/AutoBC

ICSE22-自动BC修复

Primary LanguageC++

RefSolver

Environment Setup

Use Docker

If you have install docker, use Dockerfile directly is recommanded. Try to build the image:

docker build . -t autobc:latest

After building is successful, start your container and then attach:

docker run -itd --rm --name autobc autobc:latest
docker exec -it autobc /bin/bash

The docker image is based on Ubuntu:18.04 and some basic tools(like gcc, g++, java) are automatically installed. Note that you may change Dockerfile sometimes if something went wrong in your environment.

Setup Directly

These components should be installed in your system:

Java Jdk 8 (1.8)  # for likelihood.jar
Java Jdk 16       # for modelcounting.jar
g++11, cmake, make
ABC               # see https://github.com/vlab-cs-ucsb/ABC
nuXmv             # see https://nuxmv.fbk.eu/

Some installed tips have been written in Dockerfile

Then, mkdir -p build to make building directory. Use following bash codes to build our program:

cd build
cmake ..
make

After all, these binary files have been built in build:

ref_solver
random_solver
ranking

Congratulations, you have deployed the environment needed for the experiment.

Run Experiment

All you need to run the experiment is use a sh that we provided. Try:

chmod +x *.sh
./experiment.sh

The experiment will be running in background. After done, the experiment will create folder experiment1.3 (if not exists) and our experiment results will be listed in it.