Software Engineering and Object Oriented Programming

Welcome to the Software Engineering and Object Oriented Programming part of the SCIF20002 Programming and Data Analysis for Scientists course.

This part of the course focuses on developing your software engineering skills and your knowledge of object-oriented programming through the achievement of a practical simulation project. Over the course of five sessions, you will acquire the tools and explore the instruments to:

Create your own git repository locally and synchronise it on Github
Learn to track changes, construct commits and collaborate with others
Write an object-oriented code in C++ and Python
Write Python wrapper code to interface with C++ using pybind

Requirements

Knowledge base

You will need:

consolidated knowledge of how to interact with a Unix shell
consolidated knowledge of Python
consolidated knowledge of basic C++ syntax
initial knowledge of classes in C++
knowledge of the C++ compilation process

Software

You will need access to a complete environment with

a Unix shell (e.g. a Terminal)
a C++ compiler (gcc/g++)
an installation of git
a Python environment
suitable editor (e.g. VSCode).

The Noteable environment accessible from Blackboard provides all of these.

However, you are encouraged to have your own software development environment installed locally on your machines.

Here are some architecture-dependent recommendations:

Shell

The recommended way to do software development on Windows is to install the Windows Linux Subsystem (WSL, version 2). This installs a well-integrated local Linux distribution. ↗.

In brief, you will need to open the Windows Command Prompt in administrator mode by right-clicking and selecting Run as administrator and then type the single command

wsl --install

Restart the machine and then launch WSL.

C++ compiler

From the WSL shell, install the C/C++ compiler with

sudo apt install gcc

You will be prompted to enter the password that you will have set for your WSL user.

Git

The version control software git is installed with

sudo apt-get install git

To interact with Github, it is recommended to also install the Github command-line utility gh. To do so, first add the official sources (copy and paste command)

(type -p wget >/dev/null || (sudo apt update && sudo apt-get install wget -y)) \
    && sudo mkdir -p -m 755 /etc/apt/keyrings \
        && out=$(mktemp) && wget -nv -O$out https://cli.github.com/packages/githubcli-archive-keyring.gpg \
        && cat $out | sudo tee /etc/apt/keyrings/githubcli-archive-keyring.gpg > /dev/null \
    && sudo chmod go+r /etc/apt/keyrings/githubcli-archive-keyring.gpg \
    && echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/githubcli-archive-keyring.gpg] https://cli.github.com/packages stable main" | sudo tee /etc/apt/sources.list.d/github-cli.list > /dev/null \
    && sudo apt update \
    && sudo apt install gh -y

Then update the pakcage list and install gh

sudo apt update
sudo apt install gh

Python environment

A basic working environment in Python is installed from the WSL shell via

sudo apt install python3 python3-pip ipython3

You can then install scientific computing packages via pip

pip install numpy matplotlib scipy

Code editor

The simplest option is to install Visual Studio Code and configure it to work with WSL.

Here is the offical guidance ↗️

Shell

The Mac already has a proper shell, called Terminal.

It is useful however to install a package manager that simplifies the installation of software. The most suitable is homebrew.

To install it, open a Terminal and type (you can copy and paste the command below)

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

C++

The Mac already comes with a good compiler installed called clang, with aliases for commands such as gcc and g++.

You may want to install a GNU compiler ( the standard gcc compiler that you have, for example, on the Bristol High Performance Computing clusters) you can install it with homebrew using:

brew install gcc

This will install the latest version of gcc, which you will access with the command gcc-xx where xx is the version number. For example gcc-14.

Git

The Mac should have git pre-installed.

To interact with Github, it is recommended to also install the Github command-line utility gh via

brew install gh

Python

The Mac already has Python installed as well. You can use it and readily use pip to install the packages that you need using

pip3 install numpy matplotlib scipy

However, a more flexible way to manage different version of Python and Python environments is to install pyenv. More information here ↗️

Code editor

You can install whatever editor you prefer. A simple option is Visual Studio Code, which you can download from this link ↗️

A telemetry-free version of VSCode can be installed via homewbrew

brew install --cask vscodium

Depending on the Linux distribution, you may have different package managers to install software. Here we assume a Debian-based distribution such as Ubuntu, where the package manager is apt (or apt-get).

Shell

All Linux distributions have an application to launch a shell, often called Terminal.

C++

The C/C++ compilers are normally already installed. If not, just type

sudo apt-get install gcc

Git

git is normally available. If not, just type

sudo apt-get install git

To interact with Github, it is recommended to also install the Github command-line utility gh. To do so, first add the official sources (copy and paste command)

(type -p wget >/dev/null || (sudo apt update && sudo apt-get install wget -y)) \
    && sudo mkdir -p -m 755 /etc/apt/keyrings \
        && out=$(mktemp) && wget -nv -O$out https://cli.github.com/packages/githubcli-archive-keyring.gpg \
        && cat $out | sudo tee /etc/apt/keyrings/githubcli-archive-keyring.gpg > /dev/null \
    && sudo chmod go+r /etc/apt/keyrings/githubcli-archive-keyring.gpg \
    && echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/githubcli-archive-keyring.gpg] https://cli.github.com/packages stable main" | sudo tee /etc/apt/sources.list.d/github-cli.list > /dev/null \
    && sudo apt update \
    && sudo apt install gh -y

Then update the pakcage list and install gh

sudo apt update
sudo apt install gh

Python

You install Python via

sudo apt-get install python3 python3-pip ipython3

You can then install scientific computing packages via pip

pip install numpy matplotlib scipy

Code editor

You can install whatever editor you prefer, including VSCode ↗️.

On Ubuntu, a telemetry-free version of VSCode can be installed via snaps

snap install codium --classic

Chromebooks have a Linux operating system running under the hood.

It is possible to access it by activating the Linux development environment, see the documentation ↗️

You can then follow the same instructions as Linux users.

In the Noteable environment (e.g. Jupyter Lab) you can open a terminal readily. It already has g++, python and git installed.

The only missing program is the Github command line utility gh. This allows you to interact with Github and modify a remote git repository. This is ultimately optional (local git repositories are sufficient to learn git).

To install gh, we simply fetch a precompiled version. Type the following in the Noteable terminal

wget https://github.com/cli/cli/releases/download/v2.60.1/gh_2.60.1_linux_amd64.tar.gz

This fetches a compressed package gh_2.60.1_linux_amd64.tar.gz (make sure you do this in teh home folder). To decompress it type

 tar -xvf gh_2.60.1_linux_amd64.tar.gz

The precompiled executable gh is inside $HOME/gh_2.60.1_linux_amd64/bin/gh. For convenience, we can add an alias by creating a Bash configuration file in the home folder (i.e. ~)

nano .bash_profile

and write the following line

alias gh="$HOME/gh_2.60.1_linux_amd64/bin/gh"

To update the configuration, just type

source .bash_profile

Now the command gh is also available.