Stephen Lewis (Burrows)

yeah, `instances.Update` was not possible in the past. It would be preferable to use where possible. However, unless that would allow these fields to be updated simultaneously, it still might...

@khajduczenia it sounds like you're saying that a single call to Update should be able to change them simultaneously, though?

Thanks! Any chance you have a reference link for the API bug?

Drive-by comment: @wyardley _will_ existing tests catch this error? or is this fixing bugs that were not caught by the existing tests?

per discussion today we actually want to generate both the `*_wo` and `*_wo_version` fields.

Desired end-user config would look like: ```yaml - name: 'password' type: String description: | foo ignore_read: true sensitive: true write_only: true ``` This would automatically generate `password_wo` and `password_wo_version`.

In terms of implementation, the new fields should be added at yaml parsing time so that the rest of the code can treat them as normal fields. This is what...

Given we'd be adding the fields at yaml parsing time, that means that we'd need to make sure that the WriteOnly value set by the YAML parser doesn't conflict with...

If you'd be adding WriteOnly as an additional check everywhere that IgnoreRead is currently set, then that seems reasonable to me. @BBBmau what do you think?

IIUC this is essentially about removing the need to have a custom decoder like https://github.com/hashicorp/terraform-provider-google-beta/blob/d426a62f8183a1f1dcd63096d69f64e73369a15f/google-beta/services/bigquerydatatransfer/resource_bigquery_data_transfer_config.go#L1104-L1114. That might go beyond ignore_read. Ignore_read sets the value from the config into the state...