Update on Actions:
Richard - Podman was not installed- received an error - command not found - Jenkins pipeline.
Richard has reached out to flag issues - command not found - docker
Pierre did not send docker file
Confirmation of extension received

Actions for this week:
Richard - Breakdown tasks add to a Kanban board by Friday - Richard to schedule
@pingou can you send on the docker file to Richard
Vipul to meet Richard on Monday for support - Richard to schedule
All meet same time next week

@pingou can you send on the docker file to Richard

I believe this refers to the antora dockerfile, if so it's located in this project: https://gitlab.com/antora/docker-antora/

Super thanks @pingou . @richardgreg please see response above

@pingou The challenge with the Jenkins build strategy is that I'm finding it tricky to build the code inside of the pipeline. I don't think running commands like dnf, yum or apt-get works inside of jenkins pipeline. Also I couldn't get the docker Jenkins agent to run inside of openshift. Vipul has suggested I use S2I, which I've been looking into, so I can enable CI using webhook triggers.

https://pagure.io/fedora-docs/fedora-docs-ci/blob/master/f/build.py is the initial code. I presume git and python will come installed in the virtual machine. node and antora might have to be installed manually or can be put in the script. Not sure about putting them in a 'requirements.txt' cause it's usually run with pip :) and they aren't python packages.

https://pagure.io/fedora-docs/fedora-docs-ci/blob/master/f/build.py is the initial code.

You may want to look at the argparse module in python to document a little bit the arguments that should be passed to the script. I see it takes 3 arguments for the moment, reading the code I can guess the first two but I have no idea what the 3rd one is (nor is it documented anywhere).

I presume git and python will come installed in the virtual machine. node and antora might have to be installed manually or can be put in the script. Not sure about putting them in a 'requirements.txt' cause it's usually run with pip :) and they aren't python packages.

You could include them in the requirements.txt just commented out so pip ignores them but if a human reads the file the information will be there.

Note that this is a start, it does not satisfy yet the criteria we've set :)

Alright. I've made some necessary corrections. You can check it out and let me know about next steps :)

You may want to look at the argparse module in python to document a little bit the arguments that should be passed to the script. I see it takes 3 arguments for the moment, reading the code I can guess the first two but I have no idea what the 3rd one is (nor is it documented anywhere).

So normally when one runs git clone <url> git clones the repo and dumps it in a folder named after the name of the repo. I suppose it will be extra work writing code to get the name of the folder and then running the rest of the build commands. It's just easier to give the folder a name during creation and use it in the code.

I've refactored the script to make it more secure and user friendly :)

Super stuff - well done Richard :)