durabletask 0.0.0.dev68__py3-none-any.whl → 0.0.0.dev69__py3-none-any.whl

durabletask/__init__.py

@@ -3,8 +3,14 @@
 
 """Durable Task SDK for Python"""
 
+from durabletask.payload.store import LargePayloadStorageOptions, PayloadStore
 from durabletask.worker import ConcurrencyOptions, VersioningOptions
 
-__all__ = ["ConcurrencyOptions", "VersioningOptions"]
+__all__ = [
+    "ConcurrencyOptions",
+    "LargePayloadStorageOptions",
+    "PayloadStore",
+    "VersioningOptions",
+]
 
 PACKAGE_NAME = "durabletask"

durabletask/client.py

@@ -4,11 +4,12 @@
 import logging
 import uuid
 from dataclasses import dataclass
-from datetime import datetime, timezone
+from datetime import datetime
 from enum import Enum
 from typing import Any, List, Optional, Sequence, TypeVar, Union
 
 import grpc
+import grpc.aio
 
 from durabletask.entities import EntityInstanceId
 from durabletask.entities.entity_metadata import EntityMetadata
@@ -16,8 +17,23 @@ import durabletask.internal.helpers as helpers
 import durabletask.internal.orchestrator_service_pb2 as pb
 import durabletask.internal.orchestrator_service_pb2_grpc as stubs
 import durabletask.internal.shared as shared
+import durabletask.internal.tracing as tracing
 from durabletask import task
-from durabletask.internal.grpc_interceptor import DefaultClientInterceptorImpl
+from durabletask.internal.client_helpers import (
+    build_query_entities_req,
+    build_query_instances_req,
+    build_purge_by_filter_req,
+    build_raise_event_req,
+    build_schedule_new_orchestration_req,
+    build_signal_entity_req,
+    build_terminate_req,
+    check_continuation_token,
+    log_completion_state,
+    prepare_async_interceptors,
+    prepare_sync_interceptors,
+)
+from durabletask.payload import helpers as payload_helpers
+from durabletask.payload.store import PayloadStore
 
 TInput = TypeVar('TInput')
 TOutput = TypeVar('TOutput')
@@ -138,27 +154,25 @@ class TaskHubGrpcClient:
                  log_formatter: Optional[logging.Formatter] = None,
                  secure_channel: bool = False,
                  interceptors: Optional[Sequence[shared.ClientInterceptor]] = None,
-                 default_version: Optional[str] = None):
-
-        # If the caller provided metadata, we need to create a new interceptor for it and
-        # add it to the list of interceptors.
-        if interceptors is not None:
-            interceptors = list(interceptors)
-            if metadata is not None:
-                interceptors.append(DefaultClientInterceptorImpl(metadata))
-        elif metadata is not None:
-            interceptors = [DefaultClientInterceptorImpl(metadata)]
-        else:
-            interceptors = None
+                 default_version: Optional[str] = None,
+                 payload_store: Optional[PayloadStore] = None):
+
+        interceptors = prepare_sync_interceptors(metadata, interceptors)
 
         channel = shared.get_grpc_channel(
             host_address=host_address,
             secure_channel=secure_channel,
             interceptors=interceptors
         )
+        self._channel = channel
         self._stub = stubs.TaskHubSidecarServiceStub(channel)
         self._logger = shared.get_logger("client", log_handler, log_formatter)
         self.default_version = default_version
+        self._payload_store = payload_store
+
+    def close(self) -> None:
+        """Closes the underlying gRPC channel."""
+        self._channel.close()
 
     def schedule_new_orchestration(self, orchestrator: Union[task.Orchestrator[TInput, TOutput], str], *,
                                    input: Optional[TInput] = None,
@@ -169,24 +183,39 @@ class TaskHubGrpcClient:
                                    version: Optional[str] = None) -> str:
 
         name = orchestrator if isinstance(orchestrator, str) else task.get_name(orchestrator)
-
-        req = pb.CreateInstanceRequest(
-            name=name,
-            instanceId=instance_id if instance_id else uuid.uuid4().hex,
-            input=helpers.get_string_value(shared.to_json(input) if input is not None else None),
-            scheduledStartTimestamp=helpers.new_timestamp(start_at) if start_at else None,
-            version=helpers.get_string_value(version if version else self.default_version),
-            orchestrationIdReusePolicy=reuse_id_policy,
-            tags=tags
-        )
-
-        self._logger.info(f"Starting new '{name}' instance with ID = '{req.instanceId}'.")
-        res: pb.CreateInstanceResponse = self._stub.StartInstance(req)
-        return res.instanceId
+        resolved_instance_id = instance_id if instance_id else uuid.uuid4().hex
+        resolved_version = version if version else self.default_version
+
+        with tracing.start_create_orchestration_span(
+            name, resolved_instance_id, version=resolved_version,
+        ):
+            req = build_schedule_new_orchestration_req(
+                orchestrator, input=input, instance_id=instance_id, start_at=start_at,
+                reuse_id_policy=reuse_id_policy, tags=tags,
+                version=version if version else self.default_version)
+
+            # Inject the active PRODUCER span context into the request so the sidecar
+            # stores it in the executionStarted event and the worker can parent all
+            # orchestration/activity/timer spans under this trace.
+            parent_trace_ctx = tracing.get_current_trace_context()
+            if parent_trace_ctx is not None:
+                req.parentTraceContext.CopyFrom(parent_trace_ctx)
+
+            self._logger.info(f"Starting new '{req.name}' instance with ID = '{req.instanceId}'.")
+            # Externalize any large payloads in the request
+            if self._payload_store is not None:
+                payload_helpers.externalize_payloads(
+                    req, self._payload_store, instance_id=req.instanceId,
+                )
+            res: pb.CreateInstanceResponse = self._stub.StartInstance(req)
+            return res.instanceId
 
     def get_orchestration_state(self, instance_id: str, *, fetch_payloads: bool = True) -> Optional[OrchestrationState]:
         req = pb.GetInstanceRequest(instanceId=instance_id, getInputsAndOutputs=fetch_payloads)
         res: pb.GetInstanceResponse = self._stub.GetInstance(req)
+        # De-externalize any large-payload tokens in the response
+        if self._payload_store is not None and res.exists:
+            payload_helpers.deexternalize_payloads(res, self._payload_store)
         return new_orchestration_state(req.instanceId, res)
 
     def get_all_orchestration_states(self,
@@ -201,24 +230,12 @@ class TaskHubGrpcClient:
         states = []
 
         while True:
-            req = pb.QueryInstancesRequest(
-                query=pb.InstanceQuery(
-                    runtimeStatus=[status.value for status in orchestration_query.runtime_status] if orchestration_query.runtime_status else None,
-                    createdTimeFrom=helpers.new_timestamp(orchestration_query.created_time_from) if orchestration_query.created_time_from else None,
-                    createdTimeTo=helpers.new_timestamp(orchestration_query.created_time_to) if orchestration_query.created_time_to else None,
-                    maxInstanceCount=orchestration_query.max_instance_count,
-                    fetchInputsAndOutputs=orchestration_query.fetch_inputs_and_outputs,
-                    continuationToken=_continuation_token
-                )
-            )
+            req = build_query_instances_req(orchestration_query, _continuation_token)
             resp: pb.QueryInstancesResponse = self._stub.QueryInstances(req)
+            if self._payload_store is not None:
+                payload_helpers.deexternalize_payloads(resp, self._payload_store)
             states += [parse_orchestration_state(res) for res in resp.orchestrationState]
-            # Check the value for continuationToken - none or "0" indicates that there are no more results.
-            if resp.continuationToken and resp.continuationToken.value and resp.continuationToken.value != "0":
-                self._logger.info(f"Received continuation token with value {resp.continuationToken.value}, fetching next list of instances...")
-                if _continuation_token and _continuation_token.value and _continuation_token.value == resp.continuationToken.value:
-                    self._logger.warning(f"Received the same continuation token value {resp.continuationToken.value} again, stopping to avoid infinite loop.")
-                    break
+            if check_continuation_token(resp.continuationToken, _continuation_token, self._logger):
                 _continuation_token = resp.continuationToken
             else:
                 break
@@ -232,6 +249,8 @@ class TaskHubGrpcClient:
         try:
             self._logger.info(f"Waiting up to {timeout}s for instance '{instance_id}' to start.")
             res: pb.GetInstanceResponse = self._stub.WaitForInstanceStart(req, timeout=timeout)
+            if self._payload_store is not None and res.exists:
+                payload_helpers.deexternalize_payloads(res, self._payload_store)
             return new_orchestration_state(req.instanceId, res)
         except grpc.RpcError as rpc_error:
             if rpc_error.code() == grpc.StatusCode.DEADLINE_EXCEEDED:  # type: ignore
@@ -247,54 +266,46 @@ class TaskHubGrpcClient:
         try:
             self._logger.info(f"Waiting {timeout}s for instance '{instance_id}' to complete.")
             res: pb.GetInstanceResponse = self._stub.WaitForInstanceCompletion(req, timeout=timeout)
+            if self._payload_store is not None and res.exists:
+                payload_helpers.deexternalize_payloads(res, self._payload_store)
             state = new_orchestration_state(req.instanceId, res)
-            if not state:
-                return None
-
-            if state.runtime_status == OrchestrationStatus.FAILED and state.failure_details is not None:
-                details = state.failure_details
-                self._logger.info(f"Instance '{instance_id}' failed: [{details.error_type}] {details.message}")
-            elif state.runtime_status == OrchestrationStatus.TERMINATED:
-                self._logger.info(f"Instance '{instance_id}' was terminated.")
-            elif state.runtime_status == OrchestrationStatus.COMPLETED:
-                self._logger.info(f"Instance '{instance_id}' completed.")
-
+            log_completion_state(self._logger, instance_id, state)
             return state
         except grpc.RpcError as rpc_error:
             if rpc_error.code() == grpc.StatusCode.DEADLINE_EXCEEDED:  # type: ignore
-                # Replace gRPC error with the built-in TimeoutError
                 raise TimeoutError("Timed-out waiting for the orchestration to complete")
             else:
                 raise
 
     def raise_orchestration_event(self, instance_id: str, event_name: str, *,
-                                  data: Optional[Any] = None):
-        req = pb.RaiseEventRequest(
-            instanceId=instance_id,
-            name=event_name,
-            input=helpers.get_string_value(shared.to_json(data) if data is not None else None)
-        )
-
-        self._logger.info(f"Raising event '{event_name}' for instance '{instance_id}'.")
-        self._stub.RaiseEvent(req)
+                                  data: Optional[Any] = None) -> None:
+        with tracing.start_raise_event_span(event_name, instance_id):
+            req = build_raise_event_req(instance_id, event_name, data)
+            self._logger.info(f"Raising event '{event_name}' for instance '{instance_id}'.")
+            if self._payload_store is not None:
+                payload_helpers.externalize_payloads(
+                    req, self._payload_store, instance_id=instance_id,
+                )
+            self._stub.RaiseEvent(req)
 
     def terminate_orchestration(self, instance_id: str, *,
                                 output: Optional[Any] = None,
-                                recursive: bool = True):
-        req = pb.TerminateRequest(
-            instanceId=instance_id,
-            output=helpers.get_string_value(shared.to_json(output) if output is not None else None),
-            recursive=recursive)
+                                recursive: bool = True) -> None:
+        req = build_terminate_req(instance_id, output, recursive)
 
         self._logger.info(f"Terminating instance '{instance_id}'.")
+        if self._payload_store is not None:
+            payload_helpers.externalize_payloads(
+                req, self._payload_store, instance_id=instance_id,
+            )
         self._stub.TerminateInstance(req)
 
-    def suspend_orchestration(self, instance_id: str):
+    def suspend_orchestration(self, instance_id: str) -> None:
         req = pb.SuspendRequest(instanceId=instance_id)
         self._logger.info(f"Suspending instance '{instance_id}'.")
         self._stub.SuspendInstance(req)
 
-    def resume_orchestration(self, instance_id: str):
+    def resume_orchestration(self, instance_id: str) -> None:
         req = pb.ResumeRequest(instanceId=instance_id)
         self._logger.info(f"Resuming instance '{instance_id}'.")
         self._stub.ResumeInstance(req)
@@ -335,30 +346,20 @@ class TaskHubGrpcClient:
                           f"created_time_to={created_time_to}, "
                           f"runtime_status={[str(status) for status in runtime_status] if runtime_status else None}, "
                           f"recursive={recursive}")
-        resp: pb.PurgeInstancesResponse = self._stub.PurgeInstances(pb.PurgeInstancesRequest(
-            purgeInstanceFilter=pb.PurgeInstanceFilter(
-                createdTimeFrom=helpers.new_timestamp(created_time_from) if created_time_from else None,
-                createdTimeTo=helpers.new_timestamp(created_time_to) if created_time_to else None,
-                runtimeStatus=[status.value for status in runtime_status] if runtime_status else None
-            ),
-            recursive=recursive
-        ))
+        req = build_purge_by_filter_req(created_time_from, created_time_to, runtime_status, recursive)
+        resp: pb.PurgeInstancesResponse = self._stub.PurgeInstances(req)
         return PurgeInstancesResult(resp.deletedInstanceCount, resp.isComplete.value)
 
     def signal_entity(self,
                       entity_instance_id: EntityInstanceId,
                       operation_name: str,
                       input: Optional[Any] = None) -> None:
-        req = pb.SignalEntityRequest(
-            instanceId=str(entity_instance_id),
-            name=operation_name,
-            input=helpers.get_string_value(shared.to_json(input) if input is not None else None),
-            requestId=str(uuid.uuid4()),
-            scheduledTime=None,
-            parentTraceContext=None,
-            requestTime=helpers.new_timestamp(datetime.now(timezone.utc))
-        )
+        req = build_signal_entity_req(entity_instance_id, operation_name, input)
         self._logger.info(f"Signaling entity '{entity_instance_id}' operation '{operation_name}'.")
+        if self._payload_store is not None:
+            payload_helpers.externalize_payloads(
+                req, self._payload_store, instance_id=str(entity_instance_id),
+            )
         self._stub.SignalEntity(req, None)  # TODO: Cancellation timeout?
 
     def get_entity(self,
@@ -370,7 +371,8 @@ class TaskHubGrpcClient:
         res: pb.GetEntityResponse = self._stub.GetEntity(req)
         if not res.exists:
             return None
-
+        if self._payload_store is not None:
+            payload_helpers.deexternalize_payloads(res, self._payload_store)
         return EntityMetadata.from_entity_metadata(res.entity, include_state)
 
     def get_all_entities(self,
@@ -384,24 +386,12 @@ class TaskHubGrpcClient:
         entities = []
 
         while True:
-            query_request = pb.QueryEntitiesRequest(
-                query=pb.EntityQuery(
-                    instanceIdStartsWith=helpers.get_string_value(entity_query.instance_id_starts_with),
-                    lastModifiedFrom=helpers.new_timestamp(entity_query.last_modified_from) if entity_query.last_modified_from else None,
-                    lastModifiedTo=helpers.new_timestamp(entity_query.last_modified_to) if entity_query.last_modified_to else None,
-                    includeState=entity_query.include_state,
-                    includeTransient=entity_query.include_transient,
-                    pageSize=helpers.get_int_value(entity_query.page_size),
-                    continuationToken=_continuation_token
-                )
-            )
+            query_request = build_query_entities_req(entity_query, _continuation_token)
             resp: pb.QueryEntitiesResponse = self._stub.QueryEntities(query_request)
+            if self._payload_store is not None:
+                payload_helpers.deexternalize_payloads(resp, self._payload_store)
             entities += [EntityMetadata.from_entity_metadata(entity, query_request.query.includeState) for entity in resp.entities]
-            if resp.continuationToken and resp.continuationToken.value and resp.continuationToken.value != "0":
-                self._logger.info(f"Received continuation token with value {resp.continuationToken.value}, fetching next page of entities...")
-                if _continuation_token and _continuation_token.value and _continuation_token.value == resp.continuationToken.value:
-                    self._logger.warning(f"Received the same continuation token value {resp.continuationToken.value} again, stopping to avoid infinite loop.")
-                    break
+            if check_continuation_token(resp.continuationToken, _continuation_token, self._logger):
                 _continuation_token = resp.continuationToken
             else:
                 break
@@ -427,11 +417,290 @@ class TaskHubGrpcClient:
             empty_entities_removed += resp.emptyEntitiesRemoved
             orphaned_locks_released += resp.orphanedLocksReleased
 
-            if resp.continuationToken and resp.continuationToken.value and resp.continuationToken.value != "0":
-                self._logger.info(f"Received continuation token with value {resp.continuationToken.value}, cleaning next page...")
-                if _continuation_token and _continuation_token.value and _continuation_token.value == resp.continuationToken.value:
-                    self._logger.warning(f"Received the same continuation token value {resp.continuationToken.value} again, stopping to avoid infinite loop.")
-                    break
+            if check_continuation_token(resp.continuationToken, _continuation_token, self._logger):
+                _continuation_token = resp.continuationToken
+            else:
+                break
+
+        return CleanEntityStorageResult(empty_entities_removed, orphaned_locks_released)
+
+
+class AsyncTaskHubGrpcClient:
+    """Async version of TaskHubGrpcClient using grpc.aio for asyncio-based applications."""
+
+    def __init__(self, *,
+                 host_address: Optional[str] = None,
+                 metadata: Optional[list[tuple[str, str]]] = None,
+                 log_handler: Optional[logging.Handler] = None,
+                 log_formatter: Optional[logging.Formatter] = None,
+                 secure_channel: bool = False,
+                 interceptors: Optional[Sequence[shared.AsyncClientInterceptor]] = None,
+                 default_version: Optional[str] = None,
+                 payload_store: Optional[PayloadStore] = None):
+
+        interceptors = prepare_async_interceptors(metadata, interceptors)
+
+        channel = shared.get_async_grpc_channel(
+            host_address=host_address,
+            secure_channel=secure_channel,
+            interceptors=interceptors
+        )
+        self._channel = channel
+        self._stub = stubs.TaskHubSidecarServiceStub(channel)
+        self._logger = shared.get_logger("async_client", log_handler, log_formatter)
+        self.default_version = default_version
+        self._payload_store = payload_store
+
+    async def close(self) -> None:
+        """Closes the underlying gRPC channel."""
+        await self._channel.close()
+
+    async def __aenter__(self):
+        return self
+
+    async def __aexit__(self, exc_type, exc_val, exc_tb):
+        await self.close()
+
+    async def schedule_new_orchestration(self, orchestrator: Union[task.Orchestrator[TInput, TOutput], str], *,
+                                         input: Optional[TInput] = None,
+                                         instance_id: Optional[str] = None,
+                                         start_at: Optional[datetime] = None,
+                                         reuse_id_policy: Optional[pb.OrchestrationIdReusePolicy] = None,
+                                         tags: Optional[dict[str, str]] = None,
+                                         version: Optional[str] = None) -> str:
+
+        name = orchestrator if isinstance(orchestrator, str) else task.get_name(orchestrator)
+        resolved_instance_id = instance_id if instance_id else uuid.uuid4().hex
+        resolved_version = version if version else self.default_version
+
+        with tracing.start_create_orchestration_span(
+            name, resolved_instance_id, version=resolved_version,
+        ):
+            req = build_schedule_new_orchestration_req(
+                orchestrator, input=input, instance_id=instance_id, start_at=start_at,
+                reuse_id_policy=reuse_id_policy, tags=tags,
+                version=version if version else self.default_version)
+
+            parent_trace_ctx = tracing.get_current_trace_context()
+            if parent_trace_ctx is not None:
+                req.parentTraceContext.CopyFrom(parent_trace_ctx)
+
+            self._logger.info(f"Starting new '{req.name}' instance with ID = '{req.instanceId}'.")
+            # Externalize any large payloads in the request
+            if self._payload_store is not None:
+                await payload_helpers.externalize_payloads_async(
+                    req, self._payload_store, instance_id=req.instanceId,
+                )
+            res: pb.CreateInstanceResponse = await self._stub.StartInstance(req)
+            return res.instanceId
+
+    async def get_orchestration_state(self, instance_id: str, *,
+                                      fetch_payloads: bool = True) -> Optional[OrchestrationState]:
+        req = pb.GetInstanceRequest(instanceId=instance_id, getInputsAndOutputs=fetch_payloads)
+        res: pb.GetInstanceResponse = await self._stub.GetInstance(req)
+        if self._payload_store is not None and res.exists:
+            await payload_helpers.deexternalize_payloads_async(res, self._payload_store)
+        return new_orchestration_state(req.instanceId, res)
+
+    async def get_all_orchestration_states(self,
+                                           orchestration_query: Optional[OrchestrationQuery] = None
+                                           ) -> List[OrchestrationState]:
+        if orchestration_query is None:
+            orchestration_query = OrchestrationQuery()
+        _continuation_token = None
+
+        self._logger.info(f"Querying orchestration instances with query: {orchestration_query}")
+
+        states = []
+
+        while True:
+            req = build_query_instances_req(orchestration_query, _continuation_token)
+            resp: pb.QueryInstancesResponse = await self._stub.QueryInstances(req)
+            if self._payload_store is not None:
+                await payload_helpers.deexternalize_payloads_async(resp, self._payload_store)
+            states += [parse_orchestration_state(res) for res in resp.orchestrationState]
+            if check_continuation_token(resp.continuationToken, _continuation_token, self._logger):
+                _continuation_token = resp.continuationToken
+            else:
+                break
+
+        return states
+
+    async def wait_for_orchestration_start(self, instance_id: str, *,
+                                           fetch_payloads: bool = False,
+                                           timeout: int = 60) -> Optional[OrchestrationState]:
+        req = pb.GetInstanceRequest(instanceId=instance_id, getInputsAndOutputs=fetch_payloads)
+        try:
+            self._logger.info(f"Waiting up to {timeout}s for instance '{instance_id}' to start.")
+            res: pb.GetInstanceResponse = await self._stub.WaitForInstanceStart(req, timeout=timeout)
+            if self._payload_store is not None and res.exists:
+                await payload_helpers.deexternalize_payloads_async(res, self._payload_store)
+            return new_orchestration_state(req.instanceId, res)
+        except grpc.aio.AioRpcError as rpc_error:
+            if rpc_error.code() == grpc.StatusCode.DEADLINE_EXCEEDED:
+                raise TimeoutError("Timed-out waiting for the orchestration to start")
+            else:
+                raise
+
+    async def wait_for_orchestration_completion(self, instance_id: str, *,
+                                                fetch_payloads: bool = True,
+                                                timeout: int = 60) -> Optional[OrchestrationState]:
+        req = pb.GetInstanceRequest(instanceId=instance_id, getInputsAndOutputs=fetch_payloads)
+        try:
+            self._logger.info(f"Waiting {timeout}s for instance '{instance_id}' to complete.")
+            res: pb.GetInstanceResponse = await self._stub.WaitForInstanceCompletion(req, timeout=timeout)
+            if self._payload_store is not None and res.exists:
+                await payload_helpers.deexternalize_payloads_async(res, self._payload_store)
+            state = new_orchestration_state(req.instanceId, res)
+            log_completion_state(self._logger, instance_id, state)
+            return state
+        except grpc.aio.AioRpcError as rpc_error:
+            if rpc_error.code() == grpc.StatusCode.DEADLINE_EXCEEDED:
+                raise TimeoutError("Timed-out waiting for the orchestration to complete")
+            else:
+                raise
+
+    async def raise_orchestration_event(self, instance_id: str, event_name: str, *,
+                                        data: Optional[Any] = None) -> None:
+        with tracing.start_raise_event_span(event_name, instance_id):
+            req = build_raise_event_req(instance_id, event_name, data)
+            self._logger.info(f"Raising event '{event_name}' for instance '{instance_id}'.")
+            if self._payload_store is not None:
+                await payload_helpers.externalize_payloads_async(
+                    req, self._payload_store, instance_id=instance_id,
+                )
+            await self._stub.RaiseEvent(req)
+
+    async def terminate_orchestration(self, instance_id: str, *,
+                                      output: Optional[Any] = None,
+                                      recursive: bool = True) -> None:
+        req = build_terminate_req(instance_id, output, recursive)
+
+        self._logger.info(f"Terminating instance '{instance_id}'.")
+        if self._payload_store is not None:
+            await payload_helpers.externalize_payloads_async(
+                req, self._payload_store, instance_id=instance_id,
+            )
+        await self._stub.TerminateInstance(req)
+
+    async def suspend_orchestration(self, instance_id: str) -> None:
+        req = pb.SuspendRequest(instanceId=instance_id)
+        self._logger.info(f"Suspending instance '{instance_id}'.")
+        await self._stub.SuspendInstance(req)
+
+    async def resume_orchestration(self, instance_id: str) -> None:
+        req = pb.ResumeRequest(instanceId=instance_id)
+        self._logger.info(f"Resuming instance '{instance_id}'.")
+        await self._stub.ResumeInstance(req)
+
+    async def restart_orchestration(self, instance_id: str, *,
+                                    restart_with_new_instance_id: bool = False) -> str:
+        """Restarts an existing orchestration instance.
+
+        Args:
+            instance_id: The ID of the orchestration instance to restart.
+            restart_with_new_instance_id: If True, the restarted orchestration will use a new instance ID.
+                If False (default), the restarted orchestration will reuse the same instance ID.
+
+        Returns:
+            The instance ID of the restarted orchestration.
+        """
+        req = pb.RestartInstanceRequest(
+            instanceId=instance_id,
+            restartWithNewInstanceId=restart_with_new_instance_id)
+
+        self._logger.info(f"Restarting instance '{instance_id}'.")
+        res: pb.RestartInstanceResponse = await self._stub.RestartInstance(req)
+        return res.instanceId
+
+    async def purge_orchestration(self, instance_id: str, recursive: bool = True) -> PurgeInstancesResult:
+        req = pb.PurgeInstancesRequest(instanceId=instance_id, recursive=recursive)
+        self._logger.info(f"Purging instance '{instance_id}'.")
+        resp: pb.PurgeInstancesResponse = await self._stub.PurgeInstances(req)
+        return PurgeInstancesResult(resp.deletedInstanceCount, resp.isComplete.value)
+
+    async def purge_orchestrations_by(self,
+                                      created_time_from: Optional[datetime] = None,
+                                      created_time_to: Optional[datetime] = None,
+                                      runtime_status: Optional[List[OrchestrationStatus]] = None,
+                                      recursive: bool = False) -> PurgeInstancesResult:
+        self._logger.info("Purging orchestrations by filter: "
+                          f"created_time_from={created_time_from}, "
+                          f"created_time_to={created_time_to}, "
+                          f"runtime_status={[str(status) for status in runtime_status] if runtime_status else None}, "
+                          f"recursive={recursive}")
+        req = build_purge_by_filter_req(created_time_from, created_time_to, runtime_status, recursive)
+        resp: pb.PurgeInstancesResponse = await self._stub.PurgeInstances(req)
+        return PurgeInstancesResult(resp.deletedInstanceCount, resp.isComplete.value)
+
+    async def signal_entity(self,
+                            entity_instance_id: EntityInstanceId,
+                            operation_name: str,
+                            input: Optional[Any] = None) -> None:
+        req = build_signal_entity_req(entity_instance_id, operation_name, input)
+        self._logger.info(f"Signaling entity '{entity_instance_id}' operation '{operation_name}'.")
+        if self._payload_store is not None:
+            await payload_helpers.externalize_payloads_async(
+                req, self._payload_store, instance_id=str(entity_instance_id),
+            )
+        await self._stub.SignalEntity(req, None)
+
+    async def get_entity(self,
+                         entity_instance_id: EntityInstanceId,
+                         include_state: bool = True
+                         ) -> Optional[EntityMetadata]:
+        req = pb.GetEntityRequest(instanceId=str(entity_instance_id), includeState=include_state)
+        self._logger.info(f"Getting entity '{entity_instance_id}'.")
+        res: pb.GetEntityResponse = await self._stub.GetEntity(req)
+        if not res.exists:
+            return None
+        if self._payload_store is not None:
+            await payload_helpers.deexternalize_payloads_async(res, self._payload_store)
+        return EntityMetadata.from_entity_metadata(res.entity, include_state)
+
+    async def get_all_entities(self,
+                               entity_query: Optional[EntityQuery] = None) -> List[EntityMetadata]:
+        if entity_query is None:
+            entity_query = EntityQuery()
+        _continuation_token = None
+
+        self._logger.info(f"Retrieving entities by filter: {entity_query}")
+
+        entities = []
+
+        while True:
+            query_request = build_query_entities_req(entity_query, _continuation_token)
+            resp: pb.QueryEntitiesResponse = await self._stub.QueryEntities(query_request)
+            if self._payload_store is not None:
+                await payload_helpers.deexternalize_payloads_async(resp, self._payload_store)
+            entities += [EntityMetadata.from_entity_metadata(entity, query_request.query.includeState) for entity in resp.entities]
+            if check_continuation_token(resp.continuationToken, _continuation_token, self._logger):
+                _continuation_token = resp.continuationToken
+            else:
+                break
+        return entities
+
+    async def clean_entity_storage(self,
+                                   remove_empty_entities: bool = True,
+                                   release_orphaned_locks: bool = True
+                                   ) -> CleanEntityStorageResult:
+        self._logger.info("Cleaning entity storage")
+
+        empty_entities_removed = 0
+        orphaned_locks_released = 0
+        _continuation_token = None
+
+        while True:
+            req = pb.CleanEntityStorageRequest(
+                removeEmptyEntities=remove_empty_entities,
+                releaseOrphanedLocks=release_orphaned_locks,
+                continuationToken=_continuation_token
+            )
+            resp: pb.CleanEntityStorageResponse = await self._stub.CleanEntityStorage(req)
+            empty_entities_removed += resp.emptyEntitiesRemoved
+            orphaned_locks_released += resp.orphanedLocksReleased
+
+            if check_continuation_token(resp.continuationToken, _continuation_token, self._logger):
                 _continuation_token = resp.continuationToken
             else:
                 break

durabletask/extensions/__init__.py

@@ -0,0 +1,4 @@
+# Copyright (c) Microsoft Corporation.
+# Licensed under the MIT License.
+
+"""Durable Task SDK extension packages."""