Don't infer file size right away

Importing the first 10 studies on IDC has been taking around 20 minutes. About 65% of that time has been spent on inferring file sizes. Even though we don't stream the file data, the request from which we get the content length must be taking some time on the server. Skip doing this for now. We can add it back in if we figure out a way to make it much faster. Signed-off-by: Patrick Avery <[email protected]>
girder · Feb 6, 2024 · f0e3617 · f0e3617
1 parent 8713390
commit f0e3617
Showing 1 changed file with 3 additions and 2 deletions.
diff --git a/sources/dicom/large_image_source_dicom/assetstore/dicomweb_assetstore_adapter.py b/sources/dicom/large_image_source_dicom/assetstore/dicomweb_assetstore_adapter.py
@@ -503,8 +503,9 @@ def importData(self, parent, parentType, params, progress, user, **kwargs):
                 }
                 file['imported'] = True
 
-                # Try to infer the file size without streaming, if possible.
-                file['size'] = self._infer_file_size(file)
+                # Inferring the file size can take a long time, so don't
+                # do it right away, unless we figure out a way to make it faster.
+                # file['size'] = self._infer_file_size(file)
                 file = File().save(file)
 
             items.append(item)