chore(orchestration): add support for autoRetry, maxAwaitingRetries, retryStep (#13391)

RESOLVES CORE-1163
RESOLVES CORE-1164

**What**

### Add support for non auto retryable steps.

When marking a step with `maxRetries`, when it will fail it will be marked as temporary failure and then retry itself automatically. Thats the default behaviour, if you now add `autoRetry: false`, when the step will fail it will be marked as temporary failure but not retry automatically. you can now call the workflow engine run to resume the workflow from the failing step to be retried.

### Add support for `maxAwaitingRetries`

When setting `retyIntervalAwaiting` a step that is awaiting will be retried after the specified interval without maximun retry. Now you can set `maxAwaitingRetries` to force a maximum awaiting retry number

### Add support to manually retry an awaiting step

In some scenario, either a machine dies while a step is executing or a step is taking longer than expected, you can now call `retryStep` on the workflow engine to force a retry of the step that is supposedly stucked

packages/core/workflows-sdk/src/helper/type.ts

@@ -25,6 +25,10 @@ export type FlowRegisterStepSuccessOptions<TData = unknown> =
     response?: TData
   }
 
+export type FlowRetryStepOptions = Omit<BaseFlowRunOptions, "resultFrom"> & {
+  idempotencyKey: string
+}
+
 export type FlowRegisterStepFailureOptions<TData = unknown> =
   BaseFlowRunOptions & {
     idempotencyKey: string
@@ -93,6 +97,7 @@ export type ExportedWorkflow<
       TResultOverride extends undefined ? TResult : TResultOverride
     >
   >
+  retryStep: (args?: FlowRetryStepOptions) => Promise<WorkflowResult>
   cancel: (args?: FlowCancelOptions) => Promise<WorkflowResult>
 }
 

packages/core/workflows-sdk/src/helper/workflow-export.ts

@@ -28,6 +28,7 @@ import {
   FlowCancelOptions,
   FlowRegisterStepFailureOptions,
   FlowRegisterStepSuccessOptions,
+  FlowRetryStepOptions,
   FlowRunOptions,
   MainExportedWorkflow,
   WorkflowResult,
@@ -53,7 +54,7 @@ function createContextualWorkflowRunner<
   container?: LoadedModule[] | MedusaContainer
 }): Omit<
   LocalWorkflow,
-  "run" | "registerStepSuccess" | "registerStepFailure" | "cancel"
+  "run" | "registerStepSuccess" | "registerStepFailure" | "cancel" | "retryStep"
 > &
   ExportedWorkflow<TData, TResult, TDataOverride, TResultOverride> {
   const flow = new LocalWorkflow(workflowId, container!)
@@ -62,6 +63,7 @@ function createContextualWorkflowRunner<
   const originalRegisterStepSuccess = flow.registerStepSuccess.bind(flow)
   const originalRegisterStepFailure = flow.registerStepFailure.bind(flow)
   const originalCancel = flow.cancel.bind(flow)
+  const originalRetryStep = flow.retryStep.bind(flow)
 
   const originalExecution = async (
     method,
@@ -310,6 +312,46 @@ function createContextualWorkflowRunner<
   }
   flow.registerStepFailure = newRegisterStepFailure as any
 
+  const newRetryStep = async (
+    {
+      idempotencyKey,
+      context: outerContext,
+      throwOnError,
+      logOnError,
+      events,
+      container,
+    }: FlowRetryStepOptions = {
+      idempotencyKey: "",
+    }
+  ) => {
+    idempotencyKey ??= ""
+    throwOnError ??= true
+    logOnError ??= false
+
+    const [, transactionId] = idempotencyKey.split(":")
+    const context = {
+      ...outerContext,
+      transactionId,
+      __type: MedusaContextType as Context["__type"],
+    }
+
+    context.eventGroupId ??= ulid()
+
+    return await originalExecution(
+      originalRetryStep,
+      {
+        throwOnError,
+        container,
+        logOnError,
+      },
+      idempotencyKey,
+      undefined,
+      context,
+      events
+    )
+  }
+  flow.retryStep = newRetryStep as any
+
   const newCancel = async ({
     transaction,
     transactionId,
@@ -367,7 +409,11 @@ export const exportWorkflow = <TData = unknown, TResult = unknown>(
     container?: LoadedModule[] | MedusaContainer
   ): Omit<
     LocalWorkflow,
-    "run" | "registerStepSuccess" | "registerStepFailure" | "cancel"
+    | "run"
+    | "registerStepSuccess"
+    | "registerStepFailure"
+    | "cancel"
+    | "retryStep"
   > &
     ExportedWorkflow<TData, TResult, TDataOverride, TResultOverride> {
     return createContextualWorkflowRunner<
@@ -388,11 +434,17 @@ export const exportWorkflow = <TData = unknown, TResult = unknown>(
       | "run"
       | "registerStepSuccess"
       | "registerStepFailure"
-      | "cancel",
+      | "cancel"
+      | "retryStep",
     TDataOverride,
     TResultOverride
   >(
-    action: "run" | "registerStepSuccess" | "registerStepFailure" | "cancel",
+    action:
+      | "run"
+      | "registerStepSuccess"
+      | "registerStepFailure"
+      | "cancel"
+      | "retryStep",
     container?: LoadedModule[] | MedusaContainer
   ) => {
     const contextualRunner = createContextualWorkflowRunner<
@@ -495,6 +547,22 @@ export const exportWorkflow = <TData = unknown, TResult = unknown>(
     )(inputArgs)
   }
 
+  exportedWorkflow.retryStep = async <
+    TDataOverride = undefined,
+    TResultOverride = undefined
+  >(
+    args?: FlowRetryStepOptions
+  ): Promise<WorkflowResult> => {
+    const container = args?.container
+    delete args?.container
+    const inputArgs = { ...args } as FlowRetryStepOptions
+
+    return await buildRunnerFn<"retryStep", TDataOverride, TResultOverride>(
+      "retryStep",
+      container
+    )(inputArgs)
+  }
+
   exportedWorkflow.cancel = async (
     args?: FlowCancelOptions
   ): Promise<WorkflowResult> => {

packages/core/workflows-sdk/src/utils/composer/type.ts

@@ -268,7 +268,11 @@ export type ReturnWorkflow<TData, TResult, THooks extends any[]> = {
     container?: LoadedModule[] | MedusaContainer
   ): Omit<
     LocalWorkflow,
-    "run" | "registerStepSuccess" | "registerStepFailure" | "cancel"
+    | "run"
+    | "registerStepSuccess"
+    | "registerStepFailure"
+    | "cancel"
+    | "retryStep"
   > &
     ExportedWorkflow<TData, TResult, TDataOverride, TResultOverride>
 } & {