Initial ports service

[willejs]

Hopefully i have covered how i have approached the task in the README.md over here.

I usually would break a big changeset like this down into multiple small PRs, and keep getting feedback from others as i progressed. In some cases, if i was using a new approach or technology I would usually write an RFC and get some comments inline in the PR with the code along side it, and later ratify it with an ADR, either in this repo or another depending on what its covering.

Tasks:

• Add a simple readme
• Add Dockerfile
• Add CI to test the PR
• Add vscode config
• Add basic folder structure
• Add basic service that works
• Add otel telemetry and expose the providers
• Add logs
• Instrument http handlers via some middleware
• Inject the logger and otel providers into parts of the service
• Add a helm chart
• Add a config package and use it to set port and the ports data file
• Add details on how to test the tracing
• Add some more traces down the stack
• Add a dummy CD pipeline showing how we could progress this through envs
• Add some more tests?
• Fix up any glaring issues

Hopefully the readme and comments throughout cover this, i am trying to timebox myself to 3 hours.

[willejs]

I have spent just over two hours on this now and am going to submit the task and take a break. Tomorrow i might spend another hour or so on this and try and complete some more tasks.

This will run locally, and the tests should pass... However its not running in docker currently, its due to the file location being incorrect, and I am going to fix it.

Ignore some of those junk fix(ci) commits, i couldn't test this locally with act for some reason, so i kept pushing. I would usually do a squash and merge on a pull request or rebase the commits and squash them out of this branch.

[willejs]

I have spent around another 45 minutes or so fixing the helm chart, adding a config package, and some more traces down the stack as an example. I added some details into the readme about how to check out the traces, and what i would do in future. The tracing down the stack is not the most elegant implementation.

I am missing some more goals i had of using semver for the service via the semantic release github action and pushing the image to my registry in github actions, and also creating a dummy deploy pipeline. But I am happy to have a chat about this, and its easy to do in future.

[willejs]

I would usually squash merge this, but its an initial PR and its big.