Why these changes are being introduced:

It was discovered that some ASpace records after transformation contained multiple identifiers, of the same type,
which was throwing errors in Bento.  It looks as though ASpace records may have changed slightly and are
exposing a new string for the <unitid> element, which is the same as the OAI record identifier.

How this addresses that need:

When iterating through <unitid> elements, by skipping those with attribute type=aspace_uri, we can
skip adding that particular <unitid> as a TIMDEX record identifier.

Side effects of this change:

Believed to be none.

Relevant ticket(s):

https://mitlibraries.atlassian.net/browse/TIMX-234

tests/fixtures/ead/ead_record_attribute_and_subfield_variations.xml

@@ -59,6 +59,9 @@
                         <unitid>
                             <emph>Data enclosed in subelement</emph>
                         </unitid>
+                        <unitid type="aspace_uri">
+                            <emph>unitid-that-should-not-be-identifier</emph>
+                        </unitid>
                         <abstract>
                             <emph>Data enclosed in subelement</emph>
                         </abstract>

tests/test_ead.py

@@ -526,6 +526,15 @@ def test_ead_record_invalid_date_and_date_range_are_omitted(caplog):
     ) in caplog.text
 
 
+def test_ead_record_correct_identifiers_from_multiple_unitid(caplog):
+    ead_xml_records = parse_xml_records(
+        "tests/fixtures/ead/ead_record_attribute_and_subfield_variations.xml"
+    )
+    output_record = next(Ead("aspace", ead_xml_records))
+    for identifier in output_record.identifiers:
+        assert identifier.value != "unitid-that-should-not-be-identifier"
+
+
 def test_ead_record_with_missing_optional_fields_transforms_correctly():
     ead_xml_records = parse_xml_records(
         "tests/fixtures/ead/ead_record_missing_optional_fields.xml"

transmogrifier/sources/ead.py

@@ -152,6 +152,8 @@ def get_optional_fields(self, xml: Tag) -> Optional[dict]:
         for id_element in collection_description_did.find_all(
             "unitid", recursive=False
         ):
+            if id_element.get("type") == "aspace_uri":
+                continue
             if id_value := self.create_string_from_mixed_value(
                 id_element,
                 " ",