Fix logic of determining file run number when previously inserted files are already pushed to S3 and not on filesystem anymore

python/lib/dcm2bids_imaging_pipeline_lib/nifti_insertion_pipeline.py

@@ -404,7 +404,10 @@ def _determine_new_nifti_assembly_rel_path(self):
 
         # determine NIfTI file name
         new_nifti_name = self._construct_nifti_filename(file_bids_entities_dict)
-        while os.path.exists(os.path.join(self.data_dir, new_nifti_rel_dir, new_nifti_name)):
+        already_inserted_filenames = self.imaging_obj.get_list_of_files_already_inserted_for_tarchive_id(
+            self.dicom_archive_obj.tarchive_info_dict["TarchiveID"]
+        )
+        while new_nifti_name in already_inserted_filenames:
             file_bids_entities_dict['run'] += 1
             new_nifti_name = self._construct_nifti_filename(file_bids_entities_dict)
 

python/lib/imaging.py

@@ -881,6 +881,26 @@ def determine_intended_for_field_for_fmap_json_files(self, tarchive_id):
 
         return sorted_fmap_files_dict
 
+    def get_list_of_files_already_inserted_for_tarchive_id(self, tarchive_id):
+        """
+        Get the list of filenames already inserted for a given TarchiveID.
+
+        :param tarchive_id: the Tarchive ID to process
+         :type tarchive_id: int
+
+        :return: a list with file names already inserted in the files table for TarchiveID
+         :rtype: list
+        """
+
+        # get list files from a given tarchive ID
+        results = self.files_db_obj.get_files_inserted_for_tarchive_id(tarchive_id)
+
+        files_list = []
+        for entry in results:
+            files_list.append(os.path.basename(entry['File']))
+
+        return files_list
+
     def get_list_of_fmap_files_sorted_by_acq_time(self, files_list):
         """
         Get the list of fieldmap acquisitions that requires the IntendedFor field in their JSON file.