Make it possible to document objects and arrays that are empty by default

[norwoodj]

Make it possible to document objects and arrays that are empty by default

[norwoodj]

Hey @ccremer, sorry I've been so long in responding. For various reasons I have little time to work on this project. This is.... A very complicated feature that I think is a very good idea but that I don't have time to build right now

[norwoodj]

Hey @ccremer I've thought a lot about this, and I think I probably won't add anything like this at this time. This is an incredibly complex feature that expands the scope of what this tool does substantially and I don't think it adds enough value to justify the code and documentation needed.

For now you can do:

# -- List of backup plans to create/ensure on startup
# @default -- list of objects of the form `{"plan": "selfResetAfter": "3s", "delay seconds": 30, "source": {"retention": "30d", "recursive": false} ... }`

Which would get rendered into the default column of the backupPlans field.

This seems like an admittedly clunkier way of achieving somewhat what you're trying to do here.

Is this okay for now? I'm going to close this issue for now. Happy to talk about other lighter-weight solutions

[ccremer]

That's ok

However I just had a new implementation idea. What if the complex example is actually a different file, completely uncommented with valid YAML structure?

Let's say we have this:

$ cat values.yaml
znapzend:
  # znapzend.backupPlans -- List of backup plans to create/ensure on startup, see
  # [example/backup-plans.yaml](./example/backup-plans.yaml) for an example
  # @import example/backup-plans.yaml
  backupPlans: []

and this file

$ cat example/backup-plans.yaml
znapzend:
  backup-plans:
  # -- Source dataset path
  - dataset: ""
    plan: {} etc

That should keep the values.yaml free from this implementation, but maybe it's possible to load an additional yaml structure?