Empty elements are no longer removed when deserializing a BoM file (#546

)

Co-authored-by: pyansys-ci-bot <pyansys.github.bot@ansys.com>

doc/changelog.d/546.fixed.md

@@ -0,0 +1 @@
+Empty elements are no longer removed when deserializing a BoM file

src/ansys/grantami/bomanalytics/_bom_helper.py

@@ -21,7 +21,7 @@
 # SOFTWARE.
 
 from pathlib import Path
-from typing import TYPE_CHECKING, Tuple, cast
+from typing import TYPE_CHECKING, Any, Dict, List, TextIO, Tuple, Union, cast
 
 import xmlschema
 from xmlschema import XMLSchema
@@ -63,7 +63,7 @@ def load_bom_from_file(self, file_path: Path) -> "BillOfMaterials":
         :class:`~._bom_types.BillOfMaterials`
         """
         with open(file_path, "r", encoding="utf8") as fp:
-            obj, errors = cast(Tuple, self._schema.decode(fp, validation="lax"))
+            obj, errors = self._deserialize_bom(fp)
 
         if len(errors) > 0:
             newline = "\n"
@@ -86,7 +86,7 @@ def load_bom_from_text(self, bom_text: str) -> "BillOfMaterials":
         -------
         :class:`~._bom_types.BillOfMaterials`
         """
-        obj, errors = cast(Tuple, self._schema.decode(bom_text, validation="lax", keep_empty=True))
+        obj, errors = self._deserialize_bom(bom_text)
 
         if len(errors) > 0:
             newline = "\n"
@@ -96,6 +96,23 @@ def load_bom_from_text(self, bom_text: str) -> "BillOfMaterials":
 
         return self._reader.read_bom(obj)
 
+    def _deserialize_bom(self, bom: Union[TextIO, str]) -> Tuple[Dict[str, Any], List]:
+        """
+        Deserialize either a string or I/O stream BoM.
+
+        Parameters
+        ----------
+        bom : Union[TextIO, str]
+            Object containing an XML representation of a BoM, either as text or I/O stream.
+
+        Returns
+        -------
+        Tuple[Dict[str, Any], List]
+            A tuple of the deserialized dictionary and a list of errors.
+        """
+        result = self._schema.decode(bom, validation="lax", keep_empty=True, xmlns_processing="collapsed")
+        return cast(Tuple[Dict[str, Any], List], result)
+
     def dump_bom(self, bom: "BillOfMaterials") -> str:
         """
         Convert a BillOfMaterials object into a string XML representation.

tests/inputs/__init__.py

@@ -28,25 +28,25 @@
 inputs_dir = pathlib.Path(__file__).parent
 
 _sample_bom_1711_path = inputs_dir / "bom.xml"
-with open(_sample_bom_1711_path, "r") as f:
+with open(_sample_bom_1711_path, "r", encoding="utf8") as f:
     sample_bom_1711 = f.read()
 
 _sample_compliance_bom_1711_path = (
     repository_root / "examples" / "3_Advanced_Topics" / "supporting-files" / "bom-complex.xml"
 )
-with open(_sample_compliance_bom_1711_path, "r") as f:
+with open(_sample_compliance_bom_1711_path, "r", encoding="utf8") as f:
     sample_compliance_bom_1711 = f.read()
 
 sample_bom_custom_db = sample_compliance_bom_1711.replace(
     "MI_Restricted_Substances", "MI_Restricted_Substances_Custom_Tables"
 )
 
-_sample_sustainability_bom_2301_path = (
+sample_sustainability_bom_2301_path = (
     repository_root / "examples" / "4_Sustainability" / "supporting-files" / "bom-2301-assembly.xml"
 )
-with open(_sample_sustainability_bom_2301_path, "r") as f:
+with open(sample_sustainability_bom_2301_path, "r", encoding="utf8") as f:
     sample_sustainability_bom_2301 = f.read()
 
-_large_bom_2301_path = inputs_dir / "medium-test-bom.xml"
-with open(_large_bom_2301_path, "r") as f:
+large_bom_2301_path = inputs_dir / "medium-test-bom.xml"
+with open(large_bom_2301_path, "r", encoding="utf8") as f:
     large_bom_2301 = f.read()

tests/inputs/bom.xml

@@ -24,7 +24,7 @@
                 <Part>
                     <Quantity Unit="Each">1.0</Quantity>
                     <MassPerUom Unit="kg/Part">2.0</MassPerUom>
-                    <PartNumber>3333</PartNumber>
+                    <PartNumber />
                     <Name>Part Two</Name>
                     <Materials>
                         <Material>

tests/test_bom_handler.py

@@ -30,7 +30,13 @@
 
 from ansys.grantami.bomanalytics import BoMHandler, bom_types
 
-from .inputs import large_bom_2301, sample_bom_1711, sample_sustainability_bom_2301
+from .inputs import (
+    large_bom_2301,
+    large_bom_2301_path,
+    sample_bom_1711,
+    sample_sustainability_bom_2301,
+    sample_sustainability_bom_2301_path,
+)
 
 
 class _TestableBoMHandler(BoMHandler):
@@ -80,33 +86,54 @@ def _compare_boms(*, source_bom: str, result_bom: str):
             output_lines.append(diff_item)
         return output_lines
 
+    def test_roundtrip_from_text_with_assertions(self):
+        bom_handler = _TestableBoMHandler(
+            default_namespace=self._default_namespace, namespace_mapping=self._namespace_map
+        )
+        deserialized_bom = bom_handler.load_bom_from_text(large_bom_2301)
+        output_bom = bom_handler.dump_bom(deserialized_bom)
+
+        diff = self._compare_boms(source_bom=large_bom_2301, result_bom=output_bom)
+
+        assert len(diff) == 0, "\n".join(diff)
+
     @pytest.mark.parametrize(
         "input_bom",
         [
             pytest.param(large_bom_2301, id="large_bom"),
+            pytest.param(sample_sustainability_bom_2301, id="sustainability_bom"),
         ],
     )
-    def test_roundtrip_with_assertions(self, input_bom: str):
+    def test_roundtrip_from_text_parsing_succeeds(self, input_bom: str):
+        bom_handler = BoMHandler()
+        deserialized_bom = bom_handler.load_bom_from_text(input_bom)
+
+        rendered_bom = bom_handler.dump_bom(deserialized_bom)
+        deserialized_bom_roundtriped = bom_handler.load_bom_from_text(rendered_bom)
+
+        assert deserialized_bom == deserialized_bom_roundtriped
+
+    def test_roundtrip_from_file_with_assertions(self):
         bom_handler = _TestableBoMHandler(
             default_namespace=self._default_namespace, namespace_mapping=self._namespace_map
         )
-        deserialized_bom = bom_handler.load_bom_from_text(input_bom)
+        deserialized_bom = bom_handler.load_bom_from_file(large_bom_2301_path)
         output_bom = bom_handler.dump_bom(deserialized_bom)
 
-        diff = self._compare_boms(source_bom=input_bom, result_bom=output_bom)
+        diff = self._compare_boms(source_bom=large_bom_2301, result_bom=output_bom)
 
         assert len(diff) == 0, "\n".join(diff)
 
     @pytest.mark.parametrize(
         "input_bom",
         [
-            pytest.param(large_bom_2301, id="large_bom"),
-            pytest.param(sample_sustainability_bom_2301, id="sustainability_bom"),
+            pytest.param(large_bom_2301_path, id="large_bom"),
+            pytest.param(sample_sustainability_bom_2301_path, id="sustainability_bom"),
         ],
     )
-    def test_roundtrip_parsing_succeeds(self, input_bom: str):
+    def test_roundtrip_from_file_parsing_succeeds(self, input_bom: str):
         bom_handler = BoMHandler()
-        deserialized_bom = bom_handler.load_bom_from_text(input_bom)
+        deserialized_bom = bom_handler.load_bom_from_file(input_bom)
 
         rendered_bom = bom_handler.dump_bom(deserialized_bom)
         deserialized_bom_roundtriped = bom_handler.load_bom_from_text(rendered_bom)
@@ -175,7 +202,7 @@ def get_field(self, obj: bom_types.BaseType, p_path: str) -> Any:
             ("components[0]/components[1]/quantity/value", pytest.approx(1.0)),
             ("components[0]/components[1]/mass_per_unit_of_measure/unit", "kg/Part"),
             ("components[0]/components[1]/mass_per_unit_of_measure/value", pytest.approx(2.0)),
-            ("components[0]/components[1]/part_number", "3333"),
+            ("components[0]/components[1]/part_number", ""),
             ("components[0]/components[1]/part_name", "Part Two"),
             ("components[0]/components[1]/materials[0]/percentage", pytest.approx(80.0)),
             ("components[0]/components[1]/materials[0]/mi_material_reference/db_key", "MI_Restricted_Substances"),