Merge pull request #161 from AllenNeuralDynamics/release-v0.18.2

Release v0.18.2

src/aind_metadata_mapper/__init__.py

@@ -1,3 +1,3 @@
 """Init package"""
 
-__version__ = "0.18.1"
+__version__ = "0.18.2"

src/aind_metadata_mapper/gather_metadata.py

@@ -23,6 +23,7 @@
 from aind_data_schema.core.subject import Subject
 from aind_data_schema_models.pid_names import PIDName
 from pydantic import ValidationError
+from pydantic_core import PydanticSerializationError
 
 from aind_metadata_mapper.bergamo.models import (
     JobSettings as BergamoSessionJobSettings,
@@ -343,12 +344,12 @@ def get_instrument_metadata(self) -> Optional[dict]:
         else:
             return None
 
-    def get_main_metadata(self) -> Metadata:
-        """Get main Metadata model"""
+    def get_main_metadata(self) -> dict:
+        """Get serialized main Metadata model"""
 
         def load_model(
             filepath: Optional[Path], model: Type[AindCoreModel]
-        ) -> Optional[AindCoreModel]:
+        ) -> Optional[dict]:
             """
             Validates contents of file with an AindCoreModel
             Parameters
@@ -358,16 +359,25 @@ def load_model(
 
             Returns
             -------
-            Optional[AindCodeModel]
+            Optional[dict]
 
             """
             if filepath is not None and filepath.is_file():
                 with open(filepath, "r") as f:
                     contents = json.load(f)
                 try:
-                    output = model.model_validate_json(json.dumps(contents))
-                except (ValidationError, AttributeError, ValueError, KeyError):
-                    output = model.model_construct(**contents)
+                    valid_model = model.model_validate_json(
+                        json.dumps(contents)
+                    )
+                    output = json.loads(valid_model.model_dump_json())
+                except (
+                    ValidationError,
+                    AttributeError,
+                    ValueError,
+                    KeyError,
+                    PydanticSerializationError,
+                ):
+                    output = contents
 
                 return output
             else:
@@ -410,22 +420,26 @@ def load_model(
                 acquisition=acquisition,
                 instrument=instrument,
             )
-            return metadata
+            metadata_json = json.loads(metadata.model_dump_json(by_alias=True))
+            return metadata_json
         except Exception as e:
             logging.warning(f"Issue with metadata construction! {e.args}")
-            metadata = Metadata.model_construct(
+            # Set basic parameters
+            metadata = Metadata(
                 name=self.settings.metadata_settings.name,
                 location=self.settings.metadata_settings.location,
-                subject=subject,
-                data_description=data_description,
-                procedures=procedures,
-                session=session,
-                rig=rig,
-                processing=processing,
-                acquisition=acquisition,
-                instrument=instrument,
             )
-            return metadata
+            metadata_json = json.loads(metadata.model_dump_json(by_alias=True))
+            # Attach dict objects
+            metadata_json["subject"] = subject
+            metadata_json["data_description"] = data_description
+            metadata_json["procedures"] = procedures
+            metadata_json["session"] = session
+            metadata_json["rig"] = rig
+            metadata_json["processing"] = processing
+            metadata_json["acquisition"] = acquisition
+            metadata_json["instrument"] = instrument
+            return metadata_json
 
     def _write_json_file(self, filename: str, contents: dict) -> None:
         """
@@ -504,14 +518,13 @@ def run_job(self) -> None:
         self._gather_automated_metadata()
         self._gather_non_automated_metadata()
         if self.settings.metadata_settings is not None:
-            metadata = self.get_main_metadata()
+            contents = self.get_main_metadata()
             # TODO: may need to update aind-data-schema write standard file
             #  class
             output_path = (
                 self.settings.directory_to_write_to
                 / Metadata.default_filename()
             )
-            contents = json.loads(metadata.model_dump_json(by_alias=True))
             with open(output_path, "w") as f:
                 json.dump(
                     contents,

src/aind_metadata_mapper/models.py

@@ -25,7 +25,7 @@
 )
 
 
-class SessionSettings(BaseSettings, extra='allow'):
+class SessionSettings(BaseSettings, extra="allow"):
     """Settings needed to retrieve session metadata"""
 
     job_settings: Annotated[
@@ -39,29 +39,29 @@ class SessionSettings(BaseSettings, extra='allow'):
     ]
 
 
-class AcquisitionSettings(BaseSettings, extra='allow'):
+class AcquisitionSettings(BaseSettings, extra="allow"):
     """Fields needed to retrieve acquisition metadata"""
 
     # TODO: we can change this to a tagged union once more acquisition settings
     #  are added
     job_settings: SmartSpimAcquisitionJobSettings
 
 
-class SubjectSettings(BaseSettings, extra='allow'):
+class SubjectSettings(BaseSettings, extra="allow"):
     """Fields needed to retrieve subject metadata"""
 
     subject_id: str
     metadata_service_path: str = "subject"
 
 
-class ProceduresSettings(BaseSettings, extra='allow'):
+class ProceduresSettings(BaseSettings, extra="allow"):
     """Fields needed to retrieve procedures metadata"""
 
     subject_id: str
     metadata_service_path: str = "procedures"
 
 
-class RawDataDescriptionSettings(BaseSettings, extra='allow'):
+class RawDataDescriptionSettings(BaseSettings, extra="allow"):
     """Fields needed to retrieve data description metadata"""
 
     name: str
@@ -71,7 +71,7 @@ class RawDataDescriptionSettings(BaseSettings, extra='allow'):
     metadata_service_path: str = "funding"
 
 
-class ProcessingSettings(BaseSettings, extra='allow'):
+class ProcessingSettings(BaseSettings, extra="allow"):
     """Fields needed to retrieve processing metadata"""
 
     pipeline_process: dict = Field(
@@ -83,7 +83,7 @@ class ProcessingSettings(BaseSettings, extra='allow'):
     )
 
 
-class MetadataSettings(BaseSettings, extra='allow'):
+class MetadataSettings(BaseSettings, extra="allow"):
     """Fields needed to retrieve main Metadata"""
 
     name: str
@@ -98,7 +98,7 @@ class MetadataSettings(BaseSettings, extra='allow'):
     instrument_filepath: Optional[Path] = None
 
 
-class JobSettings(BaseSettings, extra='allow'):
+class JobSettings(BaseSettings, extra="allow"):
     """Fields needed to gather all metadata"""
 
     job_settings_name: Literal["GatherMetadata"] = "GatherMetadata"