CloudQuery configuration issue with empty and non-empty array parameters

You must configure either one of those with a non-empty array. Do you mean what happens if you specify orgs to be non-empty and repos to be empty, or orgs to be empty and repos to be non-empty?

Well, I’m trying to understand how the app_auth method works with regard to the plugin.

I am using app_auth, which already configures the org that is getting accessed, so I’m wondering if I have to redundantly specify the org at the higher config level.

This is what I have roughly:

kind: source
spec:
  name: github
  path: cloudquery/github
  registry: cloudquery
  version: "v7.7.1"
  destinations: ["postgresql"]
  spec:
    app_auth:
    - org: {{ organization }}
      private_key_path: "${PRIVATE_KEY_PATH}"
      app_id: "${APP_ID}"
      installation_id: {{ installation_id }}
    orgs: [??]
    enterprise:
      base_url: ...
      upload_url: ...
  tables:
    - "*"

Where I’m generating that via Jinja template to provision the organization and installation_id.

Ah, got it.
Based on the validation in the code, I think you’d have to.
GitHub Link
Otherwise, it will fail.

We can look into populating that array, as I think there should be a 1-1 mapping between an app installation and access to the org.
Do you mind opening a feature request for it in GitHub Issues?
We’d have to test it a bit to ensure we’re not missing any edge cases.

Sounds good.
Yeah, I think there’s a 1-1 mapping, as you said, so the spec is redundant.