fix: More work on the decoder

Author: Alan Christie

tests/test_decoder.py

@@ -142,7 +142,7 @@ def test_validate_schema_for_step_specification_variable_names():
     assert error is None
 
 
-@pytest.mark.skip(reason="DO not support combination atm")
+@pytest.mark.skip(reason="We do not support combination atm")
 def test_validate_schema_for_simple_python_parallel():
     # Arrange
 

tests/test_workflow_engine_examples.py

@@ -208,37 +208,6 @@ def test_workflow_engine_example_smiles_to_file(basic_engine):
     assert project_file_exists(output_file)
 
 
-@pytest.mark.skip("Unrealistic test")
-def test_workflow_engine_shortcut_example_1(basic_engine):
-    # Arrange
-    md, da = basic_engine
-
-    # Make sure files that should be generated by the test
-    # do not exist before we run the test.
-    output_file_a = "a.sdf"
-    assert not project_file_exists(output_file_a)
-    output_file_b = "b.sdf"
-    assert not project_file_exists(output_file_b)
-
-    # Act
-    r_wfid = start_workflow(md, da, "shortcut-example-1", {})
-
-    # Assert
-    wait_for_workflow(da, r_wfid)
-    # Additional, detailed checks...
-    # Check we only have one RunningWorkflowStep, and it succeeded
-    response = da.get_running_workflow_steps(running_workflow_id=r_wfid)
-    assert response["count"] == 2
-    assert response["running_workflow_steps"][0]["done"]
-    assert response["running_workflow_steps"][0]["success"]
-    assert response["running_workflow_steps"][1]["done"]
-    assert response["running_workflow_steps"][1]["success"]
-    # This test should generate a file in the simulated project directory
-    assert project_file_exists(output_file_a)
-    assert project_file_exists(output_file_b)
-
-
-@pytest.mark.skip("temporary skip")
 def test_workflow_engine_simple_python_molprops(basic_engine):
     # Arrange
     md, da = basic_engine
@@ -249,13 +218,7 @@ def test_workflow_engine_simple_python_molprops(basic_engine):
         output="step1.out.smi",
     )
 
-    # Make sure files that should be generated by the test
-    # do not exist before we run the test.
-    output_file_1 = "step1.out.smi"
-    assert not project_file_exists(output_file_1)
-    output_file_2 = "step2.out.smi"
-    assert not project_file_exists(output_file_2)
-    # And create the test's input file.
+    # Create the test's input file.
     input_file_1 = "input1.smi"
     input_file_1_content = """O=C(CSCc1ccc(Cl)s1)N1CCC(O)CC1
         RDKit          3D
@@ -307,9 +270,22 @@ def test_workflow_engine_simple_python_molprops(basic_engine):
     ) as input_file:
         input_file.writelines(input_file_1_content)
 
+    # Make sure files that should be generated by the test
+    # do not exist before we run the test.
+    output_file_1 = "results.smi"
+    assert not project_file_exists(output_file_1)
+    output_file_2 = "clustered-results.smi"
+    assert not project_file_exists(output_file_2)
+
     # Act
     r_wfid = start_workflow(
-        md, da, "simple-python-molprops", {"candidateMolecules": input_file_1}
+        md,
+        da,
+        "simple-python-molprops",
+        {
+            "candidateMolecules": input_file_1,
+            "clusteredMolecules": "clustered-results.smi",
+        },
     )
 
     # Assert
@@ -322,12 +298,10 @@ def test_workflow_engine_simple_python_molprops(basic_engine):
     assert response["running_workflow_steps"][0]["success"]
     assert response["running_workflow_steps"][1]["done"]
     assert response["running_workflow_steps"][1]["success"]
-    # This test should generate a file in the simulated project directory
-    assert project_file_exists(output_file_1)
+    # This test should generate the expected file in the simulated project directory
     assert project_file_exists(output_file_2)
 
 
-@pytest.mark.skip("Unrealistic test")
 def test_workflow_engine_simple_python_molprops_with_options(basic_engine):
     # Arrange
     md, da = basic_engine
@@ -424,7 +398,6 @@ def test_workflow_engine_simple_python_molprops_with_options(basic_engine):
     assert project_file_exists(output_file_2)
 
 
-@pytest.mark.skip("Unrealistic test")
 def test_workflow_engine_simple_python_fanout(basic_engine):
     # Arrange
     md, da = basic_engine

tests/wapi_adapter.py

@@ -202,7 +202,6 @@ def get_running_workflow_step_by_name(
         for rwfs_id, record in running_workflow_step.items():
             if record["running_workflow"]["id"] != running_workflow_id:
                 continue
-            print("running wf step by name, record:", record)
             if record["name"] == name and record["replica"] == replica:
                 response = record
                 response["id"] = rwfs_id
@@ -425,11 +424,6 @@ def get_running_workflow_step_output_values_for_output(
             mock_output = Unpickler(pickle_file).load()
         UnitTestWorkflowAPIAdapter.lock.release()
 
-        print("mock output", mock_output)
-        print("step", step)
-        print("step_name", step_name)
-        # mock output {'first-step': {'output_variable': 'results', 'output': ['chunk_1.smi', 'chunk_2.smi']}}
-
         if step_name not in mock_output:
             return {"output": []}, 0
         # The record's output variable must match (there's only one record per step atm)

tests/workflow-definitions/simple-python-molprops.yaml

@@ -14,6 +14,7 @@ steps:
     variables:
       name: "col1"
       value: 123
+      outputFile: "results.smi"
   variable-mapping:
   - variable: inputFile
     from-workflow:

workflow/decoder.py

@@ -137,10 +137,10 @@ def get_step_prior_step_variable_mapping(
                 # Tuple is "from" -> "to"
                 if step_name in variable_mapping:
                     variable_mapping[step_name].append(
-                        (v_map["variable"], step_variable)
+                        (step_variable, v_map["variable"])
                     )
                 else:
-                    variable_mapping[step_name] = [(v_map["variable"], step_variable)]
+                    variable_mapping[step_name] = [(step_variable, v_map["variable"])]
     return variable_mapping