Investigate warning

TODO.md

@@ -122,7 +122,6 @@
   - <https://playwright.dev/python/docs/auth>
   - Unfortunately, we can't use live_server fixture inside session fixtures
   - example <https://github.com/automationneemo/PlaywrightDemoYt>
-- Get rid of dicom_connector.download_study/move_study. Do everything at the series level. That way filtering series (e.g. exlcude modalities) is much easier.
 - Evaluate if services should be better restarted with pywatchman instead of watchdog and watchmedo
   - pywatchman is used by Django autoreload
   - See <https://github.com/django/django/blob/main/django/utils/autoreload.py>

adit/batch_query/processors.py

@@ -1,4 +1,3 @@
-from django.conf import settings
 from django.template.defaultfilters import pluralize
 
 from adit.core.errors import DicomError
@@ -97,15 +96,7 @@ def _find_studies(self, patient_id: str) -> list[ResultDataset]:
         end_date = self.query_task.study_date_end
         study_date = (start_date, end_date)
 
-        modalities_query: list[str] = []
-        if self.query_task.modalities:
-            modalities_query = [
-                modality
-                for modality in self.query_task.modalities
-                if modality not in settings.EXCLUDED_MODALITIES
-            ]
-
-        if not modalities_query:
+        if not self.query_task.modalities:
             study_results = list(
                 self.operator.find_studies(
                     QueryDataset.create(
@@ -121,7 +112,9 @@ def _find_studies(self, patient_id: str) -> list[ResultDataset]:
         else:
             seen: set[str] = set()
             study_results: list[ResultDataset] = []
-            for modality in modalities_query:
+            for modality in self.query_task.modalities:
+                # ModalitiesInStudy does not support to query multiple modalities at once,
+                # so we have to query them one by one.
                 studies = list(
                     self.operator.find_studies(
                         QueryDataset.create(

adit/core/processors.py

@@ -151,6 +151,7 @@ def _download_to_folder(
         download_folder: Path,
     ) -> Path:
         pseudonym = self.transfer_task.pseudonym
+
         if pseudonym:
             patient_folder = download_folder / sanitize_filename(pseudonym)
         else:
@@ -160,9 +161,11 @@ def _download_to_folder(
         study = self._find_study()
         modalities = study.ModalitiesInStudy
 
-        modalities = [
-            modality for modality in modalities if modality not in settings.EXCLUDED_MODALITIES
-        ]
+        exclude_modalities = settings.EXCLUDE_MODALITIES
+        if pseudonym and exclude_modalities:
+            # When we download the study we will exclude the specified modalities, but only
+            # if we are pseudonymizing the study, see also _download_study().
+            modalities = [modality for modality in modalities if modality not in exclude_modalities]
 
         # If some series are explicitly chosen then check if their Series Instance UIDs
         # are correct and only use those modalities for the name of the study folder.
@@ -310,16 +313,64 @@ def callback(ds: Dataset | None) -> None:
             file_path = final_folder / file_name
             write_dataset(ds, file_path)
 
+        pseudonymize = bool(self.transfer_task.pseudonym)
+        exclude_modalities = settings.EXCLUDE_MODALITIES
+
         if series_uids:
+            # If specific series are selected we transfer only those series. When pseudonymizing
+            # we have to check if a modality should be excluded.
+            if pseudonymize and exclude_modalities:
+                filtered_series = []
+                for series_uid in series_uids:
+                    series_list = list(
+                        self.source_operator.find_series(
+                            QueryDataset.create(
+                                PatientID=patient_id,
+                                StudyInstanceUID=study_uid,
+                                SeriesInstanceUID=series_uid,
+                            )
+                        )
+                    )
+                    if not series_list:
+                        logger.warning(f"Series with UID {series_uid} not found.")
+                        continue
+
+                    assert len(series_list) == 1
+                    series = series_list[0]
+
+                    if series.Modality in exclude_modalities:
+                        continue
+
+                    filtered_series.append(series)
+
+                series_uids = [series.SeriesInstanceUID for series in filtered_series]
+
             for series_uid in series_uids:
                 self.source_operator.fetch_series(
                     patient_id=patient_id,
                     study_uid=study_uid,
                     series_uid=series_uid,
                     callback=callback,
                 )
+
+        elif pseudonymize:
+            # If the whole study should be transferred and pseudonymized, we transfer on the
+            # series level to exclude the specified modalities.
+            series_list = list(
+                self.source_operator.find_series(
+                    QueryDataset.create(PatientID=patient_id, StudyInstanceUID=study_uid)
+                )
+            )
+            for series in series_list:
+                series_uid = series.SeriesInstanceUID
+                modality = series.Modality
+                if modality in settings.EXCLUDE_MODALITIES:
+                    continue
+
+                self.source_operator.fetch_series(patient_id, study_uid, series_uid, callback)
+
         else:
-            # If no series are explicitly chosen then download all series of the study
+            # Without pseudonymization we transfer the whole study as it is.
             self.source_operator.fetch_study(
                 patient_id=patient_id,
                 study_uid=study_uid,

adit/core/templatetags/core_extras.py

@@ -1,6 +1,5 @@
 import logging
 
-from django.conf import settings
 from django.template import Library
 
 from ..models import DicomJob, DicomTask
@@ -19,12 +18,6 @@ def person_name_from_dicom(value: str) -> str:
     return value.replace("^", ", ")
 
 
-@register.simple_tag
-def filter_modalities(modalities: list[str]) -> list[str]:
-    exclude_modalities = settings.EXCLUDED_MODALITIES
-    return [modality for modality in modalities if modality not in exclude_modalities]
-
-
 @register.filter
 def dicom_job_status_css_class(status: DicomJob.Status) -> str:
     css_classes = {

adit/core/utils/dicom_dataset.py

@@ -180,6 +180,7 @@ def create(
         SeriesDescription: str | type[_NoValue] = _NoValue,
         SeriesNumber: int | str | type[_NoValue] = _NoValue,
         Modality: str | type[_NoValue] = _NoValue,
+        SOPInstanceUID: str | type[_NoValue] = _NoValue,
     ) -> "QueryDataset":
         """A helper factory method for type hinting query parameters."""
         ds = Dataset()