:
- Size: 16.8 MB
- Stars: 7
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.md
Awesome Lists containing this project
README
# Overview
LTL2TL is a tool which generates timeline visualizations for Linear temporal logic (LTL) formulas. It provides two utilities:
- `ltl2timeline`, for translating LTL formulas to timeline graphic images; and
- `ltl2regex`, for translating LTL formulas to regular expressions representing the solution set of the formula.
The tool works by transforming LTL formulae to Büchi automata (provided by [SPOT](https://spot.lre.epita.fr/)) and subsequently to $\omega$-regular expressions and timeline graphics. It is intended for validating software system specification formulas written in LTL.
The paper describing the tool is available [here](https://www.andrew.cmu.edu/user/runmingl/paper/ltl.pdf).
Installation, usage and examples are provided below.
[](https://doi.org/10.5281/zenodo.8244783)
# Install
## Option 1: Docker Image
(Recommended for trials and artifact evaluation: this method can be used on any systems with Docker installed)
- Make sure you have [Docker](https://docs.docker.com/get-docker/) installed
- Pull the docker image from Docker Hub
```
$ docker pull runmingl/ltl
```
- Run the docker image
```
$ docker run -it runmingl/ltl
```
- Run the tool
```
$ cd opt/ltl-explainability/src
$ python3 main.py ltl2timeline 'G(p xor X p)' --filename 'example' --output_format 'png'
```
This will generate a timeline image `example.gv.png` in the current directory.
- You can copy the generated images to your local machine via
```
$ docker cp :/opt/ltl-explainability/src/example.gv.png .
```
where `` is the id of the docker container you just ran.
## Option 2: Local Installation
(Recommended for development: this method requires a Unix-like system with Python 3.10+ installed)
Download the following dependencies:
### Dependencies
- SPOT (Download [here](https://spot.lre.epita.fr/install.html) and use the following command in the root directory of your installation)
```
$ ./configure --prefix ~/.local && make && sudo make install
```
- Graphviz
```
$ brew install graphviz
```
- Python 3.10+
- Other packages
```
$ pip install -r requirements.txt
```
After installing dependencies, update git submodule via
```bash
$ git submodule init
$ git submodule update
```
# Usage
We provide 2 command-line tools:
- ltl2regex
```
NAME
main.py ltl2regex
SYNOPSIS
main.py ltl2regex FORMULA
POSITIONAL ARGUMENTS
FORMULA
Type: str
```
- ltl2timeline
```
NAME
main.py ltl2timeline
SYNOPSIS
main.py ltl2timeline FORMULA
POSITIONAL ARGUMENTS
FORMULA
Type: str
FLAGS
--filename=FILENAME
Type: str
Default: 'ltl'
--output_format=OUTPUT_FORMAT
Type: str
Default: 'pdf'
Supported values: ['pdf', 'png', 'svg', 'latex']
```
## Examples
LTL Formulas are inputted following the same syntax as [SPOT](https://spot.lre.epita.fr/app/). For example,
```
python3 main.py ltl2timeline 'G(p xor X p)'
```
generates the following timeline
Some example formulas are provided in the [ltl-formulas](./ltl-formulas/) directory. For example, in [ltl-formulas/AAC_Communication_Protocol.ltl](./ltl-formulas/AAC_Communication_Protocol.ltl), we have the following formula:
```
(G (aircraft_request -> (F (! aircraft_request))))
```
which can be visualized by the tool using
```
python3 main.py ltl2timeline '(G (aircraft_request -> (F (! aircraft_request))))' --filename 'example' --output_format 'png'
```
in which case a timeline image named `example.gv.png` will be generated in the current directory.