Git & GitHub

• This is a nice tutorial / game to get acquainted with all the Git & GitHub commands (look inside the releases tab to download it)

  https://github.com/jlord/git-it-electron

• This is a video tutorial explaining how to use Git & GitHub inside the VSCode IDE (useful when working on the repository on your local device)

  https://youtu.be/i_23KUAEtUM

What is Git

Open source & free source control management (= SCM)

⇒ version control for program code

• archive of all previous versions & changes of your code

• all changes are linked to the author of the code

• everything is encrypted

• many people can work on the same code

Git

This is the actual version control system

You can use it locally in the CLI (command line interface), a GUI, or right inside your IDE

GitHub

This is the server/website used to store your code, track issues, develop code in groups, …

Uses markdown (.md) files to provide information about repositories

e.g. the README.md files but also the wiki pages

Git history: commits

The history forms a chain of snapshots (commits) of your code.

Every commit contains a hash function that links it to the chain cryptographically securing the whole history

With the git log (--oneline or -p) command you can actually view all these changes through time - with the hash function, messages, and the author (regular, short, or long version) (here 2 commits were performed)

💡

You can create code references for your issues

Git workflow

Create, commit and push new files

You can always view the state of your file with the git status command

1. Create or alter a file locally

   Use git init to initiate git inside your current working directory

   💡

   Modified state

   = when some changes have been made to a file, which are not yet saved

   ⇒ the state of file has been altered from its previous state in the committed state

   

2. git add <file> to add the file to the staging area (or use add . or add --all or add -A to add all the files in the working directory)

   💡

   Staged state

   = a file that is ready to be committed, all necessary changes have been made

   ⇒ this command doesn’t return anything but you can always view the status of your files with the git status command

   

Optionally: make git ignore certain files in the directory (e.g. sensitive data)

→ create .gitignore file

→ specify which or what type of files you wish to ignore

*.txt will make sure are .txt files are ignored

List of ways to ignore files

Now .txt files no longer show up when you ask for the git status

3. git commit to take a snapshot (commit) of all the files in the staging area and add the commit to the local repository

   ⇒ these files are now officially part of the git record

   If you commit, you have created a ‘version’ of your code that can be altered from that point, compared against, or jumped to at any time

   • You can add a message to this commit with the argument -m followed by a string with your message
   • You can also use git commit -a -m "your message" to skip the staging step and go straight to committing your file

   💡

   Committed state

   = a file is committed when all the changes made to the file have been saved in the local repo

   ⇒ ready to be pushed to the remote repo (on GitHub)

   

When you make changes to your committed repository, git will recognize that some of your files have been modified

git diff lets you view what changes have been made

4. Push the repository to GitHub

   git remote add origin <link> → creates a connection between your local repo and the remote repo on GitHub (the connection is called origin)

   git branch -M main → changes your target branch’s name to ‘main’

   git push -u origin main → pushes your repository from your local device to GitHub

   git push--all to push through all your branches

   

   Inside GitHub, you can add and edit files…

Making changes to your repository on GitHub

When making changes, you can choose to create a new branch and start a pull request

⇒ request to merge this to the main branch

From here you can discuss the pull request, link it to certain issues or milestones, …

If all is well & there are no conflicts, you can merge the request to the main branch and close all corresponding issues

There are many more repository options including wiki, insights, and release

Pull from GitHub

For this you have two options:

git fetch → download all the history from the remote tracking branches

git merge → merge this fetch with your local branch

git pull combines both of these commands

Branches

These allow you to

• develop older versions of your code

• develop new features off of the main branch (live version)

  ⇒ stable vs. dev branches

• you can merge branches or add features from one branch to another

Use git branch branchName to create a new branch to which you can make changes

Merging conflicts

This branch merger went quite smooth, nothing was changed in main so there were no conflicts…

When this happens, you get into a merger branch

Git will actually show you, inside your files, where and what the conflict it

→ HEAD indicates what is in main

→ fix conflicts and commit

UGent High-performance Computer (HPC)

https://hpcugent.github.io/vsc_user_docs/pdf/intro-HPC-linux-gent.pdf

Connecting

To submit jobs to the HPC you need a ssh (secure shell) connection

Uploading

Open FileZilla and create an SFTP connection

Submitting jobs

You can create job-files in the .pbs format

• Submit your jobs with qsub

• View the status of your job with qstat ⇒ Queued, Running, Completed, Failes

The HPC will return a job_id at the start of the run and output an error-file file_name.pbs.e.job_id which provides you with info

.pbs files

These always start with a shebang line

The following lines provide information to the server for your job

• nodes: how many nodes to use (default = 1)

• ppn: processors per node (i.e. CPU cores)

• walltime: the maximal runtime of your script

• mem: memory per code (limited depending on which cluster you use)

#!/usr/bin/env bash

#PBS -l nodes=1:ppn=1
#PBS -l walltime=00:30:00
#PBS -l mem=2gb

echo "hello world!" # prints string
cat Test.txt	      # concatenate: prints file(s) content