Metadata fixes (#160)

* Truncate dicom long dicom strings

* Remove FlattenOnLoadNestedDicomField

* Staining substance is either a single string or list of codes

* Fix missing 'allow_none' for json fields

* Do not add empty `CodingSchemeVersion` to concept code name

CHANGELOG.md

@@ -7,6 +7,15 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.20.2] - 2024-02-22
+
+### Fixed
+
+- Fixed handling of string values longer than allowed by value representation.
+- Corrected staining substances to be either a single string or a list of codes instead of a list of strings or codes.
+- Added `allow_none` for json metadata fields where it was missing.
+- Do not add empty `CodingSchemeVersion` to concept code name.
+
 ## [0.20.1] - 2024-02-22
 
 ### Fixed
@@ -344,7 +353,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Initial release of wsidicom
 
-[Unreleased]: https://github.com/imi-bigpicture/wsidicom/compare/0.20.1..HEAD
+[Unreleased]: https://github.com/imi-bigpicture/wsidicom/compare/0.20.2..HEAD
+[0.20.2]: https://github.com/imi-bigpicture/wsidicom/compare/v0.20.1..v0.20.2
 [0.20.1]: https://github.com/imi-bigpicture/wsidicom/compare/v0.20.0..v0.20.1
 [0.20.0]: https://github.com/imi-bigpicture/wsidicom/compare/v0.19.1..v0.20.0
 [0.19.1]: https://github.com/imi-bigpicture/wsidicom/compare/v0.19.0..v0.19.1

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "wsidicom"
-version = "0.20.1"
+version = "0.20.2"
 description = "Tools for handling DICOM based whole scan images"
 authors = ["Erik O Gabrielsson <[email protected]>"]
 license = "Apache-2.0"

tests/metadata/conftest.py

@@ -345,12 +345,21 @@ def processing(
 
 
 @pytest.fixture()
-def staining(date_time: datetime.datetime, description: str):
+def substances():
+    yield [
+        SpecimenStainsCode("hematoxylin stain"),
+        SpecimenStainsCode("water soluble eosin stain"),
+    ]
+
+
+@pytest.fixture()
+def staining(
+    substances: Union[str, Sequence[SpecimenStainsCode]],
+    date_time: datetime.datetime,
+    description: str,
+):
     yield Staining(
-        [
-            SpecimenStainsCode("hematoxylin stain"),
-            SpecimenStainsCode("water soluble eosin stain"),
-        ],
+        substances,
         date_time=date_time,
         description=description,
     )
@@ -421,7 +430,7 @@ def slide_sample(sample: Sample):
 
 
 @pytest.fixture()
-def slide(slide_identifier: Union[str, SpecimenIdentifier]):
+def slide(slide_identifier: Union[str, SpecimenIdentifier], staining: Staining):
     part_1 = Specimen(
         "part 1",
         Collection(
@@ -503,18 +512,8 @@ def slide(slide_identifier: Union[str, SpecimenIdentifier]):
         SampleLocalization(description="right"),
     )
 
-    stainings = [
-        Staining(
-            [
-                SpecimenStainsCode("hematoxylin stain"),
-                SpecimenStainsCode("water soluble eosin stain"),
-            ],
-            date_time=datetime.datetime(2023, 8, 5),
-        ),
-    ]
-
     yield Slide(
-        identifier=slide_identifier, stainings=stainings, samples=[sample_1, sample_2]
+        identifier=slide_identifier, stainings=[staining], samples=[sample_1, sample_2]
     )
 
 

tests/metadata/dicom_schema/helpers.py

@@ -120,10 +120,7 @@ def assert_dicom_code_dataset_equals_code(
     if expected_code.scheme_version is not None:
         assert code_dataset.CodingSchemeVersion == expected_code.scheme_version
     else:
-        assert (
-            "CodingSchemeVersion" not in code_dataset
-            or code_dataset.CodingSchemeVersion is None
-        )
+        assert "CodingSchemeVersion" not in code_dataset
 
 
 def assert_dicom_code_sequence_equals_codes(
@@ -602,12 +599,17 @@ def create_staining_dataset(
 ):
     dataset = Dataset()
     items = create_initial_common_preparation_step_items(staining_dicom, identifier)
-    for substance in staining_dicom.substances:
+    if isinstance(staining_dicom.substances, str):
         items.append(
-            create_code_item(SampleCodes.using_substance, substance)
-            if isinstance(substance, SpecimenStainsCode)
-            else create_string_item(SampleCodes.using_substance, substance)
+            create_string_item(SampleCodes.using_substance, staining_dicom.substances)
         )
+    else:
+        for substance in staining_dicom.substances:
+            items.append(
+                create_code_item(SampleCodes.using_substance, substance)
+                if isinstance(substance, SpecimenStainsCode)
+                else create_string_item(SampleCodes.using_substance, substance)
+            )
     items.extend(create_final_common_preparation_step_items(staining_dicom))
     dataset.SpecimenPreparationStepContentItemSequence = items
     return dataset
@@ -647,7 +649,7 @@ def create_specimen_preparation_sequence(
     block_sampling_method: SpecimenSamplingProcedureCode,
     block_type: AnatomicPathologySpecimenTypesCode,
     sampling_location: SamplingLocation,
-    stains: Sequence[SpecimenStainsCode],
+    stains: Union[str, Sequence[SpecimenStainsCode]],
     specimen_container: Optional[ContainerTypeCode] = None,
     block_container: Optional[ContainerTypeCode] = None,
 ):
@@ -706,7 +708,7 @@ def create_specimen_preparation_sequence(
     staining = create_staining_dataset(
         StainingDicomModel(
             identifier=slide_sample_id,
-            substances=list(stains),
+            substances=stains,
         ),
         identifier=slide_sample_id,
     )
@@ -831,7 +833,8 @@ def create_code_dataset(code: Union[Code, ConceptCode]):
     dataset.CodeValue = code.value
     dataset.CodingSchemeDesignator = code.scheme_designator
     dataset.CodeMeaning = code.meaning
-    dataset.CodingSchemeVersion = code.scheme_version
+    if code.scheme_version is not None and code.scheme_version != "":
+        dataset.CodingSchemeVersion = code.scheme_version
     return dataset
 
 

tests/metadata/dicom_schema/sample/test_dicom_sample.py

@@ -12,7 +12,7 @@
 #    See the License for the specific language governing permissions and
 #    limitations under the License.
 
-from typing import List, Optional, Sequence
+from typing import List, Optional, Sequence, Union
 
 import pytest
 from pydicom import Dataset
@@ -105,6 +105,16 @@ class TestSampleDicom:
     @pytest.mark.parametrize(
         "block_container", [None, ContainerTypeCode("Tissue cassette")]
     )
+    @pytest.mark.parametrize(
+        "substances",
+        [
+            "HE",
+            [
+                SpecimenStainsCode("hematoxylin stain"),
+                SpecimenStainsCode("water soluble eosin stain"),
+            ],
+        ],
+    )
     def test_slide_sample_from_dataset(
         self,
         slide_sample_ids: Sequence[str],
@@ -121,7 +131,7 @@ def test_slide_sample_from_dataset(
         sample_localization: SampleLocalization,
         sampling_location: SamplingLocation,
         primary_anatomic_structures: Sequence[Code],
-        stains: Sequence[SpecimenStainsCode],
+        substances: Union[str, Sequence[SpecimenStainsCode]],
         short_description: Optional[str],
         detailed_description: Optional[str],
         specimen_container: Optional[ContainerTypeCode],
@@ -145,7 +155,7 @@ def test_slide_sample_from_dataset(
                 block_sampling_method,
                 block_type,
                 sampling_location,
-                stains,
+                substances,
                 specimen_container,
                 block_container,
             )
@@ -173,6 +183,9 @@ def test_slide_sample_from_dataset(
         # Assert
         assert slide_samples is not None
         assert stainings is not None
+        for staining in stainings:
+            assert isinstance(staining, Staining)
+            assert staining.substances == substances
         assert len(slide_samples) == len(slide_sample_ids)
         for slide_sample_index, slide_sample in enumerate(slide_samples):
             assert isinstance(slide_sample, SlideSample)
@@ -246,6 +259,16 @@ def test_slide_sample_from_dataset(
     @pytest.mark.parametrize(
         "block_container", [None, ContainerTypeCode("Tissue cassette")]
     )
+    @pytest.mark.parametrize(
+        "substances",
+        [
+            "HE",
+            [
+                SpecimenStainsCode("hematoxylin stain"),
+                SpecimenStainsCode("water soluble eosin stain"),
+            ],
+        ],
+    )
     def test_slide_sample_to_dataset(
         self,
         slide_sample_ids: Sequence[str],
@@ -262,7 +285,7 @@ def test_slide_sample_to_dataset(
         sample_localization: SampleLocalization,
         sampling_location: SamplingLocation,
         primary_anatomic_structures: Sequence[Code],
-        stains: Sequence[SpecimenStainsCode],
+        substances: Union[str, Sequence[SpecimenStainsCode]],
         short_description: Optional[str],
         detailed_description: Optional[str],
         specimen_container: Optional[ContainerTypeCode],
@@ -308,7 +331,7 @@ def test_slide_sample_to_dataset(
                 detailed_description=detailed_description,
             )
             slide_samples.append(slide_sample)
-        staining = Staining(stains)
+        staining = Staining(substances)
 
         schema = SpecimenDescriptionDicomSchema()
 
@@ -545,10 +568,15 @@ def test_slide_sample_to_dataset(
                 SampleCodes.processing_type,
                 SampleCodes.staining,
             )
-            for stain in stains:
-                assert_next_item_equals_code(
-                    step_item_iterator, SampleCodes.using_substance, stain
+            if isinstance(substances, str):
+                assert_next_item_equals_string(
+                    step_item_iterator, SampleCodes.using_substance, substances
                 )
+            else:
+                for substance in substances:
+                    assert_next_item_equals_code(
+                        step_item_iterator, SampleCodes.using_substance, substance
+                    )
             assert next(step_item_iterator, None) is None
 
     def test_sampling_from_not_defined_specimen(