testing: format of datasource.yaml files

[sergenyalcin]

What problem are you facing?

In some example manifests, we need to inject some values e.g tenantID, accountID, objectID. This injection values usually corresponds to Data Sources of Terraform. Please see the following example:

tenantId: ${data.azurerm_client_config.current.tenant_id}

This convention is from the Terraform registry documentation. So we use a regular expression to extract the key of the tenantID value. For this example ,the key is: azurerm_client_config.current.tenant_id

And our source yaml file has the following format:

azurerm_client_config.current.tenant_id: ***
azurerm_client_config.current.object_id: ***

@muvaf suggested two formats for data source yaml in this PR: https://github.com/upbound/official-providers/pull/299#discussion_r910804037

data:
  azurerm_client_config:
    current:
      object_id: 847c7043-b0a5-4fdf-9285-ee084c94a206

azure:
  tenantId: b9925bc4-8383-4c37-b9d2-fa456d1bb1c7
  objectId: 847c7043-b0a5-4fdf-9285-ee084c94a206

How could Official Providers help solve your problem?

Let's discuss the possible formats and decide that a best one.

[muvaf]

FWIW, I think the simple find&replace approach we took is perfectly fine for small number of entries. So this issue is about tracking whether it's going to stay like that or grow to many that we'll want to implement fully-structured YAML.

[sergenyalcin]

Because of switching the different data sources and also supporting the repo secrets, this issue was resolved. Closing.