TACAS 22 Artifact Evaluation VM - Ubuntu 20.04 LTS

This is a virtual machine (VM) with a GNU/Linux installation to be used for evaluation of artifacts accompanying papers at computer science conferences or journals. It is based on Ubuntu 20.04 with the following additional packages installed: build-essential, cmake, clang, mono-complete, openjdk-8-jdk, python3.8, pip3, ruby, and a 32-bit libc. Moreover, VirtualBox guest additions are installed on the VM, it is therefore possible to easily connect a shared folder from a host computer running VirtualBox. The login and password of the default user are: “tacas22” / “tacas22”.

The VM is intended to be used with artifacts that are self-contained, i.e., they contain the presented software, plus all necessary dependencies, so that they can be evaluated without an Internet connection (to protect the anonymity of the reviewers and also to be reproducible even after several years when some web pages have been taken down).

This text and VM setup is heavily based on the previous VMs for the TACAS artifact evaluations, in particular Sebastian Hjort Hyberts, Peter Gjøl Jensen, & Thomas Neele. (2020). TACAS 21 Artifact Evaluation VM - Ubuntu 20.04 LTS.

When preparing an artifact for this VM, the dependencies should be also provided. You can include the required Debian packages (.deb files) in the artifact archive and ask the user to install them using, e.g.,

$ sudo dpkg -i <file>

In order to find out and download all the dependencies of some package, you can, taken the octave package as an example, run the following code:

$ mkdir packages
$ cd packages
$ sudo apt-get update
$ apt-get --print-uris install octave | grep -oP "(?<=').*(?=')" > octave.deps
$ for i in $(cat octave.deps) ; do wget -nv $i ; done

and then ask the user to install them using

$ cd packages
$ sudo dpkg -i *.deb

For Python, you can, e.g., use pip3 to download the packages. For instance, to download the bitarray package, you can run (we assume you already have pip3 installed)

$ pip3 download bitarray

The downloaded package can then be installed using

$ pip3 install bitarray-1.5.3.tar.gz