Fix problem with lineage and conditional execution

inference/core/workflows/execution_engine/v1/core.py

@@ -17,7 +17,7 @@
     validate_runtime_input,
 )
 
-EXECUTION_ENGINE_V1_VERSION = Version("1.1.0")
+EXECUTION_ENGINE_V1_VERSION = Version("1.1.1")
 
 
 class ExecutionEngineV1(BaseExecutionEngine):

inference/core/workflows/execution_engine/v1/executor/core.py

@@ -154,8 +154,9 @@ def run_simd_step_in_batch_mode(
     step_input = execution_data_manager.get_simd_step_input(step_selector=step_selector)
     if not step_input.indices:
         # no inputs - discarded either by conditional exec or by not accepting empty
-        return None
-    outputs = step_instance.run(**step_input.parameters)
+        outputs = []
+    else:
+        outputs = step_instance.run(**step_input.parameters)
     execution_data_manager.register_simd_step_output(
         step_selector=step_selector,
         indices=step_input.indices,
@@ -175,8 +176,6 @@ def run_simd_step_in_non_batch_mode(
         result = step_instance.run(**input_definition.parameters)
         results.append(result)
         indices.append(input_definition.index)
-    if not indices:
-        return None
     execution_data_manager.register_simd_step_output(
         step_selector=step_selector,
         indices=indices,

tests/inference/hosted_platform_tests/test_workflows.py

@@ -127,7 +127,7 @@ def test_get_versions_of_execution_engine(object_detection_service_url: str) ->
     # then
     response.raise_for_status()
     response_data = response.json()
-    assert response_data["versions"] == ["1.1.0"]
+    assert response_data["versions"] == ["1.1.1"]
 
 
 FUNCTION = """

tests/inference/integration_tests/test_workflow_endpoints.py

@@ -691,7 +691,7 @@ def test_get_versions_of_execution_engine(server_url: str) -> None:
     # then
     response.raise_for_status()
     response_data = response.json()
-    assert response_data["versions"] == ["1.1.0"]
+    assert response_data["versions"] == ["1.1.1"]
 
 
 def test_getting_block_schema_using_get_endpoint(server_url) -> None:

tests/inference/models_predictions_tests/test_florence2.py

@@ -1,6 +1,7 @@
+import numpy as np
 import pytest
+
 from inference.models.florence2 import Florence2
-import numpy as np
 
 
 @pytest.mark.slow

tests/workflows/integration_tests/execution/test_workflow_with_branching_affecting_lineage.py

@@ -0,0 +1,108 @@
+import numpy as np
+
+from inference.core.env import WORKFLOWS_MAX_CONCURRENT_STEPS
+from inference.core.managers.base import ModelManager
+from inference.core.workflows.core_steps.common.entities import StepExecutionMode
+from inference.core.workflows.execution_engine.core import ExecutionEngine
+
+PROBLEMATIC_WORKFLOW = {
+    "version": "1.0",
+    "inputs": [{"type": "WorkflowImage", "name": "image"}],
+    "steps": [
+        {
+            "type": "ObjectDetectionModel",
+            "name": "general_detection",
+            "image": "$inputs.image",
+            "model_id": "yolov8n-640",
+            "class_filter": ["dog"],
+        },
+        {
+            "type": "ContinueIf",
+            "name": "continue_if",
+            "condition_statement": {
+                "type": "StatementGroup",
+                "statements": [
+                    {
+                        "type": "BinaryStatement",
+                        "left_operand": {
+                            "type": "DynamicOperand",
+                            "operand_name": "prediction",
+                            "operations": [{"type": "SequenceLength"}],
+                        },
+                        "comparator": {"type": "(Number) =="},
+                        "right_operand": {
+                            "type": "StaticOperand",
+                            "value": 5,
+                        },
+                    }
+                ],
+            },
+            "evaluation_parameters": {
+                "prediction": "$steps.general_detection.predictions"
+            },
+            "next_steps": ["$steps.cropping"],
+        },
+        {
+            "type": "Crop",
+            "name": "cropping",
+            "image": "$inputs.image",
+            "predictions": "$steps.general_detection.predictions",
+        },
+        {
+            "type": "ClassificationModel",
+            "name": "breds_classification",
+            "image": "$steps.cropping.crops",
+            "model_id": "dog-breed-xpaq6/1",
+        },
+    ],
+    "outputs": [
+        {
+            "type": "JsonField",
+            "name": "predictions",
+            "selector": "$steps.breds_classification.predictions",
+        },
+    ],
+}
+
+
+def test_workflow_with_flow_control_eliminating_step_changing_lineage(
+    model_manager: ModelManager,
+    dogs_image: np.ndarray,
+) -> None:
+    """
+    This test case covers bug that in Execution Engine versions <=1.1.0.
+    The bug was not registering outputs from steps given no inputs provided
+    (inputs may have been filtered by flow-control steps). As a result,
+    given that step changes dimensionality (registers new data lineage) -
+    registration could not happen and downstream steps was raising error:
+    "Lineage ['<workflow_input>', 'XXX'] not found. [...]"
+    """
+    # given
+    workflow_init_parameters = {
+        "workflows_core.model_manager": model_manager,
+        "workflows_core.api_key": None,
+        "workflows_core.step_execution_mode": StepExecutionMode.LOCAL,
+    }
+    execution_engine = ExecutionEngine.init(
+        workflow_definition=PROBLEMATIC_WORKFLOW,
+        init_parameters=workflow_init_parameters,
+        max_concurrent_steps=WORKFLOWS_MAX_CONCURRENT_STEPS,
+    )
+
+    # when
+    result = execution_engine.run(
+        runtime_parameters={
+            "image": [dogs_image],
+        }
+    )
+
+    # then
+    assert isinstance(result, list), "Expected result to be list"
+    assert len(result) == 1, "1 image provided, so 1 output elements expected"
+    assert result[0].keys() == {
+        "predictions"
+    }, "Expected all declared outputs to be delivered for first result"
+    assert len([e for e in result[0]["predictions"] if e]) == 0, (
+        "Expected no predictions, due to conditional execution applied, effectively preventing "
+        "`cropping` step from running"
+    )