Struct CloudFormationService 

pub struct CloudFormationService { /* private fields */ }

Implementations

impl CloudFormationService

pub fn new(state: SharedCloudFormationState, deps: CloudFormationDeps) -> Self

pub fn with_snapshot_store(self, store: Arc<dyn SnapshotStore>) -> Self

pub fn with_s3_store(self, store: Arc<dyn S3Store>) -> Self

Wire the fine-grained S3 disk store so CFN-provisioned buckets are written through to disk (see ResourceProvisioner::s3_store).

pub fn with_snapshot_hooks( self, hooks: BTreeMap<&'static str, SnapshotHook>, ) -> Self

Register the per-service snapshot persist hooks (keyed by service name) used to persist every snapshot-backed service a stack op touches.

pub fn cloudcontrol_create( &self, type_name: &str, properties: Value, account_id: &str, region: &str, ) -> Result<CloudControlOutcome, String>

Provision a single resource of type_name with concrete properties (Cloud Control DesiredState). Returns the physical id + GetAtt-style attributes on success, or the handler error message.

pub fn cloudcontrol_update( &self, type_name: &str, physical_id: &str, prior_attributes: &BTreeMap<String, String>, new_properties: Value, account_id: &str, region: &str, ) -> Result<CloudControlOutcome, String>

Apply a new desired state to an existing resource. prior_attributes is what the previous create/update returned; it lets us reconstruct the backing StackResource the update handlers expect without leaking that type across the crate boundary.

pub fn cloudcontrol_delete( &self, type_name: &str, physical_id: &str, prior_attributes: &BTreeMap<String, String>, account_id: &str, region: &str, ) -> Result<(), String>

Delete an existing resource, reaping any real backing container.

pub async fn cloudcontrol_persist_type(&self, type_name: &str)

Flush the backing service state a Cloud Control op just mutated to disk, using the SAME snapshot hooks CreateStack/ExecuteChangeSet fire. A CCAPI create/update/delete goes straight through the provisioner without touching the stack handlers, so without this the provisioned SQS queue (etc.) would never be persisted and a restart would leave Cloud Control listing a resource whose owning service lost its backing state.

Trait Implementations

impl AwsService for CloudFormationService

fn service_name(&self) -> &str

The AWS service identifier (e.g., “sqs”, “sns”, “sts”, “events”, “ssm”).

fn handle<'life0, 'async_trait>( &'life0 self, req: AwsRequest, ) -> Pin<Box<dyn Future<Output = Result<AwsResponse, AwsServiceError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Handle an incoming request.

fn supported_actions(&self) -> &[&str]

List of actions this service supports (for introspection).

fn iam_enforceable(&self) -> bool

Whether this service participates in opt-in IAM enforcement (FAKECLOUD_IAM=soft|strict).

fn iam_action_for(&self, _request: &AwsRequest) -> Option<IamAction>

Derive the IAM action + resource ARN for an incoming request.

fn iam_condition_keys_for( &self, _request: &AwsRequest, _action: &IamAction, ) -> BTreeMap<String, Vec<String>>

Derive service-specific IAM condition keys for an incoming request.

fn resource_tags_for( &self, _resource_arn: &str, ) -> Option<HashMap<String, String>>

Return the tags on the resource identified by resource_arn.

fn request_tags_from( &self, _request: &AwsRequest, _action: &str, ) -> Option<HashMap<String, String>>

Extract tags being sent in the request (e.g. on CreateQueue, PutObject with x-amz-tagging, TagResource).