fix: fetch paginated cloud execution history (#514)

* fix: fetch paginated cloud execution history

* chore: update SAM template

* test: move cloud runner import to module scope

---------

Co-authored-by: GitHub Action <action@github.com>

Author: zhongkechen

.github/workflows/cloud-tests.yml

@@ -43,7 +43,7 @@ jobs:
             python-prefix: Py314
     steps:
     - uses: actions/checkout@v7
-    
+
     - name: Setup Python
       uses: actions/setup-python@v6
       with:

packages/aws-durable-execution-sdk-python-examples/examples-catalog.json

@@ -585,6 +585,17 @@
       },
       "path": "./src/comprehensive_operations/comprehensive_operations.py"
     },
+    {
+      "name": "Many Operations",
+      "description": "Runs many durable steps and child contexts to exercise long execution history retrieval",
+      "handler": "many_operations.handler",
+      "integration": true,
+      "durableConfig": {
+        "RetentionPeriodInDays": 7,
+        "ExecutionTimeout": 300
+      },
+      "path": "./src/comprehensive_operations/many_operations.py"
+    },
     {
         "name": "Create Callback Concurrency",
         "description": "Demonstrates multiple concurrent createCallback operations using context.parallel",

packages/aws-durable-execution-sdk-python-examples/src/comprehensive_operations/many_operations.py

@@ -0,0 +1,69 @@
+"""Example with many durable step operations."""
+
+from aws_durable_execution_sdk_python.context import (
+    DurableContext,
+    durable_with_child_context,
+)
+from aws_durable_execution_sdk_python.execution import durable_execution
+
+
+TOP_LEVEL_STEP_COUNT = 12
+CHILD_CONTEXT_COUNT = 4
+CHILD_STEPS_PER_CONTEXT = 2
+
+
+def make_step(value: int):
+    return lambda _: value
+
+
+@durable_with_child_context
+def child_group(
+    child_context: DurableContext, group_index: int, steps_per_context: int
+) -> list[int]:
+    values = []
+    for step_index in range(steps_per_context):
+        value = group_index * 100 + step_index
+        values.append(
+            child_context.step(
+                make_step(value),
+                name=f"child-{group_index:02d}-step-{step_index:02d}",
+            )
+        )
+    return values
+
+
+@durable_execution
+def handler(_event, context: DurableContext) -> dict[str, object]:
+    """Run many named step operations across parent and child contexts."""
+
+    step_results = []
+    for index in range(TOP_LEVEL_STEP_COUNT):
+        step_results.append(
+            context.step(
+                make_step(index),
+                name=f"many-step-{index:03d}",
+            )
+        )
+
+    child_results = []
+    for group_index in range(CHILD_CONTEXT_COUNT):
+        child_results.append(
+            context.run_in_child_context(
+                child_group(group_index, CHILD_STEPS_PER_CONTEXT),
+                name=f"child-context-{group_index:02d}",
+            )
+        )
+
+    child_values = [value for child_result in child_results for value in child_result]
+
+    return {
+        "topLevelStepCount": TOP_LEVEL_STEP_COUNT,
+        "childContextCount": CHILD_CONTEXT_COUNT,
+        "childStepCount": CHILD_CONTEXT_COUNT * CHILD_STEPS_PER_CONTEXT,
+        "totalStepCount": TOP_LEVEL_STEP_COUNT
+        + CHILD_CONTEXT_COUNT * CHILD_STEPS_PER_CONTEXT,
+        "firstStep": step_results[0] if step_results else None,
+        "lastStep": step_results[-1] if step_results else None,
+        "topLevelStepTotal": sum(step_results),
+        "childStepTotal": sum(child_values),
+    }

packages/aws-durable-execution-sdk-python-examples/test/comprehensive_operations/test_many_operations.py

@@ -0,0 +1,70 @@
+"""Tests for many_operations."""
+
+import pytest
+from aws_durable_execution_sdk_python.execution import InvocationStatus
+
+from src.comprehensive_operations import many_operations
+from test.conftest import deserialize_operation_payload
+
+
+TOP_LEVEL_STEP_COUNT = 12
+CHILD_CONTEXT_COUNT = 4
+CHILD_STEPS_PER_CONTEXT = 2
+TOTAL_STEP_COUNT = TOP_LEVEL_STEP_COUNT + CHILD_CONTEXT_COUNT * CHILD_STEPS_PER_CONTEXT
+
+
+@pytest.mark.example
+@pytest.mark.durable_execution(
+    handler=many_operations.handler,
+    lambda_function_name="Many Operations",
+)
+def test_fetches_many_steps_and_child_context_operations(durable_runner):
+    """Verify result history includes many steps and child context operations."""
+    with durable_runner:
+        result = durable_runner.run(input=None, timeout=60)
+
+    assert result.status is InvocationStatus.SUCCEEDED
+
+    result_data = deserialize_operation_payload(result.result)
+    assert result_data == {
+        "topLevelStepCount": TOP_LEVEL_STEP_COUNT,
+        "childContextCount": CHILD_CONTEXT_COUNT,
+        "childStepCount": CHILD_CONTEXT_COUNT * CHILD_STEPS_PER_CONTEXT,
+        "totalStepCount": TOTAL_STEP_COUNT,
+        "firstStep": 0,
+        "lastStep": TOP_LEVEL_STEP_COUNT - 1,
+        "topLevelStepTotal": sum(range(TOP_LEVEL_STEP_COUNT)),
+        "childStepTotal": sum(
+            group_index * 100 + step_index
+            for group_index in range(CHILD_CONTEXT_COUNT)
+            for step_index in range(CHILD_STEPS_PER_CONTEXT)
+        ),
+    }
+
+    all_ops = result.get_all_operations()
+    top_level_step_ops = [
+        op
+        for op in all_ops
+        if op.operation_type.value == "STEP"
+        and op.name is not None
+        and op.name.startswith("many-step-")
+    ]
+    assert len(top_level_step_ops) == TOP_LEVEL_STEP_COUNT
+
+    child_context_ops = [
+        op
+        for op in all_ops
+        if op.operation_type.value == "CONTEXT"
+        and op.name is not None
+        and op.name.startswith("child-context-")
+    ]
+    assert len(child_context_ops) == CHILD_CONTEXT_COUNT
+
+    child_step_ops = [
+        op
+        for op in all_ops
+        if op.operation_type.value == "STEP"
+        and op.name is not None
+        and op.name.startswith("child-")
+    ]
+    assert len(child_step_ops) == CHILD_CONTEXT_COUNT * CHILD_STEPS_PER_CONTEXT

packages/aws-durable-execution-sdk-python-testing/src/aws_durable_execution_sdk_python_testing/runner.py

@@ -1246,11 +1246,28 @@ def _fetch_execution_history(
         Raises:
             ClientError: If lambda client encounter error
         """
-        history_dict = self.lambda_client.get_durable_execution_history(
-            DurableExecutionArn=execution_arn,
-            IncludeExecutionData=True,
-        )
-        history_response = GetDurableExecutionHistoryResponse.from_dict(history_dict)
+        events: list[Event] = []
+        next_marker: str | None = None
+
+        while True:
+            request = {
+                "DurableExecutionArn": execution_arn,
+                "IncludeExecutionData": True,
+            }
+            if next_marker:
+                request["Marker"] = next_marker
+
+            history_dict = self.lambda_client.get_durable_execution_history(**request)
+            history_response = GetDurableExecutionHistoryResponse.from_dict(
+                history_dict
+            )
+            events.extend(history_response.events)
+
+            next_marker = history_response.next_marker
+            if not next_marker:
+                break
+
+        history_response = GetDurableExecutionHistoryResponse(events=events)
 
         logger.info("Retrieved %d events from history", len(history_response.events))