Import duplication when object type mapping has two identifiers and one is another referenced object

Issue Summary

During object import, objects are duplicated when object type mapping has two identifiers and one is another referenced object

Steps to Reproduce

1. Create object type called "City" in your schema (with just default attributes) 
2. Also create an object type called "Employee" with the default attributes and an attribute of "City" of the City object type created in step 1
3. Create an import configuration in your schema with the attached CSV file
4. After the configuration is created, create an object type mapping within the configuration selecting the Employee object type created in step 2 and set the identifiers of the mapping to be "Name" and "City" and ensure both are ticked (See screenshot for end result of import configuration)
   1. Ensure that for the Employee object type mapping, Missing objects are ignored
5. Execute import and check that the objects were correctly imported
6. Execute import another time without making any changes

Expected Results

We should see no duplicates

Actual Results

We see duplicates of all the objects

Workaround

This seems to only happen if you use "Label" in the import AQL of the referenced object. Using the attribute name instead can fix this. For example:

• Name = ${Placeholder}