Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -1079,8 +1079,50 @@ def cleanup_and_assert_status(self, data=None, expected_status=status.HTTP_204_N
return response

def test_simple_success(self):
"""
Test basic cleanup with default redacted values.
"""
# Verify redaction happens (records exist before cleanup)
assert UserRetirementStatus.objects.count() == 9

# Make the cleanup request
self.cleanup_and_assert_status()
assert not UserRetirementStatus.objects.all()

# Records should be deleted after redaction
retirements = UserRetirementStatus.objects.all()
assert retirements.count() == 0

def test_redaction_before_deletion(self):
"""
Verify that redaction (UPDATE) happens before deletion (DELETE).
Uses assertNumQueries to verify UPDATE queries execute before DELETE queries.
This protects PII from being exposed in soft-deletes to downstream data warehouses.
"""
# Use assertNumQueries to capture and verify the SQL queries execute in correct order.
with self.assertNumQueries(53): # Full request with 9 UPDATEs (redaction) + 9 DELETEs
self.cleanup_and_assert_status()

# Verify records are deleted after redaction
retirements = UserRetirementStatus.objects.all()
assert retirements.count() == 0

def test_custom_redacted_values(self):
"""Test that custom redacted values are applied before deletion."""
custom_username = 'username-redacted-12345'
custom_email = 'email-redacted-67890'
custom_name = 'name-redacted-abcde'

data = {
'usernames': self.usernames,
'redacted_username': custom_username,
'redacted_email': custom_email,
'redacted_name': custom_name
}
self.cleanup_and_assert_status(data=data)

# Records should be deleted after redaction
retirements = UserRetirementStatus.objects.all()
assert retirements.count() == 0

def test_leaves_other_users(self):
remaining_usernames = []
Expand Down
21 changes: 18 additions & 3 deletions openedx/core/djangoapps/user_api/accounts/views.py
Original file line number Diff line number Diff line change
Expand Up @@ -1024,14 +1024,20 @@ def cleanup(self, request):

```
{
'usernames': ['user1', 'user2', ...]
'usernames': ['user1', 'user2', ...],
'redacted_username': 'Value to store in username field',
'redacted_email': 'Value to store in email field',
'redacted_name': 'Value to store in name field'
}
```

Deletes a batch of retirement requests by username.
Redacts a batch of retirement requests by redacting PII fields.
"""
try:
usernames = request.data["usernames"]
redacted_username = request.data.get("redacted_username", "redacted")
redacted_email = request.data.get("redacted_email", "redacted")
redacted_name = request.data.get("redacted_name", "redacted")

if not isinstance(usernames, list):
raise TypeError("Usernames should be an array.")
Expand All @@ -1045,7 +1051,16 @@ def cleanup(self, request):
if len(usernames) != len(retirements):
raise UserRetirementStatus.DoesNotExist("Not all usernames exist in the COMPLETE state.")

retirements.delete()
# Redact PII fields first, then delete. In case an ETL tool is syncing data
# to a downstream data warehouse, and treats the deletes as soft-deletes,
# the data will have first been redacted, protecting the sensitive PII.
for retirement in retirements:
retirement.original_username = redacted_username
retirement.original_email = redacted_email
retirement.original_name = redacted_name
retirement.save()
retirement.delete()

return Response(status=status.HTTP_204_NO_CONTENT)
except (RetirementStateError, UserRetirementStatus.DoesNotExist, TypeError) as exc:
return Response(str(exc), status=status.HTTP_400_BAD_REQUEST)
Expand Down
Loading