Non fxa migrated deletions #6611
Conversation
This migration removes user accounts that meet both criteria: 1. Have not been migrated to Firefox Accounts (is_fxa_migrated=False) 2. Have no content or content relationships (no posts, votes, messages, etc.)
7d627e8 to
698e714
Compare
There was a problem hiding this comment.
Pull Request Overview
This PR implements a migration to delete non-FXA migrated users who have no associated content and adds a test suite to verify the query logic.
- Adds a test case for the migration query logic in kitsune/users/tests/test_migration_0033.py
- Implements the migration in kitsune/users/migrations/0033_batch_delete_non_migrated_users.py that deletes users in batches, with logging for progress reporting
Reviewed Changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.
| File | Description |
|---|---|
| kitsune/users/tests/test_migration_0033.py | Introduces tests for various user content scenarios to validate deletion logic |
| kitsune/users/migrations/0033_batch_delete_non_migrated_users.py | Implements batch deletion of non-migrated users with additional exclusion conditions |
Comments suppressed due to low confidence (1)
kitsune/users/migrations/0033_batch_delete_non_migrated_users.py:34
- Consider adding tests to cover the newly added exclusion conditions (locales_leader, locales_reviewer, locales_editor, and wiki_contributions) to ensure they are correctly preventing deletions when associated content exists.
| Q(locales_leader__isnull=False)
| | Q(wiki_contributions__isnull=False) | ||
| ) | ||
|
|
||
| user_ids = list(users_to_delete.values_list("id", flat=True)) |
There was a problem hiding this comment.
This will load all million + into memory. Is this a good place to use iterator? Something like:
user_ids = users_to_delete.values_list("id", flat=True).iterator(chunk_size=2000)
There was a problem hiding this comment.
it is much better - that was my initial approach. The batching became more complex because of the slicing and calculating where to break out. The query will round up 800K which should be around 7MB
|
I am taking another shot at this - I will ping you both when it's ready |
698e714 to
d0060f4
Compare
No description provided.