Struct ExecutionState 

pub struct ExecutionState { /* private fields */ }

Manages the execution state for a durable execution.

This struct tracks all checkpointed operations, handles replay logic, and manages communication with the Lambda durable execution service.

Implementations

impl ExecutionState

pub fn new( durable_execution_arn: impl Into<String>, checkpoint_token: impl Into<String>, initial_state: InitialExecutionState, service_client: SharedDurableServiceClient, ) -> Self

Creates a new ExecutionState from the Lambda invocation input.

Arguments

• durable_execution_arn - The ARN of the durable execution
• checkpoint_token - The initial checkpoint token
• initial_state - The initial execution state with operations
• service_client - The service client for Lambda communication

pub fn with_checkpointing_mode( durable_execution_arn: impl Into<String>, checkpoint_token: impl Into<String>, initial_state: InitialExecutionState, service_client: SharedDurableServiceClient, checkpointing_mode: CheckpointingMode, ) -> Self

Creates a new ExecutionState with a specific checkpointing mode.

pub fn with_batcher( durable_execution_arn: impl Into<String>, checkpoint_token: impl Into<String>, initial_state: InitialExecutionState, service_client: SharedDurableServiceClient, batcher_config: CheckpointBatcherConfig, queue_buffer_size: usize, ) -> (Self, CheckpointBatcher)

Creates a new ExecutionState with a checkpoint batcher.

This method sets up the checkpoint queue and batcher for efficient batched checkpointing. Returns the ExecutionState and a handle to the batcher that should be run in a background task.

pub fn with_batcher_and_mode( durable_execution_arn: impl Into<String>, checkpoint_token: impl Into<String>, initial_state: InitialExecutionState, service_client: SharedDurableServiceClient, batcher_config: CheckpointBatcherConfig, queue_buffer_size: usize, checkpointing_mode: CheckpointingMode, ) -> (Self, CheckpointBatcher)

Creates a new ExecutionState with a checkpoint batcher and specific checkpointing mode.

pub fn durable_execution_arn(&self) -> &str

Returns the durable execution ARN.

pub fn durable_execution_arn_typed(&self) -> ExecutionArn

Returns the durable execution ARN as an ExecutionArn newtype.

pub async fn checkpoint_token(&self) -> String

Returns the current checkpoint token.

pub async fn set_checkpoint_token(&self, token: impl Into<String>)

Updates the checkpoint token after a successful checkpoint.

pub fn replay_status(&self) -> ReplayStatus

Returns the current replay status.

Memory Ordering

Uses Ordering::Acquire to ensure that when we read the replay status, we also see all the operations that were replayed before the status was set to New. This creates a happens-before relationship with the Release store in track_replay.

Requirements: 4.2, 4.3, 4.6

pub fn is_replay(&self) -> bool

Returns true if currently in replay mode.

pub fn is_new(&self) -> bool

Returns true if executing new operations.

pub fn checkpointing_mode(&self) -> CheckpointingMode

Returns the current checkpointing mode.

pub fn is_eager_checkpointing(&self) -> bool

Returns true if eager checkpointing mode is enabled.

pub fn is_batched_checkpointing(&self) -> bool

Returns true if batched checkpointing mode is enabled.

pub fn is_optimistic_checkpointing(&self) -> bool

Returns true if optimistic checkpointing mode is enabled.

pub fn execution_operation(&self) -> Option<&Operation>

Returns a reference to the EXECUTION operation if it exists.

pub fn get_original_input_raw(&self) -> Option<&str>

Returns the original user input from the EXECUTION operation.

pub fn execution_operation_id(&self) -> Option<&str>

Returns the EXECUTION operation’s ID if it exists.

pub async fn complete_execution_success( &self, result: Option<String>, ) -> Result<(), DurableError>

Completes the execution with a successful result via checkpointing.

pub async fn complete_execution_failure( &self, error: ErrorObject, ) -> Result<(), DurableError>

Completes the execution with a failure via checkpointing.

pub async fn get_checkpoint_result( &self, operation_id: &str, ) -> CheckpointedResult

Gets the checkpoint result for an operation.

pub async fn track_replay(&self, operation_id: &str)

Tracks that an operation has been replayed.

pub async fn load_more_operations(&self) -> Result<bool, DurableError>

Loads additional operations from the service for pagination.

pub async fn load_all_operations(&self) -> Result<(), DurableError>

Loads all remaining operations from the service.

pub async fn has_more_operations(&self) -> bool

Returns true if there are more operations to load.

pub async fn operation_count(&self) -> usize

Returns the number of loaded operations.

pub fn service_client(&self) -> &SharedDurableServiceClient

Returns a reference to the service client.

pub async fn add_operation(&self, operation: Operation)

Adds an operation to the local cache.

pub async fn update_operation( &self, operation_id: &str, update_fn: impl FnOnce(&mut Operation), )

Updates an existing operation in the local cache.

pub async fn has_operation(&self, operation_id: &str) -> bool

Checks if an operation exists in the local cache.

pub async fn get_operation(&self, operation_id: &str) -> Option<Operation>

Gets a clone of an operation from the local cache.

pub async fn mark_parent_done(&self, parent_id: &str)

Marks a parent operation as done.

pub async fn is_parent_done(&self, parent_id: &str) -> bool

Checks if a parent operation has been marked as done.

pub async fn is_orphaned(&self, parent_id: Option<&str>) -> bool

Checks if an operation would be orphaned.

pub async fn create_checkpoint( &self, operation: OperationUpdate, is_sync: bool, ) -> Result<(), DurableError>

Creates a checkpoint for an operation.

pub async fn create_checkpoint_with_response( &self, operation: OperationUpdate, ) -> Result<CheckpointResponse, DurableError>

Creates a checkpoint and returns the full response including NewExecutionState. This is useful for operations like CALLBACK that need service-generated values.

pub async fn checkpoint_sync( &self, operation: OperationUpdate, ) -> Result<(), DurableError>

Creates a synchronous checkpoint (waits for confirmation).

pub async fn checkpoint_async( &self, operation: OperationUpdate, ) -> Result<(), DurableError>

Creates an asynchronous checkpoint (fire-and-forget).

pub async fn checkpoint_optimal( &self, operation: OperationUpdate, prefer_sync: bool, ) -> Result<(), DurableError>

Creates a checkpoint using the optimal sync behavior for the current mode.

pub fn should_use_async_checkpoint(&self) -> bool

Returns whether async checkpointing is recommended for the current mode.

pub fn prioritizes_performance(&self) -> bool

Returns whether the current mode prioritizes performance over durability.

pub fn prioritizes_durability(&self) -> bool

Returns whether the current mode prioritizes durability over performance.

pub fn shared_checkpoint_token(&self) -> Arc<RwLock<String>>

Returns a reference to the shared checkpoint token.

pub fn has_checkpoint_sender(&self) -> bool

Returns true if this ExecutionState has a checkpoint sender configured.

pub async fn load_child_operations( &self, parent_id: &str, ) -> Result<Vec<Operation>, DurableError>

Loads child operations for a specific parent operation.

pub async fn get_child_operations(&self, parent_id: &str) -> Vec<Operation>

Gets all child operations for a specific parent from the local cache.

pub async fn has_replay_children(&self, operation_id: &str) -> bool

Checks if a CONTEXT operation has ReplayChildren enabled.

Trait Implementations

impl Debug for ExecutionState

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.