mabdi / SmallAmp-evaluations

This repository includes the evaluation data for SmallAmp

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

SmallAmp-evaluations

This repository includes the evaluation data for SmallAmp paper titled: "Small-Amp: Test Amplification in a Dynamically Typed Language" DOI:TODO

Pull request experiment

For the details of the pull-request study, please visit this page.

Detailed amplification results

The detailed result of our quantitative study, you can visit here.

How to use

First clone me using git:

git clone https://github.com/mabdi/SmallAmp-evaluations.git
cd SmallAmp-evaluations

and clone Small-Amp inside SmallAmp-evaluations folder:

git clone https://github.com/mabdi/small-amp.git

Then unzip zip files and also move the manifest file:

mkdir pharo-projects-files
tar -xvf experiments-20211027/projects.tar -C pharo-projects-files
cp -r projects/manifest small-amp/runner/projects

Running the tool

  1. Open a fresh Pharo image and install SmallAmp.
  2. Install a project based on its loader script
  3. Run SmallAmp (you can use DrTest UI)

NOTE: the projects in manifest are referring to the cloned projects. So, you may find them outdated.

In the recent versions of SmallAmp, we have integrated the tool with Github Actions. However, you still can run it locally. Please visit here for more information.

Reporting tool

You can recreate the reports using the following commands:

cd small-amp/runner/
python3 runner.py -r longtable -d ../../pharo-projects-files/Bloc -p Bloc # Detailed reporting (tex format)
python3 runner.py -r sumtable -d ../../pharo-projects-files/Bloc -p Bloc   # Summary reporting (tex format)
python3 runner.py -r ampslog -d ../../pharo-projects-files/Bloc -p Bloc  # Input amplifiers, number of transformations

tsv format:

python3 runner.py -r longtable -d ../../pharo-projects-files/Bloc -p Bloc | tail -n +4 | sed -e 's/ & /\t/g' -e 's/\\\\$//g'

About

This repository includes the evaluation data for SmallAmp

License:MIT License


Languages

Language:HTML 53.8%Language:Python 36.4%Language:Shell 9.8%