feat(similarity): Add delete record by hash task

src/sentry/options/defaults.py

@@ -935,6 +935,12 @@
     flags=FLAG_ALLOW_EMPTY | FLAG_AUTOMATOR_MODIFIABLE,
 )
 
+register(
+    "embeddings-grouping.seer.delete-record-batch-size",
+    default=100,
+    flags=FLAG_ALLOW_EMPTY | FLAG_AUTOMATOR_MODIFIABLE,
+)
+
 # ## sentry.killswitches
 #
 # The following options are documented in sentry.killswitches in more detail

src/sentry/tasks/delete_seer_grouping_records_by_hash.py

@@ -0,0 +1,33 @@
+from typing import Any
+
+from sentry import options
+from sentry.seer.similarity.grouping_records import delete_grouping_records_by_hash
+from sentry.silo.base import SiloMode
+from sentry.tasks.base import instrumented_task
+
+
+@instrumented_task(
+    name="sentry.tasks.delete_seer_grouping_records_by_hash",
+    queue="delete_seer_grouping_records_by_hash",
+    max_retries=0,
+    silo_mode=SiloMode.REGION,
+)
+def delete_seer_grouping_records_by_hash(
+    project_id: int,
+    hashes: list[str],
+    last_deleted_index: int = 0,
+    soft_time_limit=60 * 15,
+    time_limit=60 * (15 + 5),
+    *args: Any,
+    **kwargs: Any,
+) -> None:
+    """
+    Task to delete seer grouping records by hash list.
+    Calls the seer delete by hash endpoint with batches of hashes of size `BATCH_SIZE`.
+    """
+    batch_size = options.get("embeddings-grouping.seer.delete-record-batch-size")
+    len_hashes = len(hashes)
+    end_index = min(last_deleted_index + batch_size, len_hashes)
+    delete_grouping_records_by_hash(project_id, hashes[last_deleted_index:end_index])
+    if end_index < len_hashes:
+        delete_seer_grouping_records_by_hash.apply_async(args=[project_id, hashes, end_index])

tests/sentry/tasks/test_delete_seer_grouping_records_by_hash.py

@@ -0,0 +1,24 @@
+from unittest.mock import patch
+
+from sentry.tasks.delete_seer_grouping_records_by_hash import delete_seer_grouping_records_by_hash
+from sentry.testutils.pytest.fixtures import django_db_all
+
+
+@django_db_all
+@patch("sentry.tasks.delete_seer_grouping_records_by_hash.delete_grouping_records_by_hash")
+@patch(
+    "sentry.tasks.delete_seer_grouping_records_by_hash.delete_seer_grouping_records_by_hash.apply_async"
+)
+def test_delete_seer_grouping_records_by_hash_batches(
+    mock_delete_seer_grouping_records_by_hash_apply_async, mock_delete_grouping_records_by_hash
+):
+    """
+    Test that when delete_seer_grouping_records_by_hash is called with over 20 hashes, it spawns
+    another task with the end index of the previous batch.
+    """
+    mock_delete_grouping_records_by_hash.return_value = True
+    project_id, hashes = 1, [str(i) for i in range(101)]
+    delete_seer_grouping_records_by_hash(project_id, hashes, 0)
+    assert mock_delete_seer_grouping_records_by_hash_apply_async.call_args[1] == {
+        "args": [project_id, hashes, 100]
+    }