Brian Flad

It might be best if we just unilaterally skipped the provider source merge if the terraform or any provider configuration blocks are found in the given configurations.

Another workaround in your case without any changes in the SDK would be switching from the deprecated `Providers` field to the `ProviderFactories` field. That should skip the duplicate `provider ...`...

Submitted #1092, which should prevent the testing framework from adding its own provider configuration blocks if it detects a `TestStep.Config` that already has them. 👍

Hi @Serpentiel 👋 Can you confirm if this is only an issue when using Terraform 1.3 and later?

I have submitted https://github.com/hashicorp/terraform-plugin-sdk/pull/1089 which will strip data source states (seemingly present after import with Terraform 1.3 and later) from `ImportStateCheck`. Please note as mentioned in the PR description and...

Hey @apparentlymart 👋 Thank you for raising this. Just to quickly clarify things upfront, [v1 is no longer supported as of last year](https://discuss.hashicorp.com/t/end-of-life-timeline-for-v1-of-the-terraform-plugin-sdk/24991), but I'm guessing this is being raised...

We've certainly seen providers that use this class of attribute value extraction across `TestStep`. 😄 I'm wondering if its worth investigating this enhancement against the context of these as well:...

Commenting to keep open -- I still believe this feature would be very helpful.

Hi @mcarter106 👋 Thank you for raising this, very understandable that this behavior is likely confusing right now. The resource's preference should likely be to respect the `special = false`...

Hi @ohookins 👋 Thanks for raising this and sorry you ran into trouble here. This is the first I think we're hearing of 3.4.x issues with the `random_pet` resource, so...