In my view, the most valuable aspect is test/live drift. We’ve only implemented half of this concept so far: the price catalog is hardcoded as a constant, and CI validates it against the documentation. Our check detects any discrepancies, and a Terraform-style push-reconciliation fixes them. One question usually determines the fate of such tools: how does the `plan` phase handle resources created outside the configuration—for instance, if the support team manually adds a coupon or a price via the dashboard?
"Terraform-for-anything" lives or dies based on its ability to detect external drift, not on the elegance of the `apply` operation.
Not that I know of! Though honestly Paddle kind of solves this by default, as a merchant of record there's barely any account state to configure, so there's much less to drift.
In my view, the most valuable aspect is test/live drift. We’ve only implemented half of this concept so far: the price catalog is hardcoded as a constant, and CI validates it against the documentation. Our check detects any discrepancies, and a Terraform-style push-reconciliation fixes them. One question usually determines the fate of such tools: how does the `plan` phase handle resources created outside the configuration—for instance, if the support team manually adds a coupon or a price via the dashboard? "Terraform-for-anything" lives or dies based on its ability to detect external drift, not on the elegance of the `apply` operation.
Quite interesting! Is there a PaddleKit which helps configuring Paddle?
Not that I know of! Though honestly Paddle kind of solves this by default, as a merchant of record there's barely any account state to configure, so there's much less to drift.
Sweet
thanks!