AP-Spotter: CPS SPA Detection Tool

Setup

For portability and replicability of this tool, we use docker.

! Note: Make sure Docker is installed before continuing.

! Note: For Windows, first install (and have it running) Docker for Windows. Then use Git Bash to run the following scripts.

Project List

The list of project to be run with AP-Spotter is available in dt/dict_repo_list.py. In this file, the location of each project is noted. This method makes it easy to have multiple projects in different locations. Each project can be in different locations, a default value is set (which is adjustable in this file), in case all the projects are in the same location.

For each project, the tool runs against a pre-stated selection of directories. This way specific directories, that might not of interest of the project, can be excluded from the run.

Docker image setup

Execute the following script for building the docker image:

./docker_scripts/build-cps-spa-detection-tool.sh

Docker image container

To create the Docker image container, run the script docker_scripts/run-cps-spa-detection-tool-container.sh, with the project name and pattern.

./docker_scripts/run-cps-spa-detection-tool-container.sh project_name pattern programming_language

• project_name: name of the project as stated in dt/dict_repo_list.py.
• project_name: hcft for the Hard Coded Fine Tuning antipattern, mwn for the Magical Waiting Number antipattern.
• programming_language: [optional] Default value is set for C++ projects (for manual setting, use cpp), for Python projects use python.

The container will stop and clean-up after finishing the run. In the meantime, to check status, run the command docker container ls.

C++ projects

For example, for running against the Arduino_IRremote project in search for the Hard Coded Fine Tuning antipattern, run:

./docker_scripts/run-cps-spa-detection-tool-container.sh Arduino_IRremote hcft

Python projects

For example, for running against the pypilot project in search for the Magical Waiting Number antipattern, run:

./docker_scripts/run-cps-spa-detection-tool-container.sh pypilot mwn python

Manual Validation

To generate the .md files for manual validation of this tool, run results_analyses_to_md_csv.py.

To be updated: In this file (results_analyses_to_md_csv.py), adjust list_analyse_files in the main function to contain the project up for manual validation. The path to the project also needs to be updated in this file under file_path.

After the .md files have been generated, analysis_cleanup.py needs to be run.

Development information

Python 3.9 is used in this project.