azurerm_stack_hci_cluster - support for identity, export cloud_id, service_endpoint, resource_provider_object_id
#25031
Conversation
github-actions
bot
added
documentation
service/azure-stack-hci
size/L
waiting-response
labels
| @@ -132,3 +158,21 @@ func (r StackHCIClusterDataSource) Read() sdk.ResourceFunc { | |||
| }, | |||
| } | |||
| } | |||
|
|
|||
| func flattenSystemAssignedToModel(input *identity.SystemAndUserAssignedMap) []identity.ModelSystemAssigned { | |||
There was a problem hiding this comment.
Is there a reason we can't use the existing flatten functions in the identity package?
There was a problem hiding this comment.
The service API only supports systemAssigned Identity, which means there should be no identity_ids field in the tf schema, but the swagger and SDK uses identity.SystemAndUserAssignedMap, so need mannually skip the identity_ids in expand/flatten function.
There was a problem hiding this comment.
Then we should raise an issue on the swagger since the definition doesn't align with the API behaviour and link it here. Can you please do that?
There was a problem hiding this comment.
change this in swagger would be considered as breaking change, and from service team the user assigned identity will be supported in the future, can we leave it for now, and remove this when user assigned identity is supported later.
There was a problem hiding this comment.
@teowa we should always raise an issue on the swagger spec when what is defined in the spec does not correspond to the behaviour of the API, which is the case here. Whether the service team actually make the breaking change to fix it or not is another question, this is about tracking for when the service is brought in line. Please open an issue on the swagger spec and link it here, like we do for all other discrepancies.
There was a problem hiding this comment.
@stephybun thanks for clarifying , I thought I need to modify the swagger. The issue has been raised: Azure/azure-rest-api-specs#28260
| func expandSystemAssigned(input []interface{}) (*identity.SystemAndUserAssignedMap, error) { | ||
| if len(input) == 0 || input[0] == nil { | ||
| return &identity.SystemAndUserAssignedMap{ | ||
| Type: identity.TypeNone, | ||
| }, nil | ||
| } | ||
|
|
||
| return &identity.SystemAndUserAssignedMap{ | ||
| Type: identity.TypeSystemAssigned, | ||
| }, nil | ||
| } | ||
|
|
||
| func flattenSystemAssigned(input *identity.SystemAndUserAssignedMap) []interface{} { | ||
| if input == nil { | ||
| return []interface{}{} | ||
| } | ||
|
|
||
| if input.Type == identity.TypeNone { | ||
| return []interface{}{} | ||
| } | ||
|
|
||
| return []interface{}{ | ||
| map[string]interface{}{ | ||
| "type": input.Type, | ||
| "principal_id": input.PrincipalId, | ||
| "tenant_id": input.TenantId, | ||
| }, | ||
| } | ||
| } |
There was a problem hiding this comment.
Same question here, why can't we use the expand and flatten functions in the identity package?
There was a problem hiding this comment.
The service API only supports systemAssigned Identity, which means there should be no identity_ids field in the tf schema, but the swagger and SDK uses identity.SystemAndUserAssignedMap, so need mannually skip the identity_ids in expand/flatten function.
…d`, `service_endpoint`, `resource_provider_object_id` (hashicorp#25031) * wip * support for identity, export cloud_id, service_endpoint, resource_provider_object_id * fix * fix golangci-lint error * link swagger issue * fix doc
|
I'm going to lock this pull request because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active contributions. |
swagger: https://github.com/Azure/azure-rest-api-specs/blob/main/specification/azurestackhci/resource-manager/Microsoft.AzureStackHCI/stable/2023-08-01/clusters.json