fix(identity): create the identity when missing

- when the identity is missing, the sync will not create it
  without failing

Author: ntarocco

.github/workflows/pypi-publish.yml

@@ -5,7 +5,6 @@
 # Invenio-CERN-sync is free software; you can redistribute it and/or modify it
 # under the terms of the MIT License; see LICENSE file for more details.
 
-
 name: Publish
 
 on:
@@ -17,7 +16,7 @@ jobs:
   build-n-publish:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
       - name: Set up Python 3.12
         uses: actions/setup-python@v5
         with:
@@ -30,9 +29,8 @@ jobs:
         run: |
           python setup.py sdist bdist_wheel
       - name: pypi-publish
-        uses: pypa/gh-action-pypi-publish@v1.3.1
+        uses: pypa/gh-action-pypi-publish@v1.13
         with:
           user: __token__
 
           password: ${{ secrets.pypi_token }}
-

invenio_cern_sync/users/sync.py

@@ -21,6 +21,7 @@
 from ..ldap.client import LdapClient
 from ..ldap.serializer import serialize_ldap_users
 from ..logging import log_info, log_warning
+from ..sso import cern_remote_app_name
 from .api import create_user, update_existing_user
 
 
@@ -131,16 +132,17 @@ def _update_existing(users, serializer_fn, log_uuid, log_name, persist_every=500
                 )
                 invenio_user["remote_account_extra_data"]["changes"] = ra_extra_data
             elif user and (not user_identity or user_identity.id_user != user.id):
-                # The `identity_id` changed.
-                # The `identity_id` of the UserIdentity associated to the User
-                # will have to be updated.
+                # The `identity_id` changed or it does not exist yet.
                 try:
                     user_identity = UserIdentity.query.filter_by(id_user=user.id).one()
                 except NoResultFound:
-                    current_app.logger.error(
-                        f"UserIdentity not found for user.id={user.id}. Skipping this user..."
+                    UserIdentity.create(
+                        user,
+                        cern_remote_app_name,
+                        invenio_user["user_identity_id"],
                     )
-                    continue
+                    db.session.flush()
+                    user_identity = UserIdentity.query.filter_by(id_user=user.id).one()
 
                 _ra_extra_data = invenio_user["remote_account_extra_data"].get(
                     "changes", []
@@ -211,7 +213,7 @@ def _insert_missing(invenio_users, log_uuid, log_name, persist_every=500):
                 db.session.commit()
 
         except Exception as e:
-            current_app.logger.error(
+            current_app.logger.warning(
                 f"Error creating user from CERN data: {e}. Skipping this user... User: {invenio_user}"
             )
             continue

tests/test_sync_users.py

@@ -259,7 +259,44 @@ def test_sync_person_id_change(
         "users-sync",
         dict(action="updating-existing-users", msg=expected_log_msg),
         log_uuid=mock.ANY,
-    ),
+    )
+
+
+@patch("invenio_cern_sync.users.sync.KeycloakService")
+@patch("invenio_cern_sync.users.sync.AuthZService")
+@patch("invenio_cern_sync.users.sync.log_warning")
+def test_sync_identity_missing(
+    mock_log_warning,
+    MockAuthZService,
+    MockKeycloakService,
+    app,
+    cern_identities,
+    db,
+):
+    """Test sync identity missing."""
+    # prepare the db with the initial data
+    client_id = app.config["CERN_APP_CREDENTIALS"]["consumer_key"]
+
+    MockAuthZService.return_value.get_identities.return_value = cern_identities
+    sync(method="AuthZ")
+
+    # delete the first identity from the DB
+    first = cern_identities[0]
+    previous_person_id = first["personId"]
+    UserIdentity.query.filter_by(id=first["personId"]).delete()
+    db.session.commit()
+    MockAuthZService.return_value.get_identities.return_value = cern_identities
+    sync(method="AuthZ")
+
+    # check that the user identity was updated, but the user was not duplicated
+    assert UserIdentity.query.filter_by(id=first["personId"]).one()
+    user = User.query.filter_by(email=first["primaryAccountEmail"]).one()
+    expected_log_msg = f"Identity Id changed for User `{user.username}` `{user.email}`. Previous UserIdentity.id in the local DB: `{previous_person_id}` - New Identity Id from CERN DB: `{first['personId']}`."
+    mock_log_warning.assert_any_call(
+        "users-sync",
+        dict(action="updating-existing-users", msg=expected_log_msg),
+        log_uuid=mock.ANY,
+    )
 
 
 @patch("invenio_cern_sync.users.sync.KeycloakService")
@@ -308,4 +345,4 @@ def test_sync_username_email_change(
         "users-sync",
         dict(action="updating-existing-users", msg=expected_log_msg),
         log_uuid=mock.ANY,
-    ),
+    )