From 86ec2fbf843c8e2bba0704f337e5b26423615bc4 Mon Sep 17 00:00:00 2001 From: Ryan Crabbe Date: Tue, 24 Feb 2026 09:56:16 -0800 Subject: [PATCH 01/15] perf(proxy): batch 11 create_task() calls into 1 in update_database() Replace 11 separate asyncio.create_task() calls per request with a single batched task that runs all spend-update helpers sequentially. This reduces task scheduling overhead at high RPS (11,000 -> 1,000 tasks/sec at 1K RPS) and cuts 5 copy.deepcopy(payload) calls to 1 shared copy. Also fixes a mutation bug where the daily agent spend handler received the raw payload without deepcopy, unlike all other daily helpers. --- litellm/proxy/db/db_spend_update_writer.py | 284 +++++++++++------ .../proxy/db/test_db_spend_update_writer.py | 293 +++++++++++++----- 2 files changed, 419 insertions(+), 158 deletions(-) diff --git a/litellm/proxy/db/db_spend_update_writer.py b/litellm/proxy/db/db_spend_update_writer.py index 03628fda47f..d7d065d9064 100644 --- a/litellm/proxy/db/db_spend_update_writer.py +++ b/litellm/proxy/db/db_spend_update_writer.py @@ -124,53 +124,20 @@ class DBSpendUpdateWriter: payload["startTime"] = payload["startTime"].isoformat() if isinstance(payload["endTime"], datetime): payload["endTime"] = payload["endTime"].isoformat() - + if org_id is not None and org_id != "": payload["organization_id"] = org_id if team_id is not None and team_id != "": payload["team_id"] = team_id - asyncio.create_task( - self._update_user_db( - response_cost=response_cost, - user_id=user_id, - prisma_client=prisma_client, - user_api_key_cache=user_api_key_cache, - litellm_proxy_budget_name=litellm_proxy_budget_name, - end_user_id=end_user_id, - ) - ) - asyncio.create_task( - self._update_key_db( - response_cost=response_cost, - hashed_token=hashed_token, - prisma_client=prisma_client, - ) - ) - asyncio.create_task( - self._update_team_db( - response_cost=response_cost, - team_id=team_id, - user_id=user_id, - prisma_client=prisma_client, - ) - ) - asyncio.create_task( - self._update_org_db( - response_cost=response_cost, - org_id=org_id, - prisma_client=prisma_client, - ) - ) - asyncio.create_task( - self._update_tag_db( - response_cost=response_cost, - request_tags=copy.deepcopy(payload.get("request_tags")), - prisma_client=prisma_client, - ) - ) + # One deepcopy shared by all 6 daily spend helpers (was 5, fixes agent bug) + payload_copy = copy.deepcopy(payload) + # Deepcopy request_tags for _update_tag_db + request_tags = copy.deepcopy(payload.get("request_tags")) + + # Keep _insert_spend_log_to_db awaited inline (not a task, preserve current behavior) if disable_spend_logs is False: await self._insert_spend_log_to_db( payload=copy.deepcopy(payload), @@ -181,44 +148,20 @@ class DBSpendUpdateWriter: "disable_spend_logs=True. Skipping writing spend logs to db. Other spend updates - Key/User/Team table will still occur." ) + # Single task replaces 11 create_task() calls asyncio.create_task( - self.add_spend_log_transaction_to_daily_user_transaction( - payload=copy.deepcopy(payload), - prisma_client=prisma_client, - ) - ) - - asyncio.create_task( - self.add_spend_log_transaction_to_daily_end_user_transaction( - payload=copy.deepcopy(payload), - prisma_client=prisma_client, - ) - ) - - asyncio.create_task( - self.add_spend_log_transaction_to_daily_agent_transaction( - payload=payload, - prisma_client=prisma_client, - ) - ) - - asyncio.create_task( - self.add_spend_log_transaction_to_daily_team_transaction( - payload=copy.deepcopy(payload), - prisma_client=prisma_client, - ) - ) - asyncio.create_task( - self.add_spend_log_transaction_to_daily_org_transaction( - payload=copy.deepcopy(payload), + self._batch_database_updates( + response_cost=response_cost, + user_id=user_id, + hashed_token=hashed_token, + team_id=team_id, org_id=org_id, + end_user_id=end_user_id, prisma_client=prisma_client, - ) - ) - asyncio.create_task( - self.add_spend_log_transaction_to_daily_tag_transaction( - payload=copy.deepcopy(payload), - prisma_client=prisma_client, + user_api_key_cache=user_api_key_cache, + litellm_proxy_budget_name=litellm_proxy_budget_name, + payload_copy=payload_copy, + request_tags=request_tags, ) ) @@ -228,6 +171,157 @@ class DBSpendUpdateWriter: f"Error updating Prisma database: {traceback.format_exc()}" ) + async def _batch_database_updates( + self, + *, + response_cost: Optional[float], + user_id: Optional[str], + hashed_token: Optional[str], + team_id: Optional[str], + org_id: Optional[str], + end_user_id: Optional[str], + prisma_client: Optional[PrismaClient], + user_api_key_cache: DualCache, + litellm_proxy_budget_name: Optional[str], + payload_copy: dict, + request_tags: Optional[Any], + ): + """ + Runs all 11 spend-update helpers sequentially inside a single asyncio task. + + Each helper is wrapped in try/except so one failure doesn't prevent the others. + """ + try: + await self._update_user_db( + response_cost=response_cost, + user_id=user_id, + prisma_client=prisma_client, + user_api_key_cache=user_api_key_cache, + litellm_proxy_budget_name=litellm_proxy_budget_name, + end_user_id=end_user_id, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: _update_user_db failed: %s", + traceback.format_exc(), + ) + + try: + await self._update_key_db( + response_cost=response_cost, + hashed_token=hashed_token, + prisma_client=prisma_client, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: _update_key_db failed: %s", + traceback.format_exc(), + ) + + try: + await self._update_team_db( + response_cost=response_cost, + team_id=team_id, + user_id=user_id, + prisma_client=prisma_client, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: _update_team_db failed: %s", + traceback.format_exc(), + ) + + try: + await self._update_org_db( + response_cost=response_cost, + org_id=org_id, + prisma_client=prisma_client, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: _update_org_db failed: %s", + traceback.format_exc(), + ) + + try: + await self._update_tag_db( + response_cost=response_cost, + request_tags=request_tags, + prisma_client=prisma_client, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: _update_tag_db failed: %s", + traceback.format_exc(), + ) + + try: + await self.add_spend_log_transaction_to_daily_user_transaction( + payload=payload_copy, + prisma_client=prisma_client, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: add_spend_log_transaction_to_daily_user_transaction failed: %s", + traceback.format_exc(), + ) + + try: + await self.add_spend_log_transaction_to_daily_end_user_transaction( + payload=payload_copy, + prisma_client=prisma_client, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: add_spend_log_transaction_to_daily_end_user_transaction failed: %s", + traceback.format_exc(), + ) + + try: + await self.add_spend_log_transaction_to_daily_agent_transaction( + payload=payload_copy, + prisma_client=prisma_client, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: add_spend_log_transaction_to_daily_agent_transaction failed: %s", + traceback.format_exc(), + ) + + try: + await self.add_spend_log_transaction_to_daily_team_transaction( + payload=payload_copy, + prisma_client=prisma_client, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: add_spend_log_transaction_to_daily_team_transaction failed: %s", + traceback.format_exc(), + ) + + try: + await self.add_spend_log_transaction_to_daily_org_transaction( + payload=payload_copy, + org_id=org_id, + prisma_client=prisma_client, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: add_spend_log_transaction_to_daily_org_transaction failed: %s", + traceback.format_exc(), + ) + + try: + await self.add_spend_log_transaction_to_daily_tag_transaction( + payload=payload_copy, + prisma_client=prisma_client, + ) + except Exception: + verbose_proxy_logger.debug( + "_batch_database_updates: add_spend_log_transaction_to_daily_tag_transaction failed: %s", + traceback.format_exc(), + ) + async def _update_key_db( self, response_cost: Optional[float], @@ -880,7 +974,7 @@ class DBSpendUpdateWriter: team_id = key.split("::")[1] user_id = key.split("::")[3] team_memberships_to_invalidate.append((user_id, team_id)) - + for i in range(n_retry_times + 1): start_time = time.time() try: @@ -917,11 +1011,13 @@ class DBSpendUpdateWriter: _raise_failed_update_spend_exception( e=e, start_time=start_time, proxy_logging_obj=proxy_logging_obj ) - + # Invalidate cache for updated team memberships # This ensures budget checks read fresh spend data from the database if team_memberships_to_invalidate and proxy_logging_obj is not None: - user_api_key_cache = proxy_logging_obj.call_details.get("user_api_key_cache") + user_api_key_cache = proxy_logging_obj.call_details.get( + "user_api_key_cache" + ) if user_api_key_cache is not None: for user_id, team_id in team_memberships_to_invalidate: cache_key = "team_membership:{}:{}".format(user_id, team_id) @@ -1233,7 +1329,9 @@ class DBSpendUpdateWriter: ), "endpoint": transaction.get("endpoint") or "", "prompt_tokens": transaction["prompt_tokens"], - "completion_tokens": transaction["completion_tokens"], + "completion_tokens": transaction[ + "completion_tokens" + ], "spend": transaction["spend"], "api_requests": transaction["api_requests"], "successful_requests": transaction[ @@ -1244,12 +1342,14 @@ class DBSpendUpdateWriter: # Add cache-related fields if they exist if "cache_read_input_tokens" in transaction: - common_data["cache_read_input_tokens"] = ( - transaction.get("cache_read_input_tokens", 0) - ) + common_data[ + "cache_read_input_tokens" + ] = transaction.get("cache_read_input_tokens", 0) if "cache_creation_input_tokens" in transaction: - common_data["cache_creation_input_tokens"] = ( - transaction.get("cache_creation_input_tokens", 0) + common_data[ + "cache_creation_input_tokens" + ] = transaction.get( + "cache_creation_input_tokens", 0 ) if entity_type == "tag" and "request_id" in transaction: @@ -1292,10 +1392,14 @@ class DBSpendUpdateWriter: } if entity_type == "tag" and "request_id" in transaction: - update_data["request_id"] = transaction.get("request_id") + update_data["request_id"] = transaction.get( + "request_id" + ) # Add endpoint to update_data so existing rows get their endpoint field updated - update_data["endpoint"] = transaction.get("endpoint") or "" + update_data["endpoint"] = ( + transaction.get("endpoint") or "" + ) table.upsert( where=where_clause, @@ -1479,7 +1583,9 @@ class DBSpendUpdateWriter: self, payload: Union[dict, SpendLogsPayload], prisma_client: PrismaClient, - type: Literal["user", "team", "org", "request_tags", "end_user", "agent"] = "user", + type: Literal[ + "user", "team", "org", "request_tags", "end_user", "agent" + ] = "user", ) -> Optional[BaseDailySpendTransaction]: common_expected_keys = ["startTime", "api_key"] if type == "user": @@ -1538,7 +1644,7 @@ class DBSpendUpdateWriter: endpoint = None if call_type: endpoint = ROUTE_ENDPOINT_MAPPING.get(call_type, None) - + daily_transaction = BaseDailySpendTransaction( date=date, api_key=payload["api_key"], @@ -1750,7 +1856,7 @@ class DBSpendUpdateWriter: endpoint_str = base_daily_transaction.get("endpoint") or "" daily_transaction_key = f"{payload['agent_id']}_{base_daily_transaction['date']}_{payload_with_agent_id['api_key']}_{payload_with_agent_id['model']}_{payload_with_agent_id['custom_llm_provider']}_{endpoint_str}" daily_transaction = DailyAgentSpendTransaction( - agent_id=payload['agent_id'], **base_daily_transaction + agent_id=payload["agent_id"], **base_daily_transaction ) await self.daily_agent_spend_update_queue.add_update( update={daily_transaction_key: daily_transaction} diff --git a/tests/test_litellm/proxy/db/test_db_spend_update_writer.py b/tests/test_litellm/proxy/db/test_db_spend_update_writer.py index 0fa0d4cf10b..0465f2adaf3 100644 --- a/tests/test_litellm/proxy/db/test_db_spend_update_writer.py +++ b/tests/test_litellm/proxy/db/test_db_spend_update_writer.py @@ -1,3 +1,4 @@ +import asyncio import json import os import sys @@ -51,6 +52,9 @@ async def test_daily_spend_tracking_with_disabled_spend_logs(): # Call the method await db_writer.update_database(**test_data) + # Let the single batched task run + await asyncio.sleep(0) + # Verify that _insert_spend_log_to_db was NOT called (since disable_spend_logs is True) db_writer._insert_spend_log_to_db.assert_not_called() @@ -115,7 +119,9 @@ async def test_update_daily_spend_with_null_entity_id(): # Verify the where clause contains null entity_id call_args = mock_table.upsert.call_args[1] - where_clause = call_args["where"]["user_id_date_api_key_model_custom_llm_provider_mcp_namespaced_tool_name_endpoint"] + where_clause = call_args["where"][ + "user_id_date_api_key_model_custom_llm_provider_mcp_namespaced_tool_name_endpoint" + ] assert where_clause["user_id"] is None assert where_clause["date"] == "2024-01-01" assert where_clause["api_key"] == "test-api-key" @@ -161,7 +167,7 @@ async def test_update_daily_spend_sorting(): upsert_calls = [] for i in range(50): daily_spend_transactions[f"test_key_{i}"] = { - "user_id": f"user{60-i}", # user60 ... user11, reverse order + "user_id": f"user{60-i}", # user60 ... user11, reverse order "date": "2024-01-01", "api_key": "test-api-key", "model": "gpt-4", @@ -173,46 +179,48 @@ async def test_update_daily_spend_sorting(): "successful_requests": 1, "failed_requests": 0, } - upsert_calls.append(call( - where={ - "user_id_date_api_key_model_custom_llm_provider_mcp_namespaced_tool_name_endpoint": { - "user_id": f"user{i+11}", # user11 ... user60, sorted order - "date": "2024-01-01", - "api_key": "test-api-key", - "model": "gpt-4", - "custom_llm_provider": "openai", - "mcp_namespaced_tool_name": "", - "endpoint": "", - } - }, - data={ - "create": { - "user_id": f"user{i+11}", - "date": "2024-01-01", - "api_key": "test-api-key", - "model": "gpt-4", - "model_group": None, - "mcp_namespaced_tool_name": "", - "custom_llm_provider": "openai", - "endpoint": "", - "prompt_tokens": 10, - "completion_tokens": 20, - "spend": 0.1, - "api_requests": 1, - "successful_requests": 1, - "failed_requests": 0, + upsert_calls.append( + call( + where={ + "user_id_date_api_key_model_custom_llm_provider_mcp_namespaced_tool_name_endpoint": { + "user_id": f"user{i+11}", # user11 ... user60, sorted order + "date": "2024-01-01", + "api_key": "test-api-key", + "model": "gpt-4", + "custom_llm_provider": "openai", + "mcp_namespaced_tool_name": "", + "endpoint": "", + } }, - "update": { - "prompt_tokens": {"increment": 10}, - "completion_tokens": {"increment": 20}, - "spend": {"increment": 0.1}, - "api_requests": {"increment": 1}, - "successful_requests": {"increment": 1}, - "failed_requests": {"increment": 0}, - "endpoint": "", + data={ + "create": { + "user_id": f"user{i+11}", + "date": "2024-01-01", + "api_key": "test-api-key", + "model": "gpt-4", + "model_group": None, + "mcp_namespaced_tool_name": "", + "custom_llm_provider": "openai", + "endpoint": "", + "prompt_tokens": 10, + "completion_tokens": 20, + "spend": 0.1, + "api_requests": 1, + "successful_requests": 1, + "failed_requests": 0, + }, + "update": { + "prompt_tokens": {"increment": 10}, + "completion_tokens": {"increment": 20}, + "spend": {"increment": 0.1}, + "api_requests": {"increment": 1}, + "successful_requests": {"increment": 1}, + "failed_requests": {"increment": 0}, + "endpoint": "", + }, }, - }, - )) + ) + ) # Call the method await DBSpendUpdateWriter._update_daily_spend( @@ -275,7 +283,7 @@ async def test_update_daily_spend_tag_with_request_id(): # Verify that table.upsert was called mock_table.upsert.assert_called_once() - + # Verify request_id is in update_data call_args = mock_table.upsert.call_args[1] update_data = call_args["data"]["update"] @@ -283,15 +291,13 @@ async def test_update_daily_spend_tag_with_request_id(): assert update_data["request_id"] == "test-request-id-123" - - @pytest.mark.asyncio async def test_update_daily_spend_with_none_values_in_sorting_fields(): """ Test that _update_daily_spend handles None values in sorting fields correctly. - + This test ensures that when fields like date, api_key, model, or custom_llm_provider - are None, the sorting doesn't crash with TypeError: '<' not supported between + are None, the sorting doesn't crash with TypeError: '<' not supported between instances of 'NoneType' and 'str'. """ # Setup @@ -509,6 +515,7 @@ async def test_update_tag_db_without_prisma_client(): assert writer.spend_update_queue.add_update.call_count == 0 + @pytest.mark.asyncio async def test_add_spend_log_transaction_to_daily_tag_transaction_with_request_id(): """ @@ -518,7 +525,7 @@ async def test_add_spend_log_transaction_to_daily_tag_transaction_with_request_i writer = DBSpendUpdateWriter() mock_prisma = MagicMock() mock_prisma.get_request_status = MagicMock(return_value="success") - + request_id = "test-request-id-123" payload = { "request_id": request_id, @@ -546,13 +553,15 @@ async def test_add_spend_log_transaction_to_daily_tag_transaction_with_request_i # Should be called twice (once for each tag) assert writer.daily_tag_spend_update_queue.add_update.call_count == 2 - + # Check that request_id is included in both transactions for call in writer.daily_tag_spend_update_queue.add_update.call_args_list: transaction_dict = call[1]["update"] # Each transaction should have one key with the format tag_date_api_key_model_provider for key, transaction in transaction_dict.items(): - assert transaction["request_id"] == request_id, f"request_id should be {request_id} but got {transaction.get('request_id')}" + assert ( + transaction["request_id"] == request_id + ), f"request_id should be {request_id} but got {transaction.get('request_id')}" @pytest.mark.asyncio @@ -866,11 +875,11 @@ async def test_endpoint_field_is_correctly_mapped_from_call_type(): call_args = writer.daily_spend_update_queue.add_update.call_args[1] update_dict = call_args["update"] assert len(update_dict) == 1 - + for key, transaction in update_dict.items(): # Verify endpoint is included in the key assert key == f"test-user_2024-01-01_test-key_gpt-4_openai_/chat/completions" - + # Verify endpoint is set in the transaction assert transaction["endpoint"] == "/chat/completions" assert transaction["user_id"] == "test-user" @@ -887,7 +896,7 @@ async def test_update_daily_spend_logs_detailed_error_on_batch_upsert_failure(): This ensures proper debugging information is available for issues like unique constraint violations. """ from litellm._logging import verbose_proxy_logger - + # Setup mock_prisma_client = MagicMock() mock_batcher = MagicMock() @@ -895,13 +904,13 @@ async def test_update_daily_spend_logs_detailed_error_on_batch_upsert_failure(): mock_batch_context = MagicMock() mock_batch_context.__aenter__ = AsyncMock(return_value=mock_batcher) mock_batcher.litellm_dailyuserspend = mock_table - + # Make the batch context manager's exit raise an exception # This simulates a batch commit failure (e.g., unique constraint violation) test_exception = Exception("Unique constraint violation") mock_batch_context.__aexit__ = AsyncMock(side_effect=test_exception) mock_prisma_client.db.batch_.return_value = mock_batch_context - + # Create a transaction daily_spend_transactions = { "test_key": { @@ -918,13 +927,13 @@ async def test_update_daily_spend_logs_detailed_error_on_batch_upsert_failure(): "failed_requests": 0, } } - + # Create a mock proxy_logging_obj with failure_handler as AsyncMock mock_proxy_logging = MagicMock() mock_proxy_logging.failure_handler = AsyncMock() - + # Mock the logger to capture exception calls - with patch.object(verbose_proxy_logger, 'exception') as mock_exception_logger: + with patch.object(verbose_proxy_logger, "exception") as mock_exception_logger: # Call the method and expect it to raise the exception with pytest.raises(Exception, match="Unique constraint violation"): await DBSpendUpdateWriter._update_daily_spend( @@ -937,13 +946,16 @@ async def test_update_daily_spend_logs_detailed_error_on_batch_upsert_failure(): table_name="litellm_dailyuserspend", unique_constraint_name="user_id_date_api_key_model_custom_llm_provider_mcp_namespaced_tool_name_endpoint", ) - + # Verify that exception was logged with detailed information assert mock_exception_logger.called call_args = mock_exception_logger.call_args[0][0] assert "Daily user spend batch upsert failed" in call_args assert "Table: litellm_dailyuserspend" in call_args - assert "Constraint: user_id_date_api_key_model_custom_llm_provider_mcp_namespaced_tool_name_endpoint" in call_args + assert ( + "Constraint: user_id_date_api_key_model_custom_llm_provider_mcp_namespaced_tool_name_endpoint" + in call_args + ) assert "Batch size: 1" in call_args assert "Unique constraint violation" in call_args @@ -961,7 +973,7 @@ async def test_update_daily_spend_re_raises_exception_after_logging(): mock_batch_context = MagicMock() mock_batch_context.__aenter__ = AsyncMock(return_value=mock_batcher) mock_batcher.litellm_dailyuserspend = mock_table - + # Create a transaction daily_spend_transactions = { "test_key": { @@ -978,16 +990,16 @@ async def test_update_daily_spend_re_raises_exception_after_logging(): "failed_requests": 0, } } - + # Create a custom exception to verify it's re-raised custom_exception = ValueError("Database connection lost") mock_batch_context.__aexit__ = AsyncMock(side_effect=custom_exception) mock_prisma_client.db.batch_.return_value = mock_batch_context - + # Create a mock proxy_logging_obj with failure_handler as AsyncMock mock_proxy_logging = MagicMock() mock_proxy_logging.failure_handler = AsyncMock() - + # Verify the exception is re-raised with pytest.raises(ValueError, match="Database connection lost"): await DBSpendUpdateWriter._update_daily_spend( @@ -1018,10 +1030,12 @@ async def test_commit_key_spend_updates_includes_last_active(): mock_transaction = AsyncMock() mock_transaction.__aenter__ = AsyncMock(return_value=mock_transaction) mock_transaction.__aexit__ = AsyncMock(return_value=False) - mock_transaction.batch_ = MagicMock(return_value=AsyncMock( - __aenter__=AsyncMock(return_value=mock_batcher), - __aexit__=AsyncMock(return_value=False), - )) + mock_transaction.batch_ = MagicMock( + return_value=AsyncMock( + __aenter__=AsyncMock(return_value=mock_batcher), + __aexit__=AsyncMock(return_value=False), + ) + ) mock_prisma_client = MagicMock() mock_prisma_client.db = MagicMock() @@ -1049,9 +1063,7 @@ async def test_commit_key_spend_updates_includes_last_active(): before_call = datetime.now(timezone.utc) - with patch( - "litellm.proxy.utils._raise_failed_update_spend_exception" - ): + with patch("litellm.proxy.utils._raise_failed_update_spend_exception"): await db_writer._commit_spend_updates_to_db( prisma_client=mock_prisma_client, n_retry_times=0, @@ -1076,3 +1088,146 @@ async def test_commit_key_spend_updates_includes_last_active(): last_active = call_kwargs["data"]["last_active"] assert isinstance(last_active, datetime) assert before_call <= last_active <= after_call + + +@pytest.mark.asyncio +async def test_update_database_creates_single_task(): + """ + Test that update_database() fires exactly 1 asyncio.create_task() call + (the batched task) instead of the previous 11. + """ + db_writer = DBSpendUpdateWriter() + + # Mock all helpers so nothing real runs + db_writer._insert_spend_log_to_db = AsyncMock() + db_writer._batch_database_updates = AsyncMock() + + with patch("litellm.proxy.proxy_server.disable_spend_logs", False), patch( + "litellm.proxy.proxy_server.prisma_client", MagicMock() + ), patch("litellm.proxy.proxy_server.user_api_key_cache", MagicMock()), patch( + "litellm.proxy.proxy_server.litellm_proxy_budget_name", "test-budget" + ), patch( + "litellm.proxy.db.db_spend_update_writer.asyncio.create_task" + ) as mock_create_task: + await db_writer.update_database( + token="test-token", + user_id="test-user", + end_user_id="test-end-user", + start_time=datetime.now(), + end_time=datetime.now(), + team_id="test-team", + org_id="test-org", + completion_response=MagicMock(), + response_cost=0.1, + kwargs={"model": "gpt-4", "custom_llm_provider": "openai"}, + ) + + # Exactly 1 create_task call (the batch), not 11 + assert mock_create_task.call_count == 1 + + +@pytest.mark.asyncio +async def test_batch_database_updates_isolation_on_failure(): + """ + Test that if one helper inside _batch_database_updates raises, + all other helpers still execute. + """ + db_writer = DBSpendUpdateWriter() + + # Make _update_key_db raise + db_writer._update_key_db = AsyncMock(side_effect=RuntimeError("key db boom")) + + # All other helpers are normal mocks + db_writer._update_user_db = AsyncMock() + db_writer._update_team_db = AsyncMock() + db_writer._update_org_db = AsyncMock() + db_writer._update_tag_db = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_user_transaction = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_end_user_transaction = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_agent_transaction = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_team_transaction = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_org_transaction = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_tag_transaction = AsyncMock() + + await db_writer._batch_database_updates( + response_cost=0.1, + user_id="u1", + hashed_token="t1", + team_id="team1", + org_id="org1", + end_user_id="eu1", + prisma_client=MagicMock(), + user_api_key_cache=MagicMock(), + litellm_proxy_budget_name="budget", + payload_copy={"key": "value"}, + request_tags=None, + ) + + # _update_key_db raised, but all others should still have been called + db_writer._update_user_db.assert_awaited_once() + db_writer._update_key_db.assert_awaited_once() + db_writer._update_team_db.assert_awaited_once() + db_writer._update_org_db.assert_awaited_once() + db_writer._update_tag_db.assert_awaited_once() + db_writer.add_spend_log_transaction_to_daily_user_transaction.assert_awaited_once() + db_writer.add_spend_log_transaction_to_daily_end_user_transaction.assert_awaited_once() + db_writer.add_spend_log_transaction_to_daily_agent_transaction.assert_awaited_once() + db_writer.add_spend_log_transaction_to_daily_team_transaction.assert_awaited_once() + db_writer.add_spend_log_transaction_to_daily_org_transaction.assert_awaited_once() + db_writer.add_spend_log_transaction_to_daily_tag_transaction.assert_awaited_once() + + +@pytest.mark.asyncio +async def test_daily_agent_receives_deepcopied_payload(): + """ + Test that the daily agent handler receives a deepcopied payload (not the original). + + Previously, add_spend_log_transaction_to_daily_agent_transaction received the raw + payload without a deepcopy, which was a mutation bug. + """ + db_writer = DBSpendUpdateWriter() + + original_payload = {"key": "value", "nested": {"a": 1}} + captured_payloads = [] + + async def capture_payload(**kwargs): + captured_payloads.append(kwargs.get("payload")) + + # Mock all helpers + db_writer._update_user_db = AsyncMock() + db_writer._update_key_db = AsyncMock() + db_writer._update_team_db = AsyncMock() + db_writer._update_org_db = AsyncMock() + db_writer._update_tag_db = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_user_transaction = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_end_user_transaction = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_agent_transaction = AsyncMock( + side_effect=capture_payload + ) + db_writer.add_spend_log_transaction_to_daily_team_transaction = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_org_transaction = AsyncMock() + db_writer.add_spend_log_transaction_to_daily_tag_transaction = AsyncMock() + + import copy + + payload_copy = copy.deepcopy(original_payload) + + await db_writer._batch_database_updates( + response_cost=0.1, + user_id="u1", + hashed_token="t1", + team_id="team1", + org_id="org1", + end_user_id="eu1", + prisma_client=MagicMock(), + user_api_key_cache=MagicMock(), + litellm_proxy_budget_name="budget", + payload_copy=payload_copy, + request_tags=None, + ) + + # The payload passed to the agent handler should NOT be the original object + assert len(captured_payloads) == 1 + assert captured_payloads[0] is not original_payload + # But it should have the same content + assert captured_payloads[0] == original_payload From 4c5963fdb90fd33b64cb4687a8ce8c41686810e7 Mon Sep 17 00:00:00 2001 From: Ryan Crabbe Date: Tue, 24 Feb 2026 10:49:16 -0800 Subject: [PATCH 02/15] test: exercise production deepcopy path in agent payload test --- .../proxy/db/test_db_spend_update_writer.py | 75 ++++++++++++------- 1 file changed, 49 insertions(+), 26 deletions(-) diff --git a/tests/test_litellm/proxy/db/test_db_spend_update_writer.py b/tests/test_litellm/proxy/db/test_db_spend_update_writer.py index 0465f2adaf3..add7293285a 100644 --- a/tests/test_litellm/proxy/db/test_db_spend_update_writer.py +++ b/tests/test_litellm/proxy/db/test_db_spend_update_writer.py @@ -1183,17 +1183,21 @@ async def test_daily_agent_receives_deepcopied_payload(): Test that the daily agent handler receives a deepcopied payload (not the original). Previously, add_spend_log_transaction_to_daily_agent_transaction received the raw - payload without a deepcopy, which was a mutation bug. + payload without a deepcopy, which was a mutation bug. This test goes through + update_database() to verify the production deepcopy path. """ db_writer = DBSpendUpdateWriter() - original_payload = {"key": "value", "nested": {"a": 1}} - captured_payloads = [] + # Capture the payload object that get_logging_payload returns (the "original") + # and the payload the agent handler receives (should be a deepcopy) + original_payload_ref = {} + captured_agent_payloads = [] - async def capture_payload(**kwargs): - captured_payloads.append(kwargs.get("payload")) + async def capture_agent_payload(**kwargs): + captured_agent_payloads.append(kwargs.get("payload")) # Mock all helpers + db_writer._insert_spend_log_to_db = AsyncMock() db_writer._update_user_db = AsyncMock() db_writer._update_key_db = AsyncMock() db_writer._update_team_db = AsyncMock() @@ -1202,32 +1206,51 @@ async def test_daily_agent_receives_deepcopied_payload(): db_writer.add_spend_log_transaction_to_daily_user_transaction = AsyncMock() db_writer.add_spend_log_transaction_to_daily_end_user_transaction = AsyncMock() db_writer.add_spend_log_transaction_to_daily_agent_transaction = AsyncMock( - side_effect=capture_payload + side_effect=capture_agent_payload ) db_writer.add_spend_log_transaction_to_daily_team_transaction = AsyncMock() db_writer.add_spend_log_transaction_to_daily_org_transaction = AsyncMock() db_writer.add_spend_log_transaction_to_daily_tag_transaction = AsyncMock() - import copy + # Mock get_logging_payload to return a known dict and capture its identity + fake_payload = { + "startTime": "2024-01-01T00:00:00", + "endTime": "2024-01-01T00:01:00", + "model": "gpt-4", + "custom_llm_provider": "openai", + "spend": 0.0, + "nested": {"a": 1}, + } + original_payload_ref["obj"] = fake_payload # store reference to the original - payload_copy = copy.deepcopy(original_payload) + with patch("litellm.proxy.proxy_server.disable_spend_logs", True), patch( + "litellm.proxy.proxy_server.prisma_client", MagicMock() + ), patch("litellm.proxy.proxy_server.user_api_key_cache", MagicMock()), patch( + "litellm.proxy.proxy_server.litellm_proxy_budget_name", "test-budget" + ), patch( + "litellm.proxy.spend_tracking.spend_tracking_utils.get_logging_payload", + return_value=fake_payload, + ): + await db_writer.update_database( + token="test-token", + user_id="test-user", + end_user_id="test-end-user", + team_id="test-team", + org_id="test-org", + kwargs={"model": "gpt-4", "custom_llm_provider": "openai"}, + completion_response=MagicMock(), + start_time=datetime.now(), + end_time=datetime.now(), + response_cost=0.1, + ) - await db_writer._batch_database_updates( - response_cost=0.1, - user_id="u1", - hashed_token="t1", - team_id="team1", - org_id="org1", - end_user_id="eu1", - prisma_client=MagicMock(), - user_api_key_cache=MagicMock(), - litellm_proxy_budget_name="budget", - payload_copy=payload_copy, - request_tags=None, - ) + # Let the single batched task run + await asyncio.sleep(0) - # The payload passed to the agent handler should NOT be the original object - assert len(captured_payloads) == 1 - assert captured_payloads[0] is not original_payload - # But it should have the same content - assert captured_payloads[0] == original_payload + # The agent handler should have been called + assert len(captured_agent_payloads) == 1 + # The payload must NOT be the same object as the original (deepcopy occurred) + assert captured_agent_payloads[0] is not original_payload_ref["obj"] + # But it should have equivalent content + assert captured_agent_payloads[0]["model"] == "gpt-4" + assert captured_agent_payloads[0]["spend"] == 0.1 From 98b496433086b5a56f87e610c0349607b7553f55 Mon Sep 17 00:00:00 2001 From: Ryan Crabbe Date: Tue, 24 Feb 2026 12:31:45 -0800 Subject: [PATCH 03/15] perf(proxy): pipeline Redis RPUSH/LPOP in spend update cycle Replace 14 sequential Redis round-trips (7 RPUSH + 7 LPOP) per spend update cycle with 2 pipelined calls (1 RPUSH pipeline + 1 LPOP pipeline). This reduces connection pool contention at scale (50+ pods). - Add RedisPipelineRpushOperation and RedisPipelineLpopOperation TypedDicts - Add async_rpush_pipeline() and async_lpop_pipeline() to RedisCache - Refactor store_in_memory_spend_updates_in_redis() to use pipeline - Add get_all_transactions_from_redis_buffer_pipeline() for batched drain - Update _commit_spend_updates_to_db_with_redis() to use pipeline drain - Existing individual methods preserved for backward compatibility --- litellm/caching/redis_cache.py | 187 ++++++++++++++- litellm/proxy/db/db_spend_update_writer.py | 31 +-- .../redis_update_buffer.py | 149 ++++++++---- litellm/types/caching.py | 18 ++ .../test_litellm/caching/test_redis_cache.py | 213 ++++++++++++++++++ .../test_redis_update_buffer.py | 194 ++++++++++++++++ .../proxy/db/test_db_spend_update_writer.py | 43 ++++ 7 files changed, 773 insertions(+), 62 deletions(-) create mode 100644 tests/test_litellm/proxy/db/db_transaction_queue/test_redis_update_buffer.py diff --git a/litellm/caching/redis_cache.py b/litellm/caching/redis_cache.py index dcc2df5f91c..9b93d890cec 100644 --- a/litellm/caching/redis_cache.py +++ b/litellm/caching/redis_cache.py @@ -22,7 +22,11 @@ from litellm._logging import print_verbose, verbose_logger from litellm.constants import DEFAULT_REDIS_MAJOR_VERSION from litellm.litellm_core_utils.core_helpers import _get_parent_otel_span_from_kwargs from litellm.litellm_core_utils.coroutine_checker import coroutine_checker -from litellm.types.caching import RedisPipelineIncrementOperation +from litellm.types.caching import ( + RedisPipelineIncrementOperation, + RedisPipelineLpopOperation, + RedisPipelineRpushOperation, +) from litellm.types.services import ServiceTypes from .base_cache import BaseCache @@ -1320,6 +1324,75 @@ class RedisCache(BaseCache): ) raise e + async def _pipeline_rpush_helper( + self, + pipe: pipeline, + rpush_list: List[RedisPipelineRpushOperation], + ) -> List[int]: + """Helper function for pipeline rpush operations""" + for rpush_op in rpush_list: + pipe.rpush(rpush_op["key"], *rpush_op["values"]) + results = await pipe.execute() + # Preserve positional correspondence — raise on per-command errors + for r in results: + if isinstance(r, Exception): + raise r + return results + + async def async_rpush_pipeline( + self, + rpush_list: List[RedisPipelineRpushOperation], + ) -> List[int]: + """ + Use Redis Pipelines for bulk RPUSH operations + + Args: + rpush_list: List of RedisPipelineRpushOperation dicts containing: + - key: str + - values: List[Any] + + Returns: + List[int]: List lengths after each push + """ + if len(rpush_list) == 0: + return [] + + _redis_client: Any = self.init_async_client() + start_time = time.time() + + try: + async with _redis_client.pipeline(transaction=False) as pipe: + results = await self._pipeline_rpush_helper(pipe, rpush_list) + + ## LOGGING ## + end_time = time.time() + _duration = end_time - start_time + asyncio.create_task( + self.service_logger_obj.async_service_success_hook( + service=ServiceTypes.REDIS, + duration=_duration, + call_type=f"async_rpush_pipeline <- {_get_call_stack_info()}", + ) + ) + return results + except Exception as e: + ## LOGGING ## + end_time = time.time() + _duration = end_time - start_time + asyncio.create_task( + self.service_logger_obj.async_service_failure_hook( + service=ServiceTypes.REDIS, + duration=_duration, + error=e, + call_type=f"async_rpush_pipeline <- {_get_call_stack_info()}", + ) + ) + verbose_logger.error( + "LiteLLM Redis Caching: async_rpush_pipeline() - Got exception from REDIS %s", + str(e), + ) + raise e + async def handle_lpop_count_for_older_redis_versions( self, pipe: pipeline, key: str, count: int ) -> List[bytes]: @@ -1400,3 +1473,115 @@ class RedisCache(BaseCache): f"LiteLLM Redis Cache LPOP: - Got exception from REDIS : {str(e)}" ) raise e + + async def _pipeline_lpop_helper( + self, + pipe: pipeline, + lpop_list: List[RedisPipelineLpopOperation], + ) -> List[Optional[List[str]]]: + """Helper function for pipeline lpop operations. + + For Redis >= 7, queues one LPOP(key, count) per operation. + For Redis < 7, queues `count` individual LPOP(key) commands per operation. + """ + major_version = self._parse_redis_major_version() + + if major_version >= 7: + for lpop_op in lpop_list: + pipe.lpop(lpop_op["key"], lpop_op["count"]) + raw_results = await pipe.execute() + else: + # For Redis < 7, LPOP doesn't support count param. + # Issue `count` individual LPOP commands per key, all in one pipeline. + counts: List[int] = [] + for lpop_op in lpop_list: + count = lpop_op["count"] or 1 + counts.append(count) + for _ in range(count): + pipe.lpop(lpop_op["key"]) + flat_results = await pipe.execute() + + # Re-group the flat results back into per-key lists + raw_results = [] + offset = 0 + for count in counts: + key_results = [ + r for r in flat_results[offset : offset + count] if r is not None + ] + raw_results.append(key_results if key_results else None) + offset += count + + # Decode bytes -> str for each result set + decoded_results: List[Optional[List[str]]] = [] + for r in raw_results: + if r is None: + decoded_results.append(None) + elif isinstance(r, list): + try: + decoded_results.append( + [ + item.decode("utf-8") if isinstance(item, bytes) else item + for item in r + if item is not None + ] + or None + ) + except Exception: + decoded_results.append(r) # type: ignore + else: + decoded_results.append(None) + return decoded_results + + async def async_lpop_pipeline( + self, + lpop_list: List[RedisPipelineLpopOperation], + ) -> List[Optional[List[str]]]: + """ + Use Redis Pipelines for bulk LPOP operations + + Args: + lpop_list: List of RedisPipelineLpopOperation dicts containing: + - key: str + - count: Optional[int] + + Returns: + List[Optional[List[str]]]: Decoded results per key, None if key was empty + """ + if len(lpop_list) == 0: + return [] + + _redis_client: Any = self.init_async_client() + start_time = time.time() + + try: + async with _redis_client.pipeline(transaction=False) as pipe: + results = await self._pipeline_lpop_helper(pipe, lpop_list) + + ## LOGGING ## + end_time = time.time() + _duration = end_time - start_time + asyncio.create_task( + self.service_logger_obj.async_service_success_hook( + service=ServiceTypes.REDIS, + duration=_duration, + call_type=f"async_lpop_pipeline <- {_get_call_stack_info()}", + ) + ) + return results + except Exception as e: + ## LOGGING ## + end_time = time.time() + _duration = end_time - start_time + asyncio.create_task( + self.service_logger_obj.async_service_failure_hook( + service=ServiceTypes.REDIS, + duration=_duration, + error=e, + call_type=f"async_lpop_pipeline <- {_get_call_stack_info()}", + ) + ) + verbose_logger.error( + "LiteLLM Redis Caching: async_lpop_pipeline() - Got exception from REDIS %s", + str(e), + ) + raise e diff --git a/litellm/proxy/db/db_spend_update_writer.py b/litellm/proxy/db/db_spend_update_writer.py index 03628fda47f..b9cd794e780 100644 --- a/litellm/proxy/db/db_spend_update_writer.py +++ b/litellm/proxy/db/db_spend_update_writer.py @@ -509,9 +509,16 @@ class DBSpendUpdateWriter: verbose_proxy_logger.debug("acquired lock for spend updates") try: - db_spend_update_transactions = ( - await self.redis_update_buffer.get_all_update_transactions_from_redis_buffer() - ) + ( + db_spend_update_transactions, + daily_spend_update_transactions, + daily_team_spend_update_transactions, + daily_org_spend_update_transactions, + daily_end_user_spend_update_transactions, + daily_agent_spend_update_transactions, + daily_tag_spend_update_transactions, + ) = await self.redis_update_buffer.get_all_transactions_from_redis_buffer_pipeline() + if db_spend_update_transactions is not None: await self._commit_spend_updates_to_db( prisma_client=prisma_client, @@ -520,9 +527,6 @@ class DBSpendUpdateWriter: db_spend_update_transactions=db_spend_update_transactions, ) - daily_spend_update_transactions = ( - await self.redis_update_buffer.get_all_daily_spend_update_transactions_from_redis_buffer() - ) if daily_spend_update_transactions is not None: await DBSpendUpdateWriter.update_daily_user_spend( n_retry_times=n_retry_times, @@ -530,9 +534,6 @@ class DBSpendUpdateWriter: proxy_logging_obj=proxy_logging_obj, daily_spend_transactions=daily_spend_update_transactions, ) - daily_team_spend_update_transactions = ( - await self.redis_update_buffer.get_all_daily_team_spend_update_transactions_from_redis_buffer() - ) if daily_team_spend_update_transactions is not None: await DBSpendUpdateWriter.update_daily_team_spend( n_retry_times=n_retry_times, @@ -541,9 +542,6 @@ class DBSpendUpdateWriter: daily_spend_transactions=daily_team_spend_update_transactions, ) - daily_org_spend_update_transactions = ( - await self.redis_update_buffer.get_all_daily_org_spend_update_transactions_from_redis_buffer() - ) if daily_org_spend_update_transactions is not None: await DBSpendUpdateWriter.update_daily_org_spend( n_retry_times=n_retry_times, @@ -552,9 +550,6 @@ class DBSpendUpdateWriter: daily_spend_transactions=daily_org_spend_update_transactions, ) - daily_tag_spend_update_transactions = ( - await self.redis_update_buffer.get_all_daily_tag_spend_update_transactions_from_redis_buffer() - ) if daily_tag_spend_update_transactions is not None: await DBSpendUpdateWriter.update_daily_tag_spend( n_retry_times=n_retry_times, @@ -562,9 +557,6 @@ class DBSpendUpdateWriter: proxy_logging_obj=proxy_logging_obj, daily_spend_transactions=daily_tag_spend_update_transactions, ) - daily_end_user_spend_update_transactions = ( - await self.redis_update_buffer.get_all_daily_end_user_spend_update_transactions_from_redis_buffer() - ) if daily_end_user_spend_update_transactions is not None: await DBSpendUpdateWriter.update_daily_end_user_spend( n_retry_times=n_retry_times, @@ -572,9 +564,6 @@ class DBSpendUpdateWriter: proxy_logging_obj=proxy_logging_obj, daily_spend_transactions=daily_end_user_spend_update_transactions, ) - daily_agent_spend_update_transactions = ( - await self.redis_update_buffer.get_all_daily_agent_spend_update_transactions_from_redis_buffer() - ) if daily_agent_spend_update_transactions is not None: await DBSpendUpdateWriter.update_daily_agent_spend( n_retry_times=n_retry_times, diff --git a/litellm/proxy/db/db_transaction_queue/redis_update_buffer.py b/litellm/proxy/db/db_transaction_queue/redis_update_buffer.py index 37b42e26bc9..ef8dbb2c7b8 100644 --- a/litellm/proxy/db/db_transaction_queue/redis_update_buffer.py +++ b/litellm/proxy/db/db_transaction_queue/redis_update_buffer.py @@ -6,7 +6,7 @@ This is to prevent deadlocks and improve reliability import asyncio import json -from typing import TYPE_CHECKING, Any, Dict, List, Optional, Union, cast +from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, Union, cast from litellm._logging import verbose_proxy_logger from litellm.caching import RedisCache @@ -36,6 +36,7 @@ from litellm.proxy.db.db_transaction_queue.daily_spend_update_queue import ( ) from litellm.proxy.db.db_transaction_queue.spend_update_queue import SpendUpdateQueue from litellm.secret_managers.main import str_to_bool +from litellm.types.caching import RedisPipelineLpopOperation, RedisPipelineRpushOperation from litellm.types.services import ServiceTypes if TYPE_CHECKING: @@ -195,47 +196,44 @@ class RedisUpdateBuffer: "ALL DAILY SPEND UPDATE TRANSACTIONS: %s", daily_spend_update_transactions ) - await self._store_transactions_in_redis( - transactions=db_spend_update_transactions, - redis_key=REDIS_UPDATE_BUFFER_KEY, - service_type=ServiceTypes.REDIS_SPEND_UPDATE_QUEUE, + # Build a list of rpush operations, skipping empty/None transaction sets + _queue_configs: List[Tuple[Any, str, ServiceTypes]] = [ + (db_spend_update_transactions, REDIS_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_SPEND_UPDATE_QUEUE), + (daily_spend_update_transactions, REDIS_DAILY_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_SPEND_UPDATE_QUEUE), + (daily_team_spend_update_transactions, REDIS_DAILY_TEAM_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_TEAM_SPEND_UPDATE_QUEUE), + (daily_org_spend_update_transactions, REDIS_DAILY_ORG_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_SPEND_UPDATE_QUEUE), + (daily_end_user_spend_update_transactions, REDIS_DAILY_END_USER_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_END_USER_SPEND_UPDATE_QUEUE), + (daily_agent_spend_update_transactions, REDIS_DAILY_AGENT_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_AGENT_SPEND_UPDATE_QUEUE), + (daily_tag_spend_update_transactions, REDIS_DAILY_TAG_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_TAG_SPEND_UPDATE_QUEUE), + ] + + rpush_list: List[RedisPipelineRpushOperation] = [] + service_types: List[ServiceTypes] = [] + for transactions, redis_key, service_type in _queue_configs: + if transactions is None or len(transactions) == 0: + continue + rpush_list.append( + RedisPipelineRpushOperation( + key=redis_key, + values=[safe_dumps(transactions)], + ) + ) + service_types.append(service_type) + + if len(rpush_list) == 0: + return + + result_lengths = await self.redis_cache.async_rpush_pipeline( + rpush_list=rpush_list, ) - await self._store_transactions_in_redis( - transactions=daily_spend_update_transactions, - redis_key=REDIS_DAILY_SPEND_UPDATE_BUFFER_KEY, - service_type=ServiceTypes.REDIS_DAILY_SPEND_UPDATE_QUEUE, - ) - - await self._store_transactions_in_redis( - transactions=daily_team_spend_update_transactions, - redis_key=REDIS_DAILY_TEAM_SPEND_UPDATE_BUFFER_KEY, - service_type=ServiceTypes.REDIS_DAILY_TEAM_SPEND_UPDATE_QUEUE, - ) - - await self._store_transactions_in_redis( - transactions=daily_org_spend_update_transactions, - redis_key=REDIS_DAILY_ORG_SPEND_UPDATE_BUFFER_KEY, - service_type=ServiceTypes.REDIS_DAILY_SPEND_UPDATE_QUEUE, - ) - - await self._store_transactions_in_redis( - transactions=daily_end_user_spend_update_transactions, - redis_key=REDIS_DAILY_END_USER_SPEND_UPDATE_BUFFER_KEY, - service_type=ServiceTypes.REDIS_DAILY_END_USER_SPEND_UPDATE_QUEUE, - ) - - await self._store_transactions_in_redis( - transactions=daily_agent_spend_update_transactions, - redis_key=REDIS_DAILY_AGENT_SPEND_UPDATE_BUFFER_KEY, - service_type=ServiceTypes.REDIS_DAILY_AGENT_SPEND_UPDATE_QUEUE, - ) - - await self._store_transactions_in_redis( - transactions=daily_tag_spend_update_transactions, - redis_key=REDIS_DAILY_TAG_SPEND_UPDATE_BUFFER_KEY, - service_type=ServiceTypes.REDIS_DAILY_TAG_SPEND_UPDATE_QUEUE, - ) + # Emit gauge events for each queue + for i, queue_size in enumerate(result_lengths): + if i < len(service_types): + await self._emit_new_item_added_to_redis_buffer_event( + queue_size=queue_size, + service=service_types[i], + ) @staticmethod def _number_of_transactions_to_store_in_redis( @@ -317,6 +315,77 @@ class RedisUpdateBuffer: return combined_transaction + async def get_all_transactions_from_redis_buffer_pipeline( + self, + ) -> Tuple[ + Optional[DBSpendUpdateTransactions], + Optional[Dict[str, DailyUserSpendTransaction]], + Optional[Dict[str, DailyTeamSpendTransaction]], + Optional[Dict[str, DailyOrganizationSpendTransaction]], + Optional[Dict[str, DailyEndUserSpendTransaction]], + Optional[Dict[str, DailyAgentSpendTransaction]], + Optional[Dict[str, DailyTagSpendTransaction]], + ]: + """ + Drains all 7 Redis buffer queues in a single pipeline round-trip. + + Returns a 7-tuple of parsed results in this order: + 0: DBSpendUpdateTransactions + 1: daily user spend + 2: daily team spend + 3: daily org spend + 4: daily end-user spend + 5: daily agent spend + 6: daily tag spend + """ + if self.redis_cache is None: + return None, None, None, None, None, None, None + + lpop_list: List[RedisPipelineLpopOperation] = [ + RedisPipelineLpopOperation(key=REDIS_UPDATE_BUFFER_KEY, count=MAX_REDIS_BUFFER_DEQUEUE_COUNT), + RedisPipelineLpopOperation(key=REDIS_DAILY_SPEND_UPDATE_BUFFER_KEY, count=MAX_REDIS_BUFFER_DEQUEUE_COUNT), + RedisPipelineLpopOperation(key=REDIS_DAILY_TEAM_SPEND_UPDATE_BUFFER_KEY, count=MAX_REDIS_BUFFER_DEQUEUE_COUNT), + RedisPipelineLpopOperation(key=REDIS_DAILY_ORG_SPEND_UPDATE_BUFFER_KEY, count=MAX_REDIS_BUFFER_DEQUEUE_COUNT), + RedisPipelineLpopOperation(key=REDIS_DAILY_END_USER_SPEND_UPDATE_BUFFER_KEY, count=MAX_REDIS_BUFFER_DEQUEUE_COUNT), + RedisPipelineLpopOperation(key=REDIS_DAILY_AGENT_SPEND_UPDATE_BUFFER_KEY, count=MAX_REDIS_BUFFER_DEQUEUE_COUNT), + RedisPipelineLpopOperation(key=REDIS_DAILY_TAG_SPEND_UPDATE_BUFFER_KEY, count=MAX_REDIS_BUFFER_DEQUEUE_COUNT), + ] + + raw_results = await self.redis_cache.async_lpop_pipeline(lpop_list=lpop_list) + + # Pad with None if pipeline returned fewer results than expected + while len(raw_results) < 7: + raw_results.append(None) + + # Slot 0: DBSpendUpdateTransactions + db_spend: Optional[DBSpendUpdateTransactions] = None + if raw_results[0] is not None: + parsed = self._parse_list_of_transactions(raw_results[0]) + if len(parsed) > 0: + db_spend = self._combine_list_of_transactions(parsed) + + # Slots 1-6: daily spend categories + daily_results: List[Optional[Dict[str, Any]]] = [] + for slot in range(1, 7): + if raw_results[slot] is None: + daily_results.append(None) + else: + list_of_daily = [json.loads(t) for t in raw_results[slot]] # type: ignore + aggregated = DailySpendUpdateQueue.get_aggregated_daily_spend_update_transactions( + list_of_daily + ) + daily_results.append(aggregated) + + return ( + db_spend, + cast(Optional[Dict[str, DailyUserSpendTransaction]], daily_results[0]), + cast(Optional[Dict[str, DailyTeamSpendTransaction]], daily_results[1]), + cast(Optional[Dict[str, DailyOrganizationSpendTransaction]], daily_results[2]), + cast(Optional[Dict[str, DailyEndUserSpendTransaction]], daily_results[3]), + cast(Optional[Dict[str, DailyAgentSpendTransaction]], daily_results[4]), + cast(Optional[Dict[str, DailyTagSpendTransaction]], daily_results[5]), + ) + async def get_all_daily_spend_update_transactions_from_redis_buffer( self, ) -> Optional[Dict[str, DailyUserSpendTransaction]]: diff --git a/litellm/types/caching.py b/litellm/types/caching.py index ad2ffeaf9bd..7126ba3e9b9 100644 --- a/litellm/types/caching.py +++ b/litellm/types/caching.py @@ -52,6 +52,24 @@ class RedisPipelineSetOperation(TypedDict): ttl: Optional[int] +class RedisPipelineRpushOperation(TypedDict): + """ + TypedDict for 1 Redis Pipeline RPUSH Operation + """ + + key: str + values: List[Any] + + +class RedisPipelineLpopOperation(TypedDict): + """ + TypedDict for 1 Redis Pipeline LPOP Operation + """ + + key: str + count: Optional[int] + + DynamicCacheControl = TypedDict( "DynamicCacheControl", { diff --git a/tests/test_litellm/caching/test_redis_cache.py b/tests/test_litellm/caching/test_redis_cache.py index f3e7953b8ae..31eeb854883 100644 --- a/tests/test_litellm/caching/test_redis_cache.py +++ b/tests/test_litellm/caching/test_redis_cache.py @@ -122,6 +122,219 @@ async def test_handle_lpop_count_for_older_redis_versions(monkeypatch): assert mock_pipeline.execute.call_count == 2 +@pytest.mark.asyncio +async def test_async_rpush_pipeline_executes_all_operations(monkeypatch, redis_no_ping): + """Verify that multiple rpush ops are batched into a single pipeline execute""" + monkeypatch.setenv("REDIS_HOST", "https://my-test-host") + redis_cache = RedisCache() + + mock_redis_instance = AsyncMock() + mock_pipeline = MagicMock() + mock_pipeline.__aenter__ = AsyncMock(return_value=mock_pipeline) + mock_pipeline.__aexit__ = AsyncMock(return_value=None) + mock_pipeline.rpush = MagicMock() + mock_pipeline.execute = AsyncMock(return_value=[3, 5, 1]) + mock_redis_instance.pipeline = MagicMock(return_value=mock_pipeline) + + from litellm.types.caching import RedisPipelineRpushOperation + + rpush_list = [ + RedisPipelineRpushOperation(key="key1", values=["a", "b"]), + RedisPipelineRpushOperation(key="key2", values=["c"]), + RedisPipelineRpushOperation(key="key3", values=["d", "e", "f"]), + ] + + with patch.object(redis_cache, "init_async_client", return_value=mock_redis_instance): + result = await redis_cache.async_rpush_pipeline(rpush_list=rpush_list) + + assert result == [3, 5, 1] + assert mock_pipeline.rpush.call_count == 3 + mock_pipeline.rpush.assert_any_call("key1", "a", "b") + mock_pipeline.rpush.assert_any_call("key2", "c") + mock_pipeline.rpush.assert_any_call("key3", "d", "e", "f") + mock_pipeline.execute.assert_called_once() + + +@pytest.mark.asyncio +async def test_async_rpush_pipeline_empty_list_returns_empty(monkeypatch, redis_no_ping): + """Empty rpush_list should return empty list without touching Redis""" + monkeypatch.setenv("REDIS_HOST", "https://my-test-host") + redis_cache = RedisCache() + + mock_redis_instance = AsyncMock() + + with patch.object(redis_cache, "init_async_client", return_value=mock_redis_instance): + result = await redis_cache.async_rpush_pipeline(rpush_list=[]) + + assert result == [] + mock_redis_instance.pipeline.assert_not_called() + + +@pytest.mark.asyncio +async def test_async_rpush_pipeline_raises_on_redis_error(monkeypatch, redis_no_ping): + """Pipeline errors should propagate""" + monkeypatch.setenv("REDIS_HOST", "https://my-test-host") + redis_cache = RedisCache() + + mock_redis_instance = AsyncMock() + mock_pipeline = MagicMock() + mock_pipeline.__aenter__ = AsyncMock(return_value=mock_pipeline) + mock_pipeline.__aexit__ = AsyncMock(return_value=None) + mock_pipeline.rpush = MagicMock() + mock_pipeline.execute = AsyncMock(side_effect=ConnectionError("Redis down")) + mock_redis_instance.pipeline = MagicMock(return_value=mock_pipeline) + + from litellm.types.caching import RedisPipelineRpushOperation + + rpush_list = [RedisPipelineRpushOperation(key="key1", values=["a"])] + + with patch.object(redis_cache, "init_async_client", return_value=mock_redis_instance): + with pytest.raises(ConnectionError, match="Redis down"): + await redis_cache.async_rpush_pipeline(rpush_list=rpush_list) + + +@pytest.mark.asyncio +async def test_async_lpop_pipeline_single_round_trip(monkeypatch, redis_no_ping): + """Verify that multiple lpop ops are batched into a single pipeline execute""" + monkeypatch.setenv("REDIS_HOST", "https://my-test-host") + redis_cache = RedisCache() + redis_cache.redis_version = "7.0.0" + + mock_redis_instance = AsyncMock() + mock_pipeline = MagicMock() + mock_pipeline.__aenter__ = AsyncMock(return_value=mock_pipeline) + mock_pipeline.__aexit__ = AsyncMock(return_value=None) + mock_pipeline.lpop = MagicMock() + mock_pipeline.execute = AsyncMock(return_value=[ + [b"val1", b"val2"], # key1 results + None, # key2 empty + [b"val3"], # key3 results + ]) + mock_redis_instance.pipeline = MagicMock(return_value=mock_pipeline) + + from litellm.types.caching import RedisPipelineLpopOperation + + lpop_list = [ + RedisPipelineLpopOperation(key="key1", count=10), + RedisPipelineLpopOperation(key="key2", count=10), + RedisPipelineLpopOperation(key="key3", count=5), + ] + + with patch.object(redis_cache, "init_async_client", return_value=mock_redis_instance): + results = await redis_cache.async_lpop_pipeline(lpop_list=lpop_list) + + assert len(results) == 3 + assert results[0] == ["val1", "val2"] + assert results[1] is None + assert results[2] == ["val3"] + mock_pipeline.execute.assert_called_once() + + +@pytest.mark.asyncio +async def test_async_lpop_pipeline_redis_lt7_regroups_flat_results(monkeypatch, redis_no_ping): + """Verify Redis < 7 fallback issues individual LPOPs and regroups correctly""" + monkeypatch.setenv("REDIS_HOST", "https://my-test-host") + redis_cache = RedisCache() + redis_cache.redis_version = "6.2.0" + + mock_redis_instance = AsyncMock() + mock_pipeline = MagicMock() + mock_pipeline.__aenter__ = AsyncMock(return_value=mock_pipeline) + mock_pipeline.__aexit__ = AsyncMock(return_value=None) + mock_pipeline.lpop = MagicMock() + + # With count=3 for key1 and count=2 for key2, we get 5 individual LPOP commands + # Simulate: key1 has 2 values then None, key2 has 1 value then None + mock_pipeline.execute = AsyncMock(return_value=[ + b"val1", b"val2", None, # 3 LPOPs for key1 + b"val3", None, # 2 LPOPs for key2 + ]) + mock_redis_instance.pipeline = MagicMock(return_value=mock_pipeline) + + from litellm.types.caching import RedisPipelineLpopOperation + + lpop_list = [ + RedisPipelineLpopOperation(key="key1", count=3), + RedisPipelineLpopOperation(key="key2", count=2), + ] + + with patch.object(redis_cache, "init_async_client", return_value=mock_redis_instance): + results = await redis_cache.async_lpop_pipeline(lpop_list=lpop_list) + + assert len(results) == 2 + assert results[0] == ["val1", "val2"] # 2 values, None filtered out + assert results[1] == ["val3"] # 1 value, None filtered out + # All 5 individual LPOPs should be queued, but only 1 execute() call + assert mock_pipeline.lpop.call_count == 5 + mock_pipeline.execute.assert_called_once() + + +@pytest.mark.asyncio +async def test_async_rpush_pipeline_raises_on_per_command_error(monkeypatch, redis_no_ping): + """Verify that per-command errors in pipeline results are raised, not silently dropped""" + monkeypatch.setenv("REDIS_HOST", "https://my-test-host") + redis_cache = RedisCache() + + mock_redis_instance = AsyncMock() + mock_pipeline = MagicMock() + mock_pipeline.__aenter__ = AsyncMock(return_value=mock_pipeline) + mock_pipeline.__aexit__ = AsyncMock(return_value=None) + mock_pipeline.rpush = MagicMock() + # Simulate: first RPUSH succeeds, second returns a per-command error + mock_pipeline.execute = AsyncMock(return_value=[3, Exception("WRONGTYPE")]) + mock_redis_instance.pipeline = MagicMock(return_value=mock_pipeline) + + from litellm.types.caching import RedisPipelineRpushOperation + + rpush_list = [ + RedisPipelineRpushOperation(key="key1", values=["a"]), + RedisPipelineRpushOperation(key="key2", values=["b"]), + ] + + with patch.object(redis_cache, "init_async_client", return_value=mock_redis_instance): + with pytest.raises(Exception, match="WRONGTYPE"): + await redis_cache.async_rpush_pipeline(rpush_list=rpush_list) + + +@pytest.mark.asyncio +async def test_async_lpop_pipeline_empty_list(monkeypatch, redis_no_ping): + """Empty lpop_list should return empty list without touching Redis""" + monkeypatch.setenv("REDIS_HOST", "https://my-test-host") + redis_cache = RedisCache() + + mock_redis_instance = AsyncMock() + + with patch.object(redis_cache, "init_async_client", return_value=mock_redis_instance): + result = await redis_cache.async_lpop_pipeline(lpop_list=[]) + + assert result == [] + mock_redis_instance.pipeline.assert_not_called() + + +@pytest.mark.asyncio +async def test_async_lpop_pipeline_propagates_redis_exception(monkeypatch, redis_no_ping): + """Pipeline errors should propagate""" + monkeypatch.setenv("REDIS_HOST", "https://my-test-host") + redis_cache = RedisCache() + redis_cache.redis_version = "7.0.0" + + mock_redis_instance = AsyncMock() + mock_pipeline = MagicMock() + mock_pipeline.__aenter__ = AsyncMock(return_value=mock_pipeline) + mock_pipeline.__aexit__ = AsyncMock(return_value=None) + mock_pipeline.lpop = MagicMock() + mock_pipeline.execute = AsyncMock(side_effect=ConnectionError("Redis down")) + mock_redis_instance.pipeline = MagicMock(return_value=mock_pipeline) + + from litellm.types.caching import RedisPipelineLpopOperation + + lpop_list = [RedisPipelineLpopOperation(key="key1", count=10)] + + with patch.object(redis_cache, "init_async_client", return_value=mock_redis_instance): + with pytest.raises(ConnectionError, match="Redis down"): + await redis_cache.async_lpop_pipeline(lpop_list=lpop_list) + + @pytest.mark.asyncio @pytest.mark.parametrize( "redis_version", diff --git a/tests/test_litellm/proxy/db/db_transaction_queue/test_redis_update_buffer.py b/tests/test_litellm/proxy/db/db_transaction_queue/test_redis_update_buffer.py new file mode 100644 index 00000000000..2a380370c30 --- /dev/null +++ b/tests/test_litellm/proxy/db/db_transaction_queue/test_redis_update_buffer.py @@ -0,0 +1,194 @@ +import json +import os +import sys +from unittest.mock import AsyncMock, MagicMock, patch + +import pytest + +sys.path.insert( + 0, os.path.abspath("../../../../..") +) # Adds the parent directory to the system path + +from litellm.proxy.db.db_transaction_queue.redis_update_buffer import RedisUpdateBuffer +from litellm.types.caching import RedisPipelineRpushOperation + + +@pytest.fixture +def mock_redis_cache(): + """Create a mock RedisCache instance""" + mock = AsyncMock() + return mock + + +@pytest.fixture +def redis_update_buffer(mock_redis_cache): + """Create a RedisUpdateBuffer with a mock RedisCache""" + return RedisUpdateBuffer(redis_cache=mock_redis_cache) + + +@pytest.mark.asyncio +async def test_store_in_memory_spend_updates_uses_pipeline(redis_update_buffer, mock_redis_cache): + """ + Verify store_in_memory_spend_updates_in_redis calls async_rpush_pipeline once + with the correct operations and skips empty queues. + """ + mock_redis_cache.async_rpush_pipeline = AsyncMock(return_value=[3, 5, 2]) + + # Create mock queues - only 3 of 7 have data + spend_update_queue = AsyncMock() + spend_update_queue.flush_and_get_aggregated_db_spend_update_transactions = AsyncMock( + return_value={"key_list_transactions": {"key1": 1.0}} + ) + + daily_spend_queue = AsyncMock() + daily_spend_queue.flush_and_get_aggregated_daily_spend_update_transactions = AsyncMock( + return_value={"user_key1": {"spend": 1.0}} + ) + + daily_team_queue = AsyncMock() + daily_team_queue.flush_and_get_aggregated_daily_spend_update_transactions = AsyncMock( + return_value={"team_key1": {"spend": 2.0}} + ) + + # Empty queues + daily_org_queue = AsyncMock() + daily_org_queue.flush_and_get_aggregated_daily_spend_update_transactions = AsyncMock( + return_value={} + ) + + daily_end_user_queue = AsyncMock() + daily_end_user_queue.flush_and_get_aggregated_daily_spend_update_transactions = AsyncMock( + return_value=None + ) + + daily_agent_queue = AsyncMock() + daily_agent_queue.flush_and_get_aggregated_daily_spend_update_transactions = AsyncMock( + return_value={} + ) + + daily_tag_queue = AsyncMock() + daily_tag_queue.flush_and_get_aggregated_daily_spend_update_transactions = AsyncMock( + return_value={} + ) + + await redis_update_buffer.store_in_memory_spend_updates_in_redis( + spend_update_queue=spend_update_queue, + daily_spend_update_queue=daily_spend_queue, + daily_team_spend_update_queue=daily_team_queue, + daily_org_spend_update_queue=daily_org_queue, + daily_end_user_spend_update_queue=daily_end_user_queue, + daily_agent_spend_update_queue=daily_agent_queue, + daily_tag_spend_update_queue=daily_tag_queue, + ) + + # Should be called exactly once (pipeline) + mock_redis_cache.async_rpush_pipeline.assert_called_once() + + # Verify only 3 operations were included (empty ones skipped) + call_args = mock_redis_cache.async_rpush_pipeline.call_args + rpush_list = call_args.kwargs["rpush_list"] + assert len(rpush_list) == 3 + + +@pytest.mark.asyncio +async def test_store_in_memory_spend_updates_all_empty_returns_early( + redis_update_buffer, mock_redis_cache +): + """ + When all queues are empty, pipeline should never be called. + """ + mock_redis_cache.async_rpush_pipeline = AsyncMock() + + # All queues return empty + empty_queue = AsyncMock() + empty_queue.flush_and_get_aggregated_db_spend_update_transactions = AsyncMock( + return_value={} + ) + empty_daily_queue = AsyncMock() + empty_daily_queue.flush_and_get_aggregated_daily_spend_update_transactions = AsyncMock( + return_value={} + ) + + await redis_update_buffer.store_in_memory_spend_updates_in_redis( + spend_update_queue=empty_queue, + daily_spend_update_queue=empty_daily_queue, + daily_team_spend_update_queue=empty_daily_queue, + daily_org_spend_update_queue=empty_daily_queue, + daily_end_user_spend_update_queue=empty_daily_queue, + daily_agent_spend_update_queue=empty_daily_queue, + daily_tag_spend_update_queue=empty_daily_queue, + ) + + mock_redis_cache.async_rpush_pipeline.assert_not_called() + + +@pytest.mark.asyncio +async def test_get_all_transactions_from_redis_buffer_pipeline( + redis_update_buffer, mock_redis_cache +): + """ + Verify get_all_transactions_from_redis_buffer_pipeline correctly parses + and aggregates results from async_lpop_pipeline. + """ + # Simulate pipeline results: slot 0 = spend updates, slots 1-6 = daily categories + db_spend_json = json.dumps( + { + "key_list_transactions": {"key1": 1.0, "key2": 2.0}, + "user_list_transactions": {"user1": 0.5}, + "end_user_list_transactions": {}, + "team_list_transactions": {}, + "team_member_list_transactions": {}, + "org_list_transactions": {}, + "tag_list_transactions": {}, + } + ) + daily_user_json = json.dumps({"user_key1": {"spend": 1.0, "api_requests": 1}}) + daily_team_json = json.dumps({"team_key1": {"spend": 2.0, "api_requests": 2}}) + + mock_redis_cache.async_lpop_pipeline = AsyncMock( + return_value=[ + [db_spend_json], # slot 0: db spend updates + [daily_user_json], # slot 1: daily user + [daily_team_json], # slot 2: daily team + None, # slot 3: daily org (empty) + None, # slot 4: daily end-user (empty) + None, # slot 5: daily agent (empty) + None, # slot 6: daily tag (empty) + ] + ) + + result = await redis_update_buffer.get_all_transactions_from_redis_buffer_pipeline() + + assert len(result) == 7 + db_spend, daily_user, daily_team, daily_org, daily_end_user, daily_agent, daily_tag = result + + # Verify db spend was parsed correctly + assert db_spend is not None + assert db_spend["key_list_transactions"]["key1"] == 1.0 + assert db_spend["key_list_transactions"]["key2"] == 2.0 + assert db_spend["user_list_transactions"]["user1"] == 0.5 + + # Verify daily user was parsed + assert daily_user is not None + assert daily_user["user_key1"]["spend"] == 1.0 + + # Verify daily team was parsed + assert daily_team is not None + assert daily_team["team_key1"]["spend"] == 2.0 + + # Verify empty slots + assert daily_org is None + assert daily_end_user is None + assert daily_agent is None + assert daily_tag is None + + # Verify pipeline was called once with correct keys + mock_redis_cache.async_lpop_pipeline.assert_called_once() + + +@pytest.mark.asyncio +async def test_get_all_transactions_from_redis_buffer_pipeline_no_redis(): + """When redis_cache is None, should return all Nones""" + buffer = RedisUpdateBuffer(redis_cache=None) + result = await buffer.get_all_transactions_from_redis_buffer_pipeline() + assert result == (None, None, None, None, None, None, None) diff --git a/tests/test_litellm/proxy/db/test_db_spend_update_writer.py b/tests/test_litellm/proxy/db/test_db_spend_update_writer.py index 0fa0d4cf10b..2a0428e2f07 100644 --- a/tests/test_litellm/proxy/db/test_db_spend_update_writer.py +++ b/tests/test_litellm/proxy/db/test_db_spend_update_writer.py @@ -1076,3 +1076,46 @@ async def test_commit_key_spend_updates_includes_last_active(): last_active = call_kwargs["data"]["last_active"] assert isinstance(last_active, datetime) assert before_call <= last_active <= after_call + + +@pytest.mark.asyncio +async def test_commit_spend_updates_uses_pipeline(): + """ + Verify that _commit_spend_updates_to_db_with_redis uses + get_all_transactions_from_redis_buffer_pipeline instead of 7 individual calls. + """ + db_writer = DBSpendUpdateWriter() + + mock_redis_update_buffer = AsyncMock() + mock_redis_update_buffer.store_in_memory_spend_updates_in_redis = AsyncMock() + # Return all-None tuple (no data to commit) + mock_redis_update_buffer.get_all_transactions_from_redis_buffer_pipeline = AsyncMock( + return_value=(None, None, None, None, None, None, None) + ) + db_writer.redis_update_buffer = mock_redis_update_buffer + + mock_pod_lock_manager = AsyncMock() + mock_pod_lock_manager.acquire_lock = AsyncMock(return_value=True) + mock_pod_lock_manager.release_lock = AsyncMock() + db_writer.pod_lock_manager = mock_pod_lock_manager + + mock_prisma_client = MagicMock() + mock_proxy_logging = MagicMock() + + await db_writer._commit_spend_updates_to_db_with_redis( + prisma_client=mock_prisma_client, + n_retry_times=1, + proxy_logging_obj=mock_proxy_logging, + ) + + # Pipeline method should be called once + mock_redis_update_buffer.get_all_transactions_from_redis_buffer_pipeline.assert_called_once() + + # Individual methods should NOT be called + mock_redis_update_buffer.get_all_update_transactions_from_redis_buffer.assert_not_called() + mock_redis_update_buffer.get_all_daily_spend_update_transactions_from_redis_buffer.assert_not_called() + mock_redis_update_buffer.get_all_daily_team_spend_update_transactions_from_redis_buffer.assert_not_called() + mock_redis_update_buffer.get_all_daily_org_spend_update_transactions_from_redis_buffer.assert_not_called() + mock_redis_update_buffer.get_all_daily_end_user_spend_update_transactions_from_redis_buffer.assert_not_called() + mock_redis_update_buffer.get_all_daily_agent_spend_update_transactions_from_redis_buffer.assert_not_called() + mock_redis_update_buffer.get_all_daily_tag_spend_update_transactions_from_redis_buffer.assert_not_called() From 68a30a39e68d631959920f7c658baaf8bea2df06 Mon Sep 17 00:00:00 2001 From: Ryan Crabbe Date: Tue, 24 Feb 2026 14:22:57 -0800 Subject: [PATCH 04/15] fix(proxy): add LPOP pipeline error checking and fix org spend ServiceType - Add per-command error check in _pipeline_lpop_helper to match _pipeline_rpush_helper, preventing silent data loss on WRONGTYPE errors - Fix pre-existing bug: org spend queue metric was using REDIS_DAILY_SPEND_UPDATE_QUEUE instead of REDIS_DAILY_ORG_SPEND_UPDATE_QUEUE - Add test for per-command LPOP pipeline error propagation --- litellm/caching/redis_cache.py | 5 ++++ .../redis_update_buffer.py | 2 +- .../test_litellm/caching/test_redis_cache.py | 30 +++++++++++++++++++ 3 files changed, 36 insertions(+), 1 deletion(-) diff --git a/litellm/caching/redis_cache.py b/litellm/caching/redis_cache.py index 9b93d890cec..fa9b94bc2ac 100644 --- a/litellm/caching/redis_cache.py +++ b/litellm/caching/redis_cache.py @@ -1511,6 +1511,11 @@ class RedisCache(BaseCache): raw_results.append(key_results if key_results else None) offset += count + # Raise on per-command errors (matches _pipeline_rpush_helper behavior) + for r in raw_results: + if isinstance(r, Exception): + raise r + # Decode bytes -> str for each result set decoded_results: List[Optional[List[str]]] = [] for r in raw_results: diff --git a/litellm/proxy/db/db_transaction_queue/redis_update_buffer.py b/litellm/proxy/db/db_transaction_queue/redis_update_buffer.py index ef8dbb2c7b8..504fc5aba02 100644 --- a/litellm/proxy/db/db_transaction_queue/redis_update_buffer.py +++ b/litellm/proxy/db/db_transaction_queue/redis_update_buffer.py @@ -201,7 +201,7 @@ class RedisUpdateBuffer: (db_spend_update_transactions, REDIS_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_SPEND_UPDATE_QUEUE), (daily_spend_update_transactions, REDIS_DAILY_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_SPEND_UPDATE_QUEUE), (daily_team_spend_update_transactions, REDIS_DAILY_TEAM_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_TEAM_SPEND_UPDATE_QUEUE), - (daily_org_spend_update_transactions, REDIS_DAILY_ORG_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_SPEND_UPDATE_QUEUE), + (daily_org_spend_update_transactions, REDIS_DAILY_ORG_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_ORG_SPEND_UPDATE_QUEUE), (daily_end_user_spend_update_transactions, REDIS_DAILY_END_USER_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_END_USER_SPEND_UPDATE_QUEUE), (daily_agent_spend_update_transactions, REDIS_DAILY_AGENT_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_AGENT_SPEND_UPDATE_QUEUE), (daily_tag_spend_update_transactions, REDIS_DAILY_TAG_SPEND_UPDATE_BUFFER_KEY, ServiceTypes.REDIS_DAILY_TAG_SPEND_UPDATE_QUEUE), diff --git a/tests/test_litellm/caching/test_redis_cache.py b/tests/test_litellm/caching/test_redis_cache.py index 31eeb854883..82606511826 100644 --- a/tests/test_litellm/caching/test_redis_cache.py +++ b/tests/test_litellm/caching/test_redis_cache.py @@ -296,6 +296,36 @@ async def test_async_rpush_pipeline_raises_on_per_command_error(monkeypatch, red await redis_cache.async_rpush_pipeline(rpush_list=rpush_list) +@pytest.mark.asyncio +async def test_async_lpop_pipeline_raises_on_per_command_error(monkeypatch, redis_no_ping): + """Verify that per-command errors in LPOP pipeline results are raised, not silently dropped""" + monkeypatch.setenv("REDIS_HOST", "https://my-test-host") + redis_cache = RedisCache() + redis_cache.redis_version = "7.0.0" + + mock_redis_instance = AsyncMock() + mock_pipeline = MagicMock() + mock_pipeline.__aenter__ = AsyncMock(return_value=mock_pipeline) + mock_pipeline.__aexit__ = AsyncMock(return_value=None) + mock_pipeline.lpop = MagicMock() + # Simulate: first LPOP succeeds, second returns a per-command error + mock_pipeline.execute = AsyncMock( + return_value=[[b"val1"], Exception("WRONGTYPE")] + ) + mock_redis_instance.pipeline = MagicMock(return_value=mock_pipeline) + + from litellm.types.caching import RedisPipelineLpopOperation + + lpop_list = [ + RedisPipelineLpopOperation(key="key1", count=10), + RedisPipelineLpopOperation(key="key2", count=10), + ] + + with patch.object(redis_cache, "init_async_client", return_value=mock_redis_instance): + with pytest.raises(Exception, match="WRONGTYPE"): + await redis_cache.async_lpop_pipeline(lpop_list=lpop_list) + + @pytest.mark.asyncio async def test_async_lpop_pipeline_empty_list(monkeypatch, redis_no_ping): """Empty lpop_list should return empty list without touching Redis""" From b78a30f7737618a2dc6b964cb96b27ccb4ed264b Mon Sep 17 00:00:00 2001 From: yuneng-jiang Date: Tue, 24 Feb 2026 21:04:53 -0800 Subject: [PATCH 05/15] [Feature] Add request_duration_ms to SpendLogs Add a `request_duration_ms` column to `LiteLLM_SpendLogs` to track request duration. New rows are computed at write time. Legacy rows use a COALESCE fallback in the `/spend/logs/ui` query to compute duration on the fly from `endTime - startTime`. The field is also sortable in the UI endpoint. Co-Authored-By: Claude Opus 4.6 (1M context) --- .../migration.sql | 2 + .../litellm_proxy_extras/schema.prisma | 1 + litellm/proxy/_types.py | 1 + litellm/proxy/schema.prisma | 1 + .../spend_management_endpoints.py | 5 +- .../spend_tracking/spend_tracking_utils.py | 11 +++ schema.prisma | 1 + .../test_spend_management_endpoints.py | 77 +++++++++++++++++++ .../test_spend_tracking_utils.py | 48 ++++++++++++ 9 files changed, 145 insertions(+), 2 deletions(-) create mode 100644 litellm-proxy-extras/litellm_proxy_extras/migrations/20260224201417_spend_logs_request_duration/migration.sql diff --git a/litellm-proxy-extras/litellm_proxy_extras/migrations/20260224201417_spend_logs_request_duration/migration.sql b/litellm-proxy-extras/litellm_proxy_extras/migrations/20260224201417_spend_logs_request_duration/migration.sql new file mode 100644 index 00000000000..604c1b16ff5 --- /dev/null +++ b/litellm-proxy-extras/litellm_proxy_extras/migrations/20260224201417_spend_logs_request_duration/migration.sql @@ -0,0 +1,2 @@ +-- AlterTable +ALTER TABLE "LiteLLM_SpendLogs" ADD COLUMN "request_duration_ms" INTEGER; \ No newline at end of file diff --git a/litellm-proxy-extras/litellm_proxy_extras/schema.prisma b/litellm-proxy-extras/litellm_proxy_extras/schema.prisma index 4af7484148c..5fa805e444e 100644 --- a/litellm-proxy-extras/litellm_proxy_extras/schema.prisma +++ b/litellm-proxy-extras/litellm_proxy_extras/schema.prisma @@ -477,6 +477,7 @@ model LiteLLM_SpendLogs { completion_tokens Int @default(0) startTime DateTime // Assuming start_time is a DateTime field endTime DateTime // Assuming end_time is a DateTime field + request_duration_ms Int? completionStartTime DateTime? // Assuming completionStartTime is a DateTime field model String @default("") model_id String? @default("") // the model id stored in proxy model db diff --git a/litellm/proxy/_types.py b/litellm/proxy/_types.py index 4053d9d077b..96c5265c0e6 100644 --- a/litellm/proxy/_types.py +++ b/litellm/proxy/_types.py @@ -3105,6 +3105,7 @@ class SpendLogsPayload(TypedDict): response: Optional[Union[str, list, dict]] proxy_server_request: Optional[str] session_id: Optional[str] + request_duration_ms: Optional[int] status: Literal["success", "failure"] diff --git a/litellm/proxy/schema.prisma b/litellm/proxy/schema.prisma index 8f746b1f9c0..8498b131df8 100644 --- a/litellm/proxy/schema.prisma +++ b/litellm/proxy/schema.prisma @@ -477,6 +477,7 @@ model LiteLLM_SpendLogs { completion_tokens Int @default(0) startTime DateTime // Assuming start_time is a DateTime field endTime DateTime // Assuming end_time is a DateTime field + request_duration_ms Int? completionStartTime DateTime? // Assuming completionStartTime is a DateTime field model String @default("") model_id String? @default("") // the model id stored in proxy model db diff --git a/litellm/proxy/spend_tracking/spend_management_endpoints.py b/litellm/proxy/spend_tracking/spend_management_endpoints.py index b4aeda62004..5b58fbe70a0 100644 --- a/litellm/proxy/spend_tracking/spend_management_endpoints.py +++ b/litellm/proxy/spend_tracking/spend_management_endpoints.py @@ -1726,7 +1726,7 @@ async def ui_view_spend_logs( # noqa: PLR0915 ) # Validate sort_by and sort_order - valid_sort_fields = {"spend", "total_tokens", "startTime", "endTime"} + valid_sort_fields = {"spend", "total_tokens", "startTime", "endTime", "request_duration_ms"} if sort_by not in valid_sort_fields: raise ProxyException( message=f"Invalid sort_by: {sort_by}. Must be one of: {', '.join(sorted(valid_sort_fields))}", @@ -1939,7 +1939,8 @@ async def ui_view_spend_logs( # noqa: PLR0915 custom_llm_provider, api_base, "user", metadata, cache_hit, cache_key, request_tags, team_id, organization_id, end_user, requester_ip_address, - session_id, status, mcp_namespaced_tool_name, agent_id + session_id, status, mcp_namespaced_tool_name, agent_id, + COALESCE(request_duration_ms, (EXTRACT(EPOCH FROM ("endTime" - "startTime")) * 1000)::INTEGER) AS request_duration_ms FROM "LiteLLM_SpendLogs" WHERE {" AND ".join(sql_conditions)} ORDER BY {_sql_col} {_sql_dir} diff --git a/litellm/proxy/spend_tracking/spend_tracking_utils.py b/litellm/proxy/spend_tracking/spend_tracking_utils.py index d517c76a08c..05dbb8ed71e 100644 --- a/litellm/proxy/spend_tracking/spend_tracking_utils.py +++ b/litellm/proxy/spend_tracking/spend_tracking_utils.py @@ -447,6 +447,7 @@ def get_logging_payload( # noqa: PLR0915 kwargs=kwargs, standard_logging_payload=standard_logging_payload, ), + request_duration_ms=_get_request_duration_ms(start_time, end_time), status=_get_status_for_spend_log( metadata=metadata, ), @@ -496,6 +497,16 @@ def _get_session_id_for_spend_log( return str(uuid.uuid4()) +def _get_request_duration_ms( + start_time: datetime, end_time: datetime +) -> Optional[int]: + """Compute request duration in milliseconds from start and end times.""" + try: + return int((end_time - start_time).total_seconds() * 1000) + except Exception: + return None + + def _ensure_datetime_utc(timestamp: datetime) -> datetime: """Helper to ensure datetime is in UTC""" timestamp = timestamp.astimezone(timezone.utc) diff --git a/schema.prisma b/schema.prisma index 4af7484148c..5fa805e444e 100644 --- a/schema.prisma +++ b/schema.prisma @@ -477,6 +477,7 @@ model LiteLLM_SpendLogs { completion_tokens Int @default(0) startTime DateTime // Assuming start_time is a DateTime field endTime DateTime // Assuming end_time is a DateTime field + request_duration_ms Int? completionStartTime DateTime? // Assuming completionStartTime is a DateTime field model String @default("") model_id String? @default("") // the model id stored in proxy model db diff --git a/tests/test_litellm/proxy/spend_tracking/test_spend_management_endpoints.py b/tests/test_litellm/proxy/spend_tracking/test_spend_management_endpoints.py index c3639e88f6e..e439dfd693c 100644 --- a/tests/test_litellm/proxy/spend_tracking/test_spend_management_endpoints.py +++ b/tests/test_litellm/proxy/spend_tracking/test_spend_management_endpoints.py @@ -274,6 +274,7 @@ ignored_keys = [ "endTime", "completionStartTime", "endTime", + "request_duration_ms", "organization_id", "metadata.model_map_information", "metadata.usage_object", @@ -606,6 +607,82 @@ async def test_ui_view_spend_logs_sort_validation_errors( app.dependency_overrides.pop(ps.user_api_key_auth, None) +@pytest.mark.asyncio +async def test_ui_view_spend_logs_sort_by_request_duration_ms(client, monkeypatch): + """Test that request_duration_ms is accepted as a valid sort_by field.""" + base_logs = [ + { + "request_id": "req_fast", + "api_key": "sk-test-key", + "user": "user1", + "spend": 0.10, + "total_tokens": 100, + "request_duration_ms": 100, + "startTime": "2025-01-01T00:00:00+00:00", + "endTime": "2025-01-01T00:00:00.100000+00:00", + "model": "gpt-4", + }, + { + "request_id": "req_slow", + "api_key": "sk-test-key", + "user": "user1", + "spend": 0.05, + "total_tokens": 50, + "request_duration_ms": 5000, + "startTime": "2025-01-01T00:00:01+00:00", + "endTime": "2025-01-01T00:00:06+00:00", + "model": "gpt-4", + }, + ] + + async def mock_count(*args, **kwargs): + return len(base_logs) + + async def mock_query_raw(sql_query, *params): + reverse = "DESC" in sql_query + sorted_logs = sorted( + base_logs, key=lambda x: x.get("request_duration_ms", 0), reverse=reverse + ) + page_size = params[-2] if len(params) >= 2 else 50 + skip = params[-1] if len(params) >= 1 else 0 + return sorted_logs[skip : skip + page_size] + + class MockPrismaClient: + def __init__(self): + self.db = MagicMock() + self.db.litellm_spendlogs = MagicMock() + self.db.litellm_spendlogs.count = AsyncMock(side_effect=mock_count) + self.db.query_raw = AsyncMock(side_effect=mock_query_raw) + + monkeypatch.setattr("litellm.proxy.proxy_server.prisma_client", MockPrismaClient()) + monkeypatch.setattr( + "litellm.proxy.spend_tracking.spend_management_endpoints._is_admin_view_safe", + lambda user_api_key_dict: True, + ) + app.dependency_overrides[ps.user_api_key_auth] = lambda: UserAPIKeyAuth( + user_role=LitellmUserRoles.PROXY_ADMIN, user_id="admin_user" + ) + + try: + response = client.get( + "/spend/logs/ui", + params={ + "start_date": "2024-12-25 00:00:00", + "end_date": "2025-01-02 23:59:59", + "sort_by": "request_duration_ms", + "sort_order": "asc", + }, + headers={"Authorization": "Bearer sk-test"}, + ) + + assert response.status_code == 200, response.text + data = response.json() + actual_ids = [log["request_id"] for log in data["data"]] + assert actual_ids == ["req_fast", "req_slow"] + finally: + app.dependency_overrides.pop(ps.user_api_key_auth, None) + + @pytest.mark.asyncio async def test_ui_view_spend_logs_with_team_id(client, monkeypatch): mock_spend_logs = [ diff --git a/tests/test_litellm/proxy/spend_tracking/test_spend_tracking_utils.py b/tests/test_litellm/proxy/spend_tracking/test_spend_tracking_utils.py index 47a327f01f6..06a544e7f4d 100644 --- a/tests/test_litellm/proxy/spend_tracking/test_spend_tracking_utils.py +++ b/tests/test_litellm/proxy/spend_tracking/test_spend_tracking_utils.py @@ -20,6 +20,7 @@ from litellm.constants import LITELLM_TRUNCATED_PAYLOAD_FIELD, REDACTED_BY_LITEL from litellm.litellm_core_utils.safe_json_dumps import safe_dumps from litellm.proxy.spend_tracking.spend_tracking_utils import ( _get_proxy_server_request_for_spend_logs_payload, + _get_request_duration_ms, _get_response_for_spend_logs_payload, _get_spend_logs_metadata, _get_vector_store_request_for_spend_logs_payload, @@ -1232,3 +1233,50 @@ def test_get_logging_payload_handles_missing_retry_info_gracefully(): metadata.get("max_retries") is None ), "max_retries should be None when not provided" + +def test_get_request_duration_ms_normal(): + """Test that request duration is correctly computed in milliseconds.""" + start = datetime.datetime(2025, 1, 1, 0, 0, 0, tzinfo=timezone.utc) + end = datetime.datetime(2025, 1, 1, 0, 0, 2, 500000, tzinfo=timezone.utc) # 2.5s later + result = _get_request_duration_ms(start, end) + assert result == 2500 + + +def test_get_request_duration_ms_sub_millisecond(): + """Test that sub-millisecond durations are truncated to int.""" + start = datetime.datetime(2025, 1, 1, 0, 0, 0, 0, tzinfo=timezone.utc) + end = datetime.datetime(2025, 1, 1, 0, 0, 0, 500, tzinfo=timezone.utc) # 0.5ms + result = _get_request_duration_ms(start, end) + assert result == 0 + + +def test_get_request_duration_ms_zero(): + """Test that identical start and end times produce 0.""" + t = datetime.datetime(2025, 1, 1, 0, 0, 0, tzinfo=timezone.utc) + result = _get_request_duration_ms(t, t) + assert result == 0 + + +def test_get_logging_payload_includes_request_duration_ms(): + """Test that get_logging_payload populates request_duration_ms.""" + start_time = datetime.datetime(2025, 1, 1, 0, 0, 0, tzinfo=timezone.utc) + end_time = datetime.datetime(2025, 1, 1, 0, 0, 3, tzinfo=timezone.utc) # 3s later + + kwargs = { + "model": "gpt-4", + "litellm_params": {"api_base": "https://api.openai.com"}, + "standard_logging_object": None, + } + response_obj = {"usage": {"prompt_tokens": 10, "completion_tokens": 5, "total_tokens": 15}} + + with patch("litellm.proxy.proxy_server.master_key", None), \ + patch("litellm.proxy.proxy_server.general_settings", {}): + payload = get_logging_payload( + kwargs=kwargs, + response_obj=response_obj, + start_time=start_time, + end_time=end_time, + ) + + assert payload["request_duration_ms"] == 3000 + From 115b9a2d29640dbf85db95e6317d0e54d024f1e7 Mon Sep 17 00:00:00 2001 From: Shin Date: Wed, 25 Feb 2026 18:09:17 +0000 Subject: [PATCH 06/15] fix(ui): remove duplicate antd import in ToolPolicies Duplicate import was causing UI build to fail: - Line 4: import { Select, Switch, Tooltip } from 'antd' - Line 5: import { Select, Tooltip } from 'antd' (duplicate) Removed the duplicate line 5. --- ui/litellm-dashboard/src/components/ToolPolicies.tsx | 1 - 1 file changed, 1 deletion(-) diff --git a/ui/litellm-dashboard/src/components/ToolPolicies.tsx b/ui/litellm-dashboard/src/components/ToolPolicies.tsx index 0e3f5434e7f..f69d9cf0c47 100644 --- a/ui/litellm-dashboard/src/components/ToolPolicies.tsx +++ b/ui/litellm-dashboard/src/components/ToolPolicies.tsx @@ -2,7 +2,6 @@ import React, { useCallback, useDeferredValue, useEffect, useState } from "react"; import { Select, Switch, Tooltip } from "antd"; -import { Select, Tooltip } from "antd"; import { Table, TableHead, From 6ae7e84a0b89ce42b35db2ee9739bb058f0c98ce Mon Sep 17 00:00:00 2001 From: yuneng-jiang Date: Wed, 25 Feb 2026 10:25:25 -0800 Subject: [PATCH 07/15] [Test] UI - Pricing Calculator: Add comprehensive unit tests Added unit tests for all pricing calculator components with 64 passing tests across 5 test files: - multi_export_utils.test.ts (16 tests for PDF/CSV export functions) - use_multi_cost_estimate.test.ts (15 tests for cost estimation hook) - multi_export_dropdown.test.tsx (8 tests for export dropdown component) - multi_cost_results.test.tsx (15 tests for results display and UI states) - index.test.tsx (10 tests for main calculator component) Also fixed missing page description for tool-policies page in page_metadata.ts. Co-Authored-By: Claude Haiku 4.5 --- ui/litellm-dashboard/package-lock.json | 93 ++++- .../pricing_calculator/index.test.tsx | 131 +++++++ .../multi_cost_results.test.tsx | 305 ++++++++++++++++ .../multi_export_dropdown.test.tsx | 146 ++++++++ .../multi_export_utils.test.ts | 274 ++++++++++++++ .../use_multi_cost_estimate.test.ts | 342 ++++++++++++++++++ .../src/components/page_metadata.ts | 1 + 7 files changed, 1276 insertions(+), 16 deletions(-) create mode 100644 ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/index.test.tsx create mode 100644 ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_cost_results.test.tsx create mode 100644 ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_export_dropdown.test.tsx create mode 100644 ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_export_utils.test.ts create mode 100644 ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/use_multi_cost_estimate.test.ts diff --git a/ui/litellm-dashboard/package-lock.json b/ui/litellm-dashboard/package-lock.json index 3787f451ad3..fc2aa1599d3 100644 --- a/ui/litellm-dashboard/package-lock.json +++ b/ui/litellm-dashboard/package-lock.json @@ -90,6 +90,7 @@ "version": "5.2.0", "resolved": "https://registry.npmjs.org/@alloc/quick-lru/-/quick-lru-5.2.0.tgz", "integrity": "sha512-UrcABB+4bUrFABwbluTIBErXwvbsU/V7TZWfmbgJfbkwiBuziS9gxdODUyuiecfdGQ85jglMW6juS3+z5TsKLw==", + "dev": true, "license": "MIT", "engines": { "node": ">=10" @@ -1771,6 +1772,7 @@ "version": "0.3.13", "resolved": "https://registry.npmjs.org/@jridgewell/gen-mapping/-/gen-mapping-0.3.13.tgz", "integrity": "sha512-2kkt/7niJ6MgEPxF0bYdQ6etZaA+fQvDcLKckhy1yIQOzaoKjBBjSj63/aLVjYE3qhRt5dvM+uUyfCg6UKCBbA==", + "dev": true, "license": "MIT", "dependencies": { "@jridgewell/sourcemap-codec": "^1.5.0", @@ -1781,6 +1783,7 @@ "version": "3.1.2", "resolved": "https://registry.npmjs.org/@jridgewell/resolve-uri/-/resolve-uri-3.1.2.tgz", "integrity": "sha512-bRISgCIjP20/tbWSPWMEi54QVPRZExkuD9lJL+UIxUKtwVJA8wW1Trb1jMs1RFXo1CBTNZ/5hpC9QvmKWdopKw==", + "dev": true, "license": "MIT", "engines": { "node": ">=6.0.0" @@ -1790,12 +1793,14 @@ "version": "1.5.5", "resolved": "https://registry.npmjs.org/@jridgewell/sourcemap-codec/-/sourcemap-codec-1.5.5.tgz", "integrity": "sha512-cYQ9310grqxueWbl+WuIUIaiUaDcj7WOq5fVhEljNVgRfOUhY9fy2zTvfoqWsnebh8Sl70VScFbICvJnLKB0Og==", + "dev": true, "license": "MIT" }, "node_modules/@jridgewell/trace-mapping": { "version": "0.3.31", "resolved": "https://registry.npmjs.org/@jridgewell/trace-mapping/-/trace-mapping-0.3.31.tgz", "integrity": "sha512-zzNR+SdQSDJzc8joaeP8QQoCQr8NuYx2dIIytl1QeBEZHJ9uW6hebsrYgbz8hJwUQao3TWCMtmfV8Nu1twOLAw==", + "dev": true, "license": "MIT", "dependencies": { "@jridgewell/resolve-uri": "^3.1.0", @@ -1973,6 +1978,7 @@ "version": "2.1.5", "resolved": "https://registry.npmjs.org/@nodelib/fs.scandir/-/fs.scandir-2.1.5.tgz", "integrity": "sha512-vq24Bq3ym5HEQm2NKCr3yXDwjc7vTsEThRDnkp2DK9p1uqLR+DHurm/NOTo0KG7HYHU7eppKZj3MyqYuMBf62g==", + "dev": true, "license": "MIT", "dependencies": { "@nodelib/fs.stat": "2.0.5", @@ -1986,6 +1992,7 @@ "version": "2.0.5", "resolved": "https://registry.npmjs.org/@nodelib/fs.stat/-/fs.stat-2.0.5.tgz", "integrity": "sha512-RkhPPp2zrqDAQA/2jNhnztcPAlv64XdhIp7a7454A5ovI7Bukxgt7MX7udwAu3zg1DcpPU0rz3VV1SeaqvY4+A==", + "dev": true, "license": "MIT", "engines": { "node": ">= 8" @@ -1995,6 +2002,7 @@ "version": "1.2.8", "resolved": "https://registry.npmjs.org/@nodelib/fs.walk/-/fs.walk-1.2.8.tgz", "integrity": "sha512-oGB+UxlgWcgQkgwo8GcEGwemoTFt3FIO9ababBmaGwXIoBKZ+GTy0pP185beGg7Llih/NSHSV2XAs1lnznocSg==", + "dev": true, "license": "MIT", "dependencies": { "@nodelib/fs.scandir": "2.1.5", @@ -2318,7 +2326,7 @@ "version": "1.58.1", "resolved": "https://registry.npmjs.org/@playwright/test/-/test-1.58.1.tgz", "integrity": "sha512-6LdVIUERWxQMmUSSQi0I53GgCBYgM2RpGngCPY7hSeju+VrKjq3lvs7HpJoPbDiY5QM5EYRtRX5fvrinnMAz3w==", - "devOptional": true, + "dev": true, "license": "Apache-2.0", "dependencies": { "playwright": "1.58.1" @@ -3423,12 +3431,14 @@ "version": "15.7.15", "resolved": "https://registry.npmjs.org/@types/prop-types/-/prop-types-15.7.15.tgz", "integrity": "sha512-F6bEyamV9jKGAFBEmlQnesRPGOQqS2+Uwi0Em15xenOxHaf2hv6L8YCVn3rPdPJOiJfPiCnLIRyvwVaqMY3MIw==", + "dev": true, "license": "MIT" }, "node_modules/@types/react": { "version": "18.2.48", "resolved": "https://registry.npmjs.org/@types/react/-/react-18.2.48.tgz", "integrity": "sha512-qboRCl6Ie70DQQG9hhNREz81jqC1cs9EVNcjQ1AU+jH6NFfSAhVVbrrY/+nSF+Bsk4AOwm9Qa61InvMCyV+H3w==", + "dev": true, "license": "MIT", "dependencies": { "@types/prop-types": "*", @@ -3470,6 +3480,7 @@ "version": "0.26.0", "resolved": "https://registry.npmjs.org/@types/scheduler/-/scheduler-0.26.0.tgz", "integrity": "sha512-WFHp9YUJQ6CKshqoC37iOlHnQSmxNc795UhB26CyBBttrN9svdIrUjl/NjnNmfcwtncN0h/0PPAFWv9ovP8mLA==", + "dev": true, "license": "MIT" }, "node_modules/@types/unist": { @@ -4330,12 +4341,14 @@ "version": "1.3.0", "resolved": "https://registry.npmjs.org/any-promise/-/any-promise-1.3.0.tgz", "integrity": "sha512-7UvmKalWRt1wgjL1RrGxoSJW/0QZFIegpeGvZG9kjp8vrRu55XTHbwnqq2GpXm9uLbcuhxm3IqX9OB4MZR1b2A==", + "dev": true, "license": "MIT" }, "node_modules/anymatch": { "version": "3.1.3", "resolved": "https://registry.npmjs.org/anymatch/-/anymatch-3.1.3.tgz", "integrity": "sha512-KMReFUr0B4t+D+OBkjR3KYqvocp2XaSzO55UcB6mgQMd3KbcE+mWTyvVV7D/zsdEbNnV6acZUutkiHQXvTr1Rw==", + "dev": true, "license": "ISC", "dependencies": { "normalize-path": "^3.0.0", @@ -4349,6 +4362,7 @@ "version": "2.3.1", "resolved": "https://registry.npmjs.org/picomatch/-/picomatch-2.3.1.tgz", "integrity": "sha512-JU3teHTNjmE2VCGFzuY8EXzCDVwEqB2a8fsIvwaStHhAWJEeVd1o1QD80CU6+ZdEXXSLbSsuLwJjkCBWqRQUVA==", + "dev": true, "license": "MIT", "engines": { "node": ">=8.6" @@ -4361,6 +4375,7 @@ "version": "5.0.2", "resolved": "https://registry.npmjs.org/arg/-/arg-5.0.2.tgz", "integrity": "sha512-PYjyFOLKQ9y57JvQ6QLo8dAgNqswh8M1RMJYdQduT6xbWSgK36P/Z/v+p888pM69jMMfS8Xd8F6I1kQ/I9HUGg==", + "dev": true, "license": "MIT" }, "node_modules/argparse": { @@ -4732,6 +4747,7 @@ "version": "2.3.0", "resolved": "https://registry.npmjs.org/binary-extensions/-/binary-extensions-2.3.0.tgz", "integrity": "sha512-Ceh+7ox5qe7LJuLHoY0feh3pHuUDHAcRUeyL2VYghZwfpkNIy/+8Ocg0a3UuSoYzavmylwuLWQOf3hl0jjMMIw==", + "dev": true, "license": "MIT", "engines": { "node": ">=8" @@ -4757,6 +4773,7 @@ "version": "3.0.3", "resolved": "https://registry.npmjs.org/braces/-/braces-3.0.3.tgz", "integrity": "sha512-yQbXgO/OSZVD2IsiLlro+7Hf6Q18EJrKSEsdoMzKePKXct3gvD8oLcOQdIzGupr5Fj+EDe8gO/lxc1BzfMpxvA==", + "dev": true, "license": "MIT", "dependencies": { "fill-range": "^7.1.1" @@ -4872,6 +4889,7 @@ "version": "2.0.1", "resolved": "https://registry.npmjs.org/camelcase-css/-/camelcase-css-2.0.1.tgz", "integrity": "sha512-QOSvevhslijgYwRx6Rv7zKdMF8lbRmx+uQGx2+vDc+KI/eBnsy9kit5aj23AgGu3pa4t9AgwbnXWqS+iOY+2aA==", + "dev": true, "license": "MIT", "engines": { "node": ">= 6" @@ -4995,6 +5013,7 @@ "version": "3.6.0", "resolved": "https://registry.npmjs.org/chokidar/-/chokidar-3.6.0.tgz", "integrity": "sha512-7VT13fmjotKpGipCW9JEQAusEPE+Ei8nl6/g4FBAmIm0GOOLMua9NDDo/DWp0ZAxCr3cPq5ZpBqmPAQgDda2Pw==", + "dev": true, "license": "MIT", "dependencies": { "anymatch": "~3.1.2", @@ -5019,6 +5038,7 @@ "version": "5.1.2", "resolved": "https://registry.npmjs.org/glob-parent/-/glob-parent-5.1.2.tgz", "integrity": "sha512-AOIgSQCepiJYwP3ARnGx+5VnTu2HBYdzbGP45eLw1vr3zB3vZLeyed1sC9hnbcOc9/SrMyM5RPQrkGz4aS9Zow==", + "dev": true, "license": "ISC", "dependencies": { "is-glob": "^4.0.1" @@ -5094,6 +5114,7 @@ "version": "4.1.1", "resolved": "https://registry.npmjs.org/commander/-/commander-4.1.1.tgz", "integrity": "sha512-NOKm8xhkzAjzFx8B2v5OAHT+u5pRQc2UCa2Vq9jYL/31o2wi9mxBA7LIFs3sV5VSC49z6pEhfbMULvShKj26WA==", + "dev": true, "license": "MIT", "engines": { "node": ">= 6" @@ -5154,6 +5175,7 @@ "version": "3.0.0", "resolved": "https://registry.npmjs.org/cssesc/-/cssesc-3.0.0.tgz", "integrity": "sha512-/Tb/JcjK111nNScGob5MNtsntNM1aCNUDipB/TkwZFhyDrrE47SOx/18wF2bbjgc3ZzCSKW1T5nt5EbFoAz/Vg==", + "dev": true, "license": "MIT", "bin": { "cssesc": "bin/cssesc" @@ -5567,12 +5589,14 @@ "version": "1.2.2", "resolved": "https://registry.npmjs.org/didyoumean/-/didyoumean-1.2.2.tgz", "integrity": "sha512-gxtyfqMg7GKyhQmb056K7M3xszy/myH8w+B4RT+QXBQsvAOdc3XymqDDPHx1BgPgsdAA5SIifona89YtRATDzw==", + "dev": true, "license": "Apache-2.0" }, "node_modules/dlv": { "version": "1.1.3", "resolved": "https://registry.npmjs.org/dlv/-/dlv-1.1.3.tgz", "integrity": "sha512-+HlytyjlPKnIG8XuRG8WvmBP8xs8P71y+SKKS6ZXWoEgLuePxtDoUEiH7WkdePWrQ5JBpE6aoVqfZfJUQkjXwA==", + "dev": true, "license": "MIT" }, "node_modules/doctrine": { @@ -6486,6 +6510,7 @@ "version": "1.20.1", "resolved": "https://registry.npmjs.org/fastq/-/fastq-1.20.1.tgz", "integrity": "sha512-GGToxJ/w1x32s/D2EKND7kTil4n8OVk/9mycTc4VDza13lOvpUZTGX3mFSCtV9ksdGBVzvsyAVLM6mHFThxXxw==", + "dev": true, "license": "ISC", "dependencies": { "reusify": "^1.0.4" @@ -6518,6 +6543,7 @@ "version": "6.5.0", "resolved": "https://registry.npmjs.org/fdir/-/fdir-6.5.0.tgz", "integrity": "sha512-tIbYtZbucOs0BRGqPJkshJUYdL+SDH7dVM8gjy+ERp3WAUjLEFJE+02kanyHtwjWOnwrKYBiwAmM0p4kLJAnXg==", + "dev": true, "license": "MIT", "engines": { "node": ">=12.0.0" @@ -6555,6 +6581,7 @@ "version": "7.1.1", "resolved": "https://registry.npmjs.org/fill-range/-/fill-range-7.1.1.tgz", "integrity": "sha512-YsGpe3WHLK8ZYi4tWDg2Jy3ebRz2rXowDxnld4bkQB00cc/1Zw9AWnC0i9ztDJitivtQvaI9KaLyKrc+hBW0yg==", + "dev": true, "license": "MIT", "dependencies": { "to-regex-range": "^5.0.1" @@ -6715,6 +6742,7 @@ "version": "2.3.2", "resolved": "https://registry.npmjs.org/fsevents/-/fsevents-2.3.2.tgz", "integrity": "sha512-xiqMQR4xAeHTuB9uWm+fFRcIOgKBMiOBP+eXiyT7jsgVCq1bkVygt00oASowB7EdtpOHaaPgKt812P9ab+DDKA==", + "dev": true, "hasInstallScript": true, "license": "MIT", "optional": true, @@ -6865,6 +6893,7 @@ "version": "6.0.2", "resolved": "https://registry.npmjs.org/glob-parent/-/glob-parent-6.0.2.tgz", "integrity": "sha512-XxwI8EOhVQgWp6iDL+3b0r86f4d6AX6zSU55HfB4ydCEuXLXc5FcYeOu+nnGftS4TEju/11rt4KJPTMgbfmv4A==", + "dev": true, "license": "ISC", "dependencies": { "is-glob": "^4.0.3" @@ -7362,6 +7391,7 @@ "version": "2.1.0", "resolved": "https://registry.npmjs.org/is-binary-path/-/is-binary-path-2.1.0.tgz", "integrity": "sha512-ZMERYes6pDydyuGidse7OsHxtbI7WVeUEozgR/g7rd0xUimYNlvZRE/K2MgZTjWy725IfelLeVcEM97mmtRGXw==", + "dev": true, "license": "MIT", "dependencies": { "binary-extensions": "^2.0.0" @@ -7414,6 +7444,7 @@ "version": "2.16.1", "resolved": "https://registry.npmjs.org/is-core-module/-/is-core-module-2.16.1.tgz", "integrity": "sha512-UfoeMA6fIJ8wTYFEUjelnaGI67v6+N7qXJEvQuIGa99l4xsCruSYOVSQ0uPANn4dAzm8lkYPaKLrrijLq7x23w==", + "dev": true, "license": "MIT", "dependencies": { "hasown": "^2.0.2" @@ -7474,6 +7505,7 @@ "version": "2.1.1", "resolved": "https://registry.npmjs.org/is-extglob/-/is-extglob-2.1.1.tgz", "integrity": "sha512-SbKbANkN603Vi4jEZv49LeVJMn4yGwsbzZworEoyEiutsN3nJYdbO36zfhGJ6QEDpOZIFkDtnq5JRxmvl3jsoQ==", + "dev": true, "license": "MIT", "engines": { "node": ">=0.10.0" @@ -7519,6 +7551,7 @@ "version": "4.0.3", "resolved": "https://registry.npmjs.org/is-glob/-/is-glob-4.0.3.tgz", "integrity": "sha512-xelSayHH36ZgE7ZWhli7pW34hNbNl8Ojv5KVmkJD4hBdD3th8Tfk9vYasLM+mXWOZhFkgZfxhLSnrwRr4elSSg==", + "dev": true, "license": "MIT", "dependencies": { "is-extglob": "^2.1.1" @@ -7567,6 +7600,7 @@ "version": "7.0.0", "resolved": "https://registry.npmjs.org/is-number/-/is-number-7.0.0.tgz", "integrity": "sha512-41Cifkg6e8TylSpdtTpeLVMqvSBEVzTttHvERD741+pnZ8ANv0004MRL43QKPDlK9cGvNp6NZWZUBlbGXYxxng==", + "dev": true, "license": "MIT", "engines": { "node": ">=0.12.0" @@ -7843,6 +7877,7 @@ "version": "1.21.7", "resolved": "https://registry.npmjs.org/jiti/-/jiti-1.21.7.tgz", "integrity": "sha512-/imKNG4EbWNrVjoNC/1H5/9GFy+tqjGBHCaSsN+P2RnPqjsLmv6UD3Ej+Kj8nBWaRAwyk7kK5ZUc+OEatnTR3A==", + "dev": true, "license": "MIT", "bin": { "jiti": "bin/jiti.js" @@ -8128,6 +8163,7 @@ "version": "3.1.3", "resolved": "https://registry.npmjs.org/lilconfig/-/lilconfig-3.1.3.tgz", "integrity": "sha512-/vlFKAoH5Cgt3Ie+JLhRbwOsCQePABiU3tJ1egGvyQ+33R/vcwM2Zl2QR/LzjsBeItPt3oSVXapn+m4nQDvpzw==", + "dev": true, "license": "MIT", "engines": { "node": ">=14" @@ -8140,6 +8176,7 @@ "version": "1.2.4", "resolved": "https://registry.npmjs.org/lines-and-columns/-/lines-and-columns-1.2.4.tgz", "integrity": "sha512-7ylylesZQ/PV29jhEDl3Ufjo6ZX7gCqJr5F7PKrqc93v7fzSymt1BpwEU8nAUXs8qzzvqhbjhK5QZg6Mt/HkBg==", + "dev": true, "license": "MIT" }, "node_modules/locate-path": { @@ -8454,6 +8491,7 @@ "version": "1.4.1", "resolved": "https://registry.npmjs.org/merge2/-/merge2-1.4.1.tgz", "integrity": "sha512-8q7VEgMJW4J8tcfVPy8g09NcQwZdbwFEqhe/WZkoIzjn/3TGDwtOCYtXGxA3O8tPzpczCCDgv+P2P5y00ZJOOg==", + "dev": true, "license": "MIT", "engines": { "node": ">= 8" @@ -8905,6 +8943,7 @@ "version": "4.0.8", "resolved": "https://registry.npmjs.org/micromatch/-/micromatch-4.0.8.tgz", "integrity": "sha512-PXwfBhYu0hBCPw8Dn0E+WDYb7af3dSLVWKi3HGv84IdF4TyFoC0ysxFd0Goxw7nSv4T/PzEJQxsYsEiFCKo2BA==", + "dev": true, "license": "MIT", "dependencies": { "braces": "^3.0.3", @@ -8918,6 +8957,7 @@ "version": "2.3.1", "resolved": "https://registry.npmjs.org/picomatch/-/picomatch-2.3.1.tgz", "integrity": "sha512-JU3teHTNjmE2VCGFzuY8EXzCDVwEqB2a8fsIvwaStHhAWJEeVd1o1QD80CU6+ZdEXXSLbSsuLwJjkCBWqRQUVA==", + "dev": true, "license": "MIT", "engines": { "node": ">=8.6" @@ -9032,6 +9072,7 @@ "version": "2.7.0", "resolved": "https://registry.npmjs.org/mz/-/mz-2.7.0.tgz", "integrity": "sha512-z81GNO7nnYMEhrGh9LeymoE4+Yr0Wn5McHIZMK5cfQCl+NDX08sCZgUc9/6MHni9IWuFLm1Z3HTCXu2z9fN62Q==", + "dev": true, "license": "MIT", "dependencies": { "any-promise": "^1.0.0", @@ -9243,6 +9284,7 @@ "version": "3.0.0", "resolved": "https://registry.npmjs.org/normalize-path/-/normalize-path-3.0.0.tgz", "integrity": "sha512-6eZs5Ls3WtCisHWp9S2GUy8dqkpGi4BVSz3GaqiE6ezub0512ESztXUwUB6C6IKbQkY2Pnb/mD4WYojCRwcwLA==", + "dev": true, "license": "MIT", "engines": { "node": ">=0.10.0" @@ -9261,6 +9303,7 @@ "version": "3.0.0", "resolved": "https://registry.npmjs.org/object-hash/-/object-hash-3.0.0.tgz", "integrity": "sha512-RSn9F68PjH9HqtltsSnqYC1XXoWe9Bju5+213R98cNGttag9q9yAOTzdbsqvIa7aNm5WffBZFpWYr2aWrklWAw==", + "dev": true, "license": "MIT", "engines": { "node": ">= 6" @@ -9605,6 +9648,7 @@ "version": "1.0.7", "resolved": "https://registry.npmjs.org/path-parse/-/path-parse-1.0.7.tgz", "integrity": "sha512-LDJzPVEEEPR+y48z93A0Ed0yXb8pAByGWo/k5YYdYgpY2/2EsOsksJrq7lOHxryrVOn1ejG6oAp8ahvOIQD8sw==", + "dev": true, "license": "MIT" }, "node_modules/path-scurry": { @@ -9651,6 +9695,7 @@ "version": "4.0.3", "resolved": "https://registry.npmjs.org/picomatch/-/picomatch-4.0.3.tgz", "integrity": "sha512-5gTmgEY/sqK6gFXLIsQNH19lWb4ebPDLA4SdLP7dsWkIXHWlG66oPuVvXSGFPppYZz8ZDZq0dYYrbHfBCVUb1Q==", + "dev": true, "license": "MIT", "engines": { "node": ">=12" @@ -9663,6 +9708,7 @@ "version": "2.3.0", "resolved": "https://registry.npmjs.org/pify/-/pify-2.3.0.tgz", "integrity": "sha512-udgsAY+fTnvv7kI7aaxbqwWNb0AHiB0qBO89PZKPkoTmGOgdbrHDKD+0B2X4uTfJ/FT1R09r9gTsjUjNJotuog==", + "dev": true, "license": "MIT", "engines": { "node": ">=0.10.0" @@ -9672,6 +9718,7 @@ "version": "4.0.7", "resolved": "https://registry.npmjs.org/pirates/-/pirates-4.0.7.tgz", "integrity": "sha512-TfySrs/5nm8fQJDcBDuUng3VOUKsd7S+zqvbOTiGXHfxX4wK31ard+hoNuvkicM/2YFzlpDgABOevKSsB4G/FA==", + "dev": true, "license": "MIT", "engines": { "node": ">= 6" @@ -9681,7 +9728,7 @@ "version": "1.58.1", "resolved": "https://registry.npmjs.org/playwright/-/playwright-1.58.1.tgz", "integrity": "sha512-+2uTZHxSCcxjvGc5C891LrS1/NlxglGxzrC4seZiVjcYVQfUa87wBL6rTDqzGjuoWNjnBzRqKmF6zRYGMvQUaQ==", - "devOptional": true, + "dev": true, "license": "Apache-2.0", "dependencies": { "playwright-core": "1.58.1" @@ -9700,7 +9747,7 @@ "version": "1.58.1", "resolved": "https://registry.npmjs.org/playwright-core/-/playwright-core-1.58.1.tgz", "integrity": "sha512-bcWzOaTxcW+VOOGBCQgnaKToLJ65d6AqfLVKEWvexyS3AS6rbXl+xdpYRMGSRBClPvyj44njOWoxjNdL/H9UNg==", - "devOptional": true, + "dev": true, "license": "Apache-2.0", "bin": { "playwright-core": "cli.js" @@ -9723,6 +9770,7 @@ "version": "8.5.6", "resolved": "https://registry.npmjs.org/postcss/-/postcss-8.5.6.tgz", "integrity": "sha512-3Ybi1tAuwAP9s0r1UQ2J4n5Y0G05bJkpUIO0/bI9MhwmD70S5aTWbXGBwxHrelT+XM1k6dM0pk+SwNkpTRN7Pg==", + "dev": true, "funding": [ { "type": "opencollective", @@ -9751,6 +9799,7 @@ "version": "15.1.0", "resolved": "https://registry.npmjs.org/postcss-import/-/postcss-import-15.1.0.tgz", "integrity": "sha512-hpr+J05B2FVYUAXHeK1YyI267J/dDDhMU6B6civm8hSY1jYJnBXxzKDKDswzJmtLHryrjhnDjqqp/49t8FALew==", + "dev": true, "license": "MIT", "dependencies": { "postcss-value-parser": "^4.0.0", @@ -9768,6 +9817,7 @@ "version": "4.1.0", "resolved": "https://registry.npmjs.org/postcss-js/-/postcss-js-4.1.0.tgz", "integrity": "sha512-oIAOTqgIo7q2EOwbhb8UalYePMvYoIeRY2YKntdpFQXNosSu3vLrniGgmH9OKs/qAkfoj5oB3le/7mINW1LCfw==", + "dev": true, "funding": [ { "type": "opencollective", @@ -9793,6 +9843,7 @@ "version": "6.0.1", "resolved": "https://registry.npmjs.org/postcss-load-config/-/postcss-load-config-6.0.1.tgz", "integrity": "sha512-oPtTM4oerL+UXmx+93ytZVN82RrlY/wPUV8IeDxFrzIjXOLF1pN+EmKPLbubvKHT2HC20xXsCAH2Z+CKV6Oz/g==", + "dev": true, "funding": [ { "type": "opencollective", @@ -9835,6 +9886,7 @@ "version": "6.2.0", "resolved": "https://registry.npmjs.org/postcss-nested/-/postcss-nested-6.2.0.tgz", "integrity": "sha512-HQbt28KulC5AJzG+cZtj9kvKB93CFCdLvog1WFLf1D+xmMvPGlBstkpTEZfK5+AN9hfJocyBFCNiqyS48bpgzQ==", + "dev": true, "funding": [ { "type": "opencollective", @@ -9860,6 +9912,7 @@ "version": "6.1.2", "resolved": "https://registry.npmjs.org/postcss-selector-parser/-/postcss-selector-parser-6.1.2.tgz", "integrity": "sha512-Q8qQfPiZ+THO/3ZrOrO0cJJKfpYCagtMUkXbnEfmgUjwXg6z/WBeOyS9APBBPCTSiDV+s4SwQGu8yFsiMRIudg==", + "dev": true, "license": "MIT", "dependencies": { "cssesc": "^3.0.0", @@ -9873,6 +9926,7 @@ "version": "4.2.0", "resolved": "https://registry.npmjs.org/postcss-value-parser/-/postcss-value-parser-4.2.0.tgz", "integrity": "sha512-1NNCs6uurfkVbeXG4S8JFT9t19m45ICnif8zWLd5oPSZ50QnwMfK+H3jv408d4jw/7Bttv5axS5IiHoLaVNHeQ==", + "dev": true, "license": "MIT" }, "node_modules/prelude-ls": { @@ -9986,6 +10040,7 @@ "version": "1.2.3", "resolved": "https://registry.npmjs.org/queue-microtask/-/queue-microtask-1.2.3.tgz", "integrity": "sha512-NuaNSa6flKT5JaSYQzJok04JzTL1CA6aGhv5rfLW3PgqA+M2ChpZQnAC8h8i4ZFkBS8X5RqkDBHA7r4hej3K9A==", + "dev": true, "funding": [ { "type": "github", @@ -10774,6 +10829,7 @@ "version": "1.0.0", "resolved": "https://registry.npmjs.org/read-cache/-/read-cache-1.0.0.tgz", "integrity": "sha512-Owdv/Ft7IjOgm/i0xvNDZ1LrRANRfew4b2prF3OWMQLxLfu3bS8FVhCsrSCMK4lR56Y9ya+AThoTpDCTxCmpRA==", + "dev": true, "license": "MIT", "dependencies": { "pify": "^2.3.0" @@ -10783,6 +10839,7 @@ "version": "3.6.0", "resolved": "https://registry.npmjs.org/readdirp/-/readdirp-3.6.0.tgz", "integrity": "sha512-hOS089on8RduqdbhvQ5Z37A0ESjsqz6qnRcffsMU3495FuTdqSm+7bhJ29JvIOsBDEEnan5DPu9t3To9VRlMzA==", + "dev": true, "license": "MIT", "dependencies": { "picomatch": "^2.2.1" @@ -10795,6 +10852,7 @@ "version": "2.3.1", "resolved": "https://registry.npmjs.org/picomatch/-/picomatch-2.3.1.tgz", "integrity": "sha512-JU3teHTNjmE2VCGFzuY8EXzCDVwEqB2a8fsIvwaStHhAWJEeVd1o1QD80CU6+ZdEXXSLbSsuLwJjkCBWqRQUVA==", + "dev": true, "license": "MIT", "engines": { "node": ">=8.6" @@ -11059,6 +11117,7 @@ "version": "1.22.11", "resolved": "https://registry.npmjs.org/resolve/-/resolve-1.22.11.tgz", "integrity": "sha512-RfqAvLnMl313r7c9oclB1HhUEAezcpLjz95wFH4LVuhk9JF/r22qmVP9AMmOU4vMX7Q8pN8jwNg/CSpdFnMjTQ==", + "dev": true, "license": "MIT", "dependencies": { "is-core-module": "^2.16.1", @@ -11099,6 +11158,7 @@ "version": "1.1.0", "resolved": "https://registry.npmjs.org/reusify/-/reusify-1.1.0.tgz", "integrity": "sha512-g6QUff04oZpHs0eG5p83rFLhHeV00ug/Yf9nZM6fLeUrPguBTkTQOdpAWWspMh55TZfVQDPaN3NQJfbVRAxdIw==", + "dev": true, "license": "MIT", "engines": { "iojs": ">=1.0.0", @@ -11154,6 +11214,7 @@ "version": "1.2.0", "resolved": "https://registry.npmjs.org/run-parallel/-/run-parallel-1.2.0.tgz", "integrity": "sha512-5l4VyZR86LZ/lDxZTR6jqL8AFE2S0IFLMP26AbjsLVADxHdhB/c0GUsH+y39UfCi3dzz8OlQuPmnaJOMoDHQBA==", + "dev": true, "funding": [ { "type": "github", @@ -11794,6 +11855,7 @@ "version": "3.35.1", "resolved": "https://registry.npmjs.org/sucrase/-/sucrase-3.35.1.tgz", "integrity": "sha512-DhuTmvZWux4H1UOnWMB3sk0sbaCVOoQZjv8u1rDoTV0HTdGem9hkAZtl4JZy8P2z4Bg0nT+YMeOFyVr4zcG5Tw==", + "dev": true, "license": "MIT", "dependencies": { "@jridgewell/gen-mapping": "^0.3.2", @@ -11829,6 +11891,7 @@ "version": "1.0.0", "resolved": "https://registry.npmjs.org/supports-preserve-symlinks-flag/-/supports-preserve-symlinks-flag-1.0.0.tgz", "integrity": "sha512-ot0WnXS9fgdkgIcePe6RHNk1WA8+muPa6cSjeR3V8K27q9BB1rTE3R1p7Hv0z1ZyAc8s6Vvv8DIyWf681MAt0w==", + "dev": true, "license": "MIT", "engines": { "node": ">= 0.4" @@ -11864,6 +11927,7 @@ "version": "3.4.19", "resolved": "https://registry.npmjs.org/tailwindcss/-/tailwindcss-3.4.19.tgz", "integrity": "sha512-3ofp+LL8E+pK/JuPLPggVAIaEuhvIz4qNcf3nA1Xn2o/7fb7s/TYpHhwGDv1ZU3PkBluUVaF8PyCHcm48cKLWQ==", + "dev": true, "license": "MIT", "dependencies": { "@alloc/quick-lru": "^5.2.0", @@ -11901,6 +11965,7 @@ "version": "3.3.3", "resolved": "https://registry.npmjs.org/fast-glob/-/fast-glob-3.3.3.tgz", "integrity": "sha512-7MptL8U0cqcFdzIzwOTHoilX9x5BrNqye7Z/LuC7kCMRio1EMSyqRK3BEAUD7sXRq4iT4AzTVuZdhgQ2TCvYLg==", + "dev": true, "license": "MIT", "dependencies": { "@nodelib/fs.stat": "^2.0.2", @@ -11917,6 +11982,7 @@ "version": "5.1.2", "resolved": "https://registry.npmjs.org/glob-parent/-/glob-parent-5.1.2.tgz", "integrity": "sha512-AOIgSQCepiJYwP3ARnGx+5VnTu2HBYdzbGP45eLw1vr3zB3vZLeyed1sC9hnbcOc9/SrMyM5RPQrkGz4aS9Zow==", + "dev": true, "license": "ISC", "dependencies": { "is-glob": "^4.0.1" @@ -11944,6 +12010,7 @@ "version": "3.3.1", "resolved": "https://registry.npmjs.org/thenify/-/thenify-3.3.1.tgz", "integrity": "sha512-RVZSIV5IG10Hk3enotrhvz0T9em6cyHBLkH/YAZuKqd8hRkKhSfCGIcP2KUY0EPxndzANBmNllzWPwak+bheSw==", + "dev": true, "license": "MIT", "dependencies": { "any-promise": "^1.0.0" @@ -11953,6 +12020,7 @@ "version": "1.6.0", "resolved": "https://registry.npmjs.org/thenify-all/-/thenify-all-1.6.0.tgz", "integrity": "sha512-RNxQH/qI8/t3thXJDwcstUO4zeqo64+Uy/+sNVRBx4Xn2OX+OZ9oP+iJnNFqplFra2ZUVeKCSa2oVWi3T4uVmA==", + "dev": true, "license": "MIT", "dependencies": { "thenify": ">= 3.1.0 < 4" @@ -11994,6 +12062,7 @@ "version": "0.2.15", "resolved": "https://registry.npmjs.org/tinyglobby/-/tinyglobby-0.2.15.tgz", "integrity": "sha512-j2Zq4NyQYG5XMST4cbs02Ak8iJUdxRM0XI5QyxXuZOzKOINmWurp3smXu3y5wDcJrptwpSjgXHzIQxR0omXljQ==", + "dev": true, "license": "MIT", "dependencies": { "fdir": "^6.5.0", @@ -12060,6 +12129,7 @@ "version": "5.0.1", "resolved": "https://registry.npmjs.org/to-regex-range/-/to-regex-range-5.0.1.tgz", "integrity": "sha512-65P7iz6X5yEr1cwcgvQxbbIw7Uk3gOy5dIdtZ4rDveLqhrdJP+Li/Hx6tyK0NEb+2GCyneCMJiGqrADCSNk8sQ==", + "dev": true, "license": "MIT", "dependencies": { "is-number": "^7.0.0" @@ -12147,6 +12217,7 @@ "version": "0.1.13", "resolved": "https://registry.npmjs.org/ts-interface-checker/-/ts-interface-checker-0.1.13.tgz", "integrity": "sha512-Y/arvbn+rrz3JCKl9C4kVNfTfSm2/mEp5FSz5EsZSANGPSlQrpRI5M4PKF+mJnE52jOO90PnPSc3Ur3bTQw0gA==", + "dev": true, "license": "Apache-2.0" }, "node_modules/tsconfig-paths": { @@ -12263,7 +12334,7 @@ "version": "5.3.3", "resolved": "https://registry.npmjs.org/typescript/-/typescript-5.3.3.tgz", "integrity": "sha512-pXWcraxM0uxAS+tN0AG/BF2TyqmHO014Z070UsJ+pFvYuRSq8KH8DmWpnbXe0pEPDHXZV3FcAbJkijJ5oNEnWw==", - "devOptional": true, + "dev": true, "license": "Apache-2.0", "bin": { "tsc": "bin/tsc", @@ -12465,6 +12536,7 @@ "version": "1.0.2", "resolved": "https://registry.npmjs.org/util-deprecate/-/util-deprecate-1.0.2.tgz", "integrity": "sha512-EPD5q1uXyFxJpCrLnCc1nHnq3gOa6DZBocAIiI2TaSCA7VCJ1UJDMagCzIkXNsUYfD1daK//LTEQ8xiIbrHtcw==", + "dev": true, "license": "MIT" }, "node_modules/uuid": { @@ -12918,7 +12990,7 @@ "version": "8.19.0", "resolved": "https://registry.npmjs.org/ws/-/ws-8.19.0.tgz", "integrity": "sha512-blAT2mjOEIi0ZzruJfIhb3nps74PRWTCz1IjglWEEpQl5XS/UNama6u2/rjFkDDouqr4L67ry+1aGIALViWjDg==", - "devOptional": true, + "dev": true, "license": "MIT", "engines": { "node": ">=10.0.0" @@ -12975,17 +13047,6 @@ "url": "https://github.com/sponsors/sindresorhus" } }, - "node_modules/zod": { - "version": "3.25.76", - "resolved": "https://registry.npmjs.org/zod/-/zod-3.25.76.tgz", - "integrity": "sha512-gzUt/qt81nXsFGKIFcC3YnfEAx5NkunCfnDlvuBSSFS02bcXu4Lmea0AFIUwbLWxWPx3d9p8S5QoaujKcNQxcQ==", - "license": "MIT", - "optional": true, - "peer": true, - "funding": { - "url": "https://github.com/sponsors/colinhacks" - } - }, "node_modules/zwitch": { "version": "2.0.4", "resolved": "https://registry.npmjs.org/zwitch/-/zwitch-2.0.4.tgz", diff --git a/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/index.test.tsx b/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/index.test.tsx new file mode 100644 index 00000000000..b5cf2d8b346 --- /dev/null +++ b/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/index.test.tsx @@ -0,0 +1,131 @@ +import React from "react"; +import { describe, it, expect, vi, beforeEach } from "vitest"; +import { screen, within } from "@testing-library/react"; +import userEvent from "@testing-library/user-event"; +import { renderWithProviders } from "../../../../tests/test-utils"; +import PricingCalculator from "./index"; +import type { ModelEntry } from "./types"; +import type { MultiModelResult } from "./types"; + +vi.mock("./use_multi_cost_estimate", () => ({ + useMultiCostEstimate: vi.fn(() => ({ + debouncedFetchForEntry: vi.fn(), + removeEntry: vi.fn(), + getMultiModelResult: vi.fn((entries: ModelEntry[]): MultiModelResult => ({ + entries: entries.map((e) => ({ entry: e, result: null, loading: false, error: null })), + totals: { + cost_per_request: 0, + daily_cost: null, + monthly_cost: null, + margin_per_request: 0, + daily_margin: null, + monthly_margin: null, + }, + })), + })), +})); + +vi.mock("./multi_export_utils", () => ({ + exportMultiToPDF: vi.fn(), + exportMultiToCSV: vi.fn(), +})); + +vi.mock("@/utils/dataUtils", () => ({ + formatNumberWithCommas: vi.fn((v: number, d: number = 0) => + Number.isFinite(v) ? v.toFixed(d) : "-" + ), +})); + +const DEFAULT_PROPS = { + accessToken: "test-token", + models: ["gpt-4", "gpt-3.5-turbo", "claude-3-sonnet"], +}; + +describe("PricingCalculator", () => { + beforeEach(() => { + vi.clearAllMocks(); + }); + + it("should render the calculator with an initial model row", () => { + renderWithProviders(); + expect(screen.getByRole("table")).toBeInTheDocument(); + }); + + it("should render the time period toggle with Per Day and Per Month options", () => { + renderWithProviders(); + expect(screen.getByText("Per Day")).toBeInTheDocument(); + expect(screen.getByText("Per Month")).toBeInTheDocument(); + }); + + it("should render an Add Another Model button", () => { + renderWithProviders(); + expect(screen.getByRole("button", { name: /add another model/i })).toBeInTheDocument(); + }); + + it("should show the Requests/Month column header by default", () => { + renderWithProviders(); + expect(screen.getByText("Requests/Month")).toBeInTheDocument(); + }); + + it("should add a new row when Add Another Model is clicked", async () => { + const user = userEvent.setup(); + renderWithProviders(); + + const table = screen.getByRole("table"); + const initialRows = within(table).getAllByRole("row"); + + await user.click(screen.getByRole("button", { name: /add another model/i })); + + const updatedRows = within(table).getAllByRole("row"); + // One new data row added (header row + data rows) + expect(updatedRows.length).toBeGreaterThan(initialRows.length); + }); + + it("should have the delete button disabled when there is only one row", () => { + renderWithProviders(); + const allButtons = screen.getAllByRole("button"); + const disabledButtons = allButtons.filter((btn) => btn.hasAttribute("disabled")); + expect(disabledButtons.length).toBeGreaterThan(0); + }); + + it("should have no disabled buttons after adding a second row", async () => { + const user = userEvent.setup(); + renderWithProviders(); + + await user.click(screen.getByRole("button", { name: /add another model/i })); + + // With two rows, no delete buttons should be disabled + const allButtons = screen.getAllByRole("button"); + const disabledButtons = allButtons.filter((btn) => btn.hasAttribute("disabled")); + expect(disabledButtons.length).toBe(0); + }); + + describe("time period toggle", () => { + it("should switch the column header to Requests/Day when Per Day is selected", async () => { + const user = userEvent.setup(); + renderWithProviders(); + + await user.click(screen.getByText("Per Day")); + + expect(screen.getByText("Requests/Day")).toBeInTheDocument(); + }); + + it("should switch the column header back to Requests/Month when Per Month is selected", async () => { + const user = userEvent.setup(); + renderWithProviders(); + + await user.click(screen.getByText("Per Day")); + expect(screen.getByText("Requests/Day")).toBeInTheDocument(); + + await user.click(screen.getByText("Per Month")); + expect(screen.getByText("Requests/Month")).toBeInTheDocument(); + }); + }); + + it("should render column headers for Model, Input Tokens, and Output Tokens", () => { + renderWithProviders(); + expect(screen.getByText("Model")).toBeInTheDocument(); + expect(screen.getByText("Input Tokens")).toBeInTheDocument(); + expect(screen.getByText("Output Tokens")).toBeInTheDocument(); + }); +}); diff --git a/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_cost_results.test.tsx b/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_cost_results.test.tsx new file mode 100644 index 00000000000..6f6522f395e --- /dev/null +++ b/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_cost_results.test.tsx @@ -0,0 +1,305 @@ +import React from "react"; +import { describe, it, expect, vi, beforeEach } from "vitest"; +import { screen, within } from "@testing-library/react"; +import userEvent from "@testing-library/user-event"; +import { renderWithProviders } from "../../../../tests/test-utils"; +import MultiCostResults from "./multi_cost_results"; +import type { MultiModelResult } from "./types"; +import type { CostEstimateResponse } from "../types"; + +vi.mock("./multi_export_utils", () => ({ + exportMultiToPDF: vi.fn(), + exportMultiToCSV: vi.fn(), +})); + +vi.mock("@/utils/dataUtils", () => ({ + formatNumberWithCommas: vi.fn((v: number, d: number = 0) => + Number.isFinite(v) ? v.toFixed(d) : "-" + ), +})); + +function makeCostResponse(overrides: Partial = {}): CostEstimateResponse { + return { + model: "gpt-4", + input_tokens: 1000, + output_tokens: 500, + num_requests_per_day: 100, + num_requests_per_month: null, + cost_per_request: 0.05, + input_cost_per_request: 0.03, + output_cost_per_request: 0.02, + margin_cost_per_request: 0, + daily_cost: 5.0, + daily_input_cost: 3.0, + daily_output_cost: 2.0, + daily_margin_cost: 0, + monthly_cost: null, + monthly_input_cost: null, + monthly_output_cost: null, + monthly_margin_cost: null, + input_cost_per_token: null, + output_cost_per_token: null, + provider: "openai", + ...overrides, + }; +} + +function makeMultiResult(overrides: Partial = {}): MultiModelResult { + return { + entries: [ + { + entry: { id: "e1", model: "gpt-4", input_tokens: 1000, output_tokens: 500 }, + result: makeCostResponse(), + loading: false, + error: null, + }, + ], + totals: { + cost_per_request: 0.05, + daily_cost: 5.0, + monthly_cost: null, + margin_per_request: 0, + daily_margin: null, + monthly_margin: null, + }, + ...overrides, + }; +} + +function emptyMultiResult(): MultiModelResult { + return { + entries: [ + { + entry: { id: "e1", model: "", input_tokens: 1000, output_tokens: 500 }, + result: null, + loading: false, + error: null, + }, + ], + totals: { + cost_per_request: 0, + daily_cost: null, + monthly_cost: null, + margin_per_request: 0, + daily_margin: null, + monthly_margin: null, + }, + }; +} + +describe("MultiCostResults", () => { + beforeEach(() => { + vi.clearAllMocks(); + }); + + describe("when no model has been selected", () => { + it("should show a prompt to select models", () => { + renderWithProviders( + + ); + expect(screen.getByText(/select models above to see cost estimates/i)).toBeInTheDocument(); + }); + }); + + describe("when results are loading and no data has arrived yet", () => { + it("should show a calculating costs spinner", () => { + const multiResult: MultiModelResult = { + entries: [ + { + entry: { id: "e1", model: "gpt-4", input_tokens: 1000, output_tokens: 500 }, + result: null, + loading: true, + error: null, + }, + ], + totals: { + cost_per_request: 0, + daily_cost: null, + monthly_cost: null, + margin_per_request: 0, + daily_margin: null, + monthly_margin: null, + }, + }; + + renderWithProviders(); + expect(screen.getByText(/calculating costs/i)).toBeInTheDocument(); + }); + }); + + describe("when there are errors but no valid results", () => { + it("should display the error message with the model name", () => { + const multiResult: MultiModelResult = { + entries: [ + { + entry: { id: "e1", model: "bad-model", input_tokens: 0, output_tokens: 0 }, + result: null, + loading: false, + error: "Pricing not found", + }, + ], + totals: { + cost_per_request: 0, + daily_cost: null, + monthly_cost: null, + margin_per_request: 0, + daily_margin: null, + monthly_margin: null, + }, + }; + + renderWithProviders(); + expect(screen.getByText(/bad-model/i)).toBeInTheDocument(); + expect(screen.getByText(/Pricing not found/i)).toBeInTheDocument(); + }); + }); + + describe("when valid results are available", () => { + it("should show the Cost Estimates heading", () => { + renderWithProviders( + + ); + expect(screen.getByText("Cost Estimates")).toBeInTheDocument(); + }); + + it("should display the Total Per Request statistic", () => { + renderWithProviders( + + ); + expect(screen.getByText("Total Per Request")).toBeInTheDocument(); + }); + + it("should display Total Daily statistic when timePeriod is day", () => { + renderWithProviders( + + ); + expect(screen.getByText("Total Daily")).toBeInTheDocument(); + }); + + it("should display Total Monthly statistic when timePeriod is month", () => { + renderWithProviders( + + ); + expect(screen.getByText("Total Monthly")).toBeInTheDocument(); + }); + + it("should show the model name in the summary table", () => { + renderWithProviders( + + ); + expect(screen.getByText("gpt-4")).toBeInTheDocument(); + }); + + it("should show the provider tag next to the model name", () => { + renderWithProviders( + + ); + expect(screen.getByText("openai")).toBeInTheDocument(); + }); + + it("should show the Export button when results are available", () => { + renderWithProviders( + + ); + expect(screen.getByRole("button", { name: /export/i })).toBeInTheDocument(); + }); + + it("should expand the model breakdown row when the expand button is clicked", async () => { + const user = userEvent.setup(); + renderWithProviders( + + ); + + // The expand column renders a button (RightOutlined icon) for rows without errors + const expandButtons = screen.getAllByRole("button"); + // Find the small expand button (not the Export button) + const expandButton = expandButtons.find( + (btn) => !btn.textContent?.toLowerCase().includes("export") + ); + expect(expandButton).toBeDefined(); + + await user.click(expandButton!); + + // After expanding, the SingleModelBreakdown should be visible + expect(screen.getByText("Total/Request")).toBeInTheDocument(); + }); + + it("should show the collapse icon after expanding a row", async () => { + const user = userEvent.setup(); + renderWithProviders( + + ); + + const getExpandButton = () => { + const allButtons = screen.getAllByRole("button"); + return allButtons.find((btn) => !btn.textContent?.toLowerCase().includes("export")); + }; + + // Before expand: button has the "down" aria-label (RightOutlined renders as down in ant icons) + // Just verify clicking works and the breakdown content appears + await user.click(getExpandButton()!); + expect(screen.getByText("Total/Request")).toBeInTheDocument(); + + // After a second click, the row collapses — content may be hidden or removed + await user.click(getExpandButton()!); + // The expanded content should no longer be visible + expect(screen.queryByText("Total/Request")).not.toBeVisible(); + }); + }); + + describe("margin section", () => { + it("should show margin fee details when margin per request is greater than zero", () => { + const multiResult = makeMultiResult({ + entries: [ + { + entry: { id: "e1", model: "gpt-4", input_tokens: 1000, output_tokens: 500 }, + result: makeCostResponse({ margin_cost_per_request: 0.01, daily_margin_cost: 1.0 }), + loading: false, + error: null, + }, + ], + totals: { + cost_per_request: 0.06, + daily_cost: 6.0, + monthly_cost: null, + margin_per_request: 0.01, + daily_margin: 1.0, + monthly_margin: null, + }, + }); + + renderWithProviders(); + expect(screen.getByText("Margin Fee/Request")).toBeInTheDocument(); + }); + + it("should not show margin fee details when margin per request is zero", () => { + renderWithProviders( + + ); + expect(screen.queryByText("Margin Fee/Request")).not.toBeInTheDocument(); + }); + }); + + describe("when a model has zero cost", () => { + it("should show a warning about missing pricing data", () => { + const multiResult = makeMultiResult({ + entries: [ + { + entry: { id: "e1", model: "custom-model", input_tokens: 1000, output_tokens: 500 }, + result: makeCostResponse({ model: "custom-model", cost_per_request: 0 }), + loading: false, + error: null, + }, + ], + }); + + renderWithProviders(); + expect(screen.getByText(/no pricing data found/i)).toBeInTheDocument(); + }); + }); +}); diff --git a/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_export_dropdown.test.tsx b/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_export_dropdown.test.tsx new file mode 100644 index 00000000000..be1a89cf77f --- /dev/null +++ b/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_export_dropdown.test.tsx @@ -0,0 +1,146 @@ +import React from "react"; +import { describe, it, expect, vi, beforeEach, afterEach } from "vitest"; +import { screen, fireEvent } from "@testing-library/react"; +import userEvent from "@testing-library/user-event"; +import { renderWithProviders } from "../../../../tests/test-utils"; +import MultiExportDropdown from "./multi_export_dropdown"; +import type { MultiModelResult } from "./types"; + +vi.mock("./multi_export_utils", () => ({ + exportMultiToPDF: vi.fn(), + exportMultiToCSV: vi.fn(), +})); + +import { exportMultiToPDF, exportMultiToCSV } from "./multi_export_utils"; + +function makeMultiResult(hasResult: boolean): MultiModelResult { + return { + entries: [ + { + entry: { id: "e1", model: "gpt-4", input_tokens: 1000, output_tokens: 500 }, + result: hasResult + ? { + model: "gpt-4", + input_tokens: 1000, + output_tokens: 500, + num_requests_per_day: null, + num_requests_per_month: null, + cost_per_request: 0.05, + input_cost_per_request: 0.03, + output_cost_per_request: 0.02, + margin_cost_per_request: 0, + daily_cost: null, + daily_input_cost: null, + daily_output_cost: null, + daily_margin_cost: null, + monthly_cost: null, + monthly_input_cost: null, + monthly_output_cost: null, + monthly_margin_cost: null, + input_cost_per_token: null, + output_cost_per_token: null, + provider: "openai", + } + : null, + loading: false, + error: null, + }, + ], + totals: { + cost_per_request: hasResult ? 0.05 : 0, + daily_cost: null, + monthly_cost: null, + margin_per_request: 0, + daily_margin: null, + monthly_margin: null, + }, + }; +} + +describe("MultiExportDropdown", () => { + beforeEach(() => { + vi.clearAllMocks(); + }); + + it("should not render anything when no entries have results", () => { + const { container } = renderWithProviders( + + ); + expect(container.firstChild).toBeNull(); + }); + + it("should render the Export button when at least one entry has a result", () => { + renderWithProviders(); + expect(screen.getByRole("button", { name: /^export$/i })).toBeInTheDocument(); + }); + + it("should show the export menu when the Export button is clicked", async () => { + const user = userEvent.setup(); + renderWithProviders(); + + await user.click(screen.getByRole("button", { name: /^export$/i })); + + expect(screen.getByText("Export as PDF")).toBeInTheDocument(); + expect(screen.getByText("Export as CSV")).toBeInTheDocument(); + }); + + it("should hide the export menu when the Export button is clicked again", async () => { + const user = userEvent.setup(); + renderWithProviders(); + + await user.click(screen.getByRole("button", { name: /^export$/i })); + expect(screen.getByText("Export as PDF")).toBeInTheDocument(); + + await user.click(screen.getByRole("button", { name: /^export$/i })); + expect(screen.queryByText("Export as PDF")).not.toBeInTheDocument(); + }); + + it("should call exportMultiToPDF and close the menu when Export as PDF is clicked", async () => { + const user = userEvent.setup(); + renderWithProviders(); + + await user.click(screen.getByRole("button", { name: /^export$/i })); + await user.click(screen.getByText("Export as PDF")); + + expect(exportMultiToPDF).toHaveBeenCalledTimes(1); + expect(screen.queryByText("Export as PDF")).not.toBeInTheDocument(); + }); + + it("should call exportMultiToCSV and close the menu when Export as CSV is clicked", async () => { + const user = userEvent.setup(); + renderWithProviders(); + + await user.click(screen.getByRole("button", { name: /^export$/i })); + await user.click(screen.getByText("Export as CSV")); + + expect(exportMultiToCSV).toHaveBeenCalledTimes(1); + expect(screen.queryByText("Export as CSV")).not.toBeInTheDocument(); + }); + + it("should pass the multiResult to the export functions", async () => { + const user = userEvent.setup(); + const multiResult = makeMultiResult(true); + renderWithProviders(); + + await user.click(screen.getByRole("button", { name: /^export$/i })); + await user.click(screen.getByText("Export as PDF")); + + expect(exportMultiToPDF).toHaveBeenCalledWith(multiResult); + }); + + it("should close the menu when clicking outside", async () => { + const user = userEvent.setup(); + renderWithProviders( +
+ +
Outside
+
+ ); + + await user.click(screen.getByRole("button", { name: /^export$/i })); + expect(screen.getByText("Export as PDF")).toBeInTheDocument(); + + fireEvent.mouseDown(screen.getByTestId("outside")); + expect(screen.queryByText("Export as PDF")).not.toBeInTheDocument(); + }); +}); diff --git a/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_export_utils.test.ts b/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_export_utils.test.ts new file mode 100644 index 00000000000..40d6879d1af --- /dev/null +++ b/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/multi_export_utils.test.ts @@ -0,0 +1,274 @@ +import { describe, it, expect, vi, beforeEach, afterEach } from "vitest"; +import { exportMultiToPDF, exportMultiToCSV } from "./multi_export_utils"; +import type { MultiModelResult } from "./types"; +import type { CostEstimateResponse } from "../types"; + +vi.mock("@/utils/dataUtils", () => ({ + formatNumberWithCommas: vi.fn((v: number, d: number = 0) => + Number.isFinite(v) ? v.toFixed(d) : "-" + ), +})); + +function makeCostResponse(overrides: Partial = {}): CostEstimateResponse { + return { + model: "gpt-4", + input_tokens: 1000, + output_tokens: 500, + num_requests_per_day: 100, + num_requests_per_month: 3000, + cost_per_request: 0.05, + input_cost_per_request: 0.03, + output_cost_per_request: 0.02, + margin_cost_per_request: 0, + daily_cost: 5.0, + daily_input_cost: 3.0, + daily_output_cost: 2.0, + daily_margin_cost: 0, + monthly_cost: 150.0, + monthly_input_cost: 90.0, + monthly_output_cost: 60.0, + monthly_margin_cost: 0, + input_cost_per_token: 0.00003, + output_cost_per_token: 0.00004, + provider: "openai", + ...overrides, + }; +} + +function makeMultiResult(overrides: Partial = {}): MultiModelResult { + return { + entries: [ + { + entry: { id: "entry-1", model: "gpt-4", input_tokens: 1000, output_tokens: 500 }, + result: makeCostResponse(), + loading: false, + error: null, + }, + ], + totals: { + cost_per_request: 0.05, + daily_cost: 5.0, + monthly_cost: 150.0, + margin_per_request: 0, + daily_margin: null, + monthly_margin: null, + }, + ...overrides, + }; +} + +describe("exportMultiToPDF", () => { + let mockPrintWindow: { + document: { write: ReturnType; close: ReturnType }; + print: ReturnType; + onload: (() => void) | null; + }; + + beforeEach(() => { + mockPrintWindow = { + document: { write: vi.fn(), close: vi.fn() }, + print: vi.fn(), + onload: null, + }; + vi.spyOn(window, "open").mockReturnValue(mockPrintWindow as unknown as Window); + }); + + afterEach(() => { + vi.restoreAllMocks(); + }); + + it("should open a new popup window", () => { + exportMultiToPDF(makeMultiResult()); + expect(window.open).toHaveBeenCalledWith("", "_blank"); + }); + + it("should write HTML containing the report title", () => { + exportMultiToPDF(makeMultiResult()); + const html = mockPrintWindow.document.write.mock.calls[0][0] as string; + expect(html).toContain("LLM Cost Estimate Report"); + }); + + it("should include model name and provider in the generated HTML", () => { + exportMultiToPDF(makeMultiResult()); + const html = mockPrintWindow.document.write.mock.calls[0][0] as string; + expect(html).toContain("gpt-4"); + expect(html).toContain("openai"); + }); + + it("should close the document after writing", () => { + exportMultiToPDF(makeMultiResult()); + expect(mockPrintWindow.document.close).toHaveBeenCalledTimes(1); + }); + + it("should call print after the window finishes loading", () => { + exportMultiToPDF(makeMultiResult()); + expect(mockPrintWindow.print).not.toHaveBeenCalled(); + mockPrintWindow.onload!(); + expect(mockPrintWindow.print).toHaveBeenCalledTimes(1); + }); + + it("should show the margin section when margin per request is greater than zero", () => { + const multiResult = makeMultiResult({ + totals: { + cost_per_request: 0.06, + daily_cost: 5.0, + monthly_cost: 150.0, + margin_per_request: 0.01, + daily_margin: 1.0, + monthly_margin: 30.0, + }, + }); + exportMultiToPDF(multiResult); + const html = mockPrintWindow.document.write.mock.calls[0][0] as string; + expect(html).toContain("Margin/Request"); + }); + + it("should not show the margin section when margin per request is zero", () => { + exportMultiToPDF(makeMultiResult()); + const html = mockPrintWindow.document.write.mock.calls[0][0] as string; + expect(html).not.toContain("Margin/Request"); + }); + + it("should alert when popup is blocked", () => { + vi.spyOn(window, "open").mockReturnValue(null); + const alertSpy = vi.spyOn(window, "alert").mockImplementation(() => {}); + exportMultiToPDF(makeMultiResult()); + expect(alertSpy).toHaveBeenCalledWith("Please allow popups to export PDF"); + }); + + it("should only include entries that have a result", () => { + const multiResult: MultiModelResult = { + entries: [ + { entry: { id: "e1", model: "gpt-4", input_tokens: 1000, output_tokens: 500 }, result: null, loading: false, error: null }, + { entry: { id: "e2", model: "claude-3", input_tokens: 500, output_tokens: 250 }, result: makeCostResponse({ model: "claude-3", provider: "anthropic" }), loading: false, error: null }, + ], + totals: { cost_per_request: 0.05, daily_cost: 5.0, monthly_cost: 150.0, margin_per_request: 0, daily_margin: null, monthly_margin: null }, + }; + exportMultiToPDF(multiResult); + const html = mockPrintWindow.document.write.mock.calls[0][0] as string; + expect(html).toContain("1 model configured"); + expect(html).toContain("claude-3"); + }); + + it("should show plural 'models' when multiple results are present", () => { + const multiResult: MultiModelResult = { + entries: [ + { entry: { id: "e1", model: "gpt-4", input_tokens: 1000, output_tokens: 500 }, result: makeCostResponse(), loading: false, error: null }, + { entry: { id: "e2", model: "claude-3", input_tokens: 500, output_tokens: 250 }, result: makeCostResponse({ model: "claude-3" }), loading: false, error: null }, + ], + totals: { cost_per_request: 0.10, daily_cost: 10.0, monthly_cost: 300.0, margin_per_request: 0, daily_margin: null, monthly_margin: null }, + }; + exportMultiToPDF(multiResult); + const html = mockPrintWindow.document.write.mock.calls[0][0] as string; + expect(html).toContain("2 models configured"); + }); +}); + +describe("exportMultiToCSV", () => { + beforeEach(() => { + document.body.innerHTML = ""; + window.URL.createObjectURL = vi.fn(() => "blob:mock-url"); + window.URL.revokeObjectURL = vi.fn(); + }); + + afterEach(() => { + vi.restoreAllMocks(); + }); + + it("should create an object URL and revoke it after download", () => { + exportMultiToCSV(makeMultiResult()); + expect(window.URL.createObjectURL).toHaveBeenCalledTimes(1); + expect(window.URL.revokeObjectURL).toHaveBeenCalledWith("blob:mock-url"); + }); + + it("should set the download filename to include today's date", () => { + const createdAnchors: HTMLAnchorElement[] = []; + const originalCreate = document.createElement.bind(document); + vi.spyOn(document, "createElement").mockImplementation((tag: string) => { + const el = originalCreate(tag); + if (tag === "a") createdAnchors.push(el as HTMLAnchorElement); + return el; + }); + + const today = new Date().toISOString().split("T")[0]; + exportMultiToCSV(makeMultiResult()); + + expect(createdAnchors[0].download).toBe(`cost_estimate_multi_model_${today}.csv`); + }); + + it("should generate CSV content containing a header row and model data", () => { + let csvContent = ""; + const OriginalBlob = globalThis.Blob; + globalThis.Blob = class extends OriginalBlob { + constructor(parts?: BlobPart[], options?: BlobPropertyBag) { + super(parts, options); + if (typeof parts?.[0] === "string") csvContent = parts[0]; + } + } as unknown as typeof Blob; + + exportMultiToCSV(makeMultiResult()); + globalThis.Blob = OriginalBlob; + + expect(csvContent).toContain("Model"); + expect(csvContent).toContain("Cost/Request"); + expect(csvContent).toContain("gpt-4"); + expect(csvContent).toContain("openai"); + }); + + it("should include the combined totals section in CSV", () => { + let csvContent = ""; + const OriginalBlob = globalThis.Blob; + globalThis.Blob = class extends OriginalBlob { + constructor(parts?: BlobPart[], options?: BlobPropertyBag) { + super(parts, options); + if (typeof parts?.[0] === "string") csvContent = parts[0]; + } + } as unknown as typeof Blob; + + exportMultiToCSV(makeMultiResult()); + globalThis.Blob = OriginalBlob; + + expect(csvContent).toContain("COMBINED TOTALS"); + }); + + it("should create a blob with the correct CSV mime type", () => { + let capturedType = ""; + const OriginalBlob = globalThis.Blob; + globalThis.Blob = class extends OriginalBlob { + constructor(parts?: BlobPart[], options?: BlobPropertyBag) { + super(parts, options); + if (options?.type) capturedType = options.type; + } + } as unknown as typeof Blob; + + exportMultiToCSV(makeMultiResult()); + globalThis.Blob = OriginalBlob; + + expect(capturedType).toBe("text/csv;charset=utf-8;"); + }); + + it("should skip entries with null results", () => { + const multiResult: MultiModelResult = { + entries: [ + { entry: { id: "e1", model: "gpt-4", input_tokens: 1000, output_tokens: 500 }, result: null, loading: false, error: null }, + ], + totals: { cost_per_request: 0, daily_cost: null, monthly_cost: null, margin_per_request: 0, daily_margin: null, monthly_margin: null }, + }; + + let csvContent = ""; + const OriginalBlob = globalThis.Blob; + globalThis.Blob = class extends OriginalBlob { + constructor(parts?: BlobPart[], options?: BlobPropertyBag) { + super(parts, options); + if (typeof parts?.[0] === "string") csvContent = parts[0]; + } + } as unknown as typeof Blob; + + exportMultiToCSV(multiResult); + globalThis.Blob = OriginalBlob; + + // CSV should have metadata rows but no model data row for gpt-4 + const lines = csvContent.split("\n").filter((l) => l.includes('"gpt-4"')); + expect(lines).toHaveLength(0); + }); +}); diff --git a/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/use_multi_cost_estimate.test.ts b/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/use_multi_cost_estimate.test.ts new file mode 100644 index 00000000000..f5715194f58 --- /dev/null +++ b/ui/litellm-dashboard/src/components/CostTrackingSettings/pricing_calculator/use_multi_cost_estimate.test.ts @@ -0,0 +1,342 @@ +import { describe, it, expect, vi, beforeEach, afterEach } from "vitest"; +import { renderHook, act } from "@testing-library/react"; +import { useMultiCostEstimate } from "./use_multi_cost_estimate"; +import type { ModelEntry } from "./types"; +import type { CostEstimateResponse } from "../types"; + +vi.mock("@/components/networking", () => ({ + getProxyBaseUrl: vi.fn(() => ""), + getGlobalLitellmHeaderName: vi.fn(() => "Authorization"), +})); + +function makeEntry(overrides: Partial = {}): ModelEntry { + return { + id: "entry-1", + model: "gpt-4", + input_tokens: 1000, + output_tokens: 500, + ...overrides, + }; +} + +function makeApiResponse(overrides: Partial = {}): CostEstimateResponse { + return { + model: "gpt-4", + input_tokens: 1000, + output_tokens: 500, + num_requests_per_day: null, + num_requests_per_month: null, + cost_per_request: 0.05, + input_cost_per_request: 0.03, + output_cost_per_request: 0.02, + margin_cost_per_request: 0, + daily_cost: null, + daily_input_cost: null, + daily_output_cost: null, + daily_margin_cost: null, + monthly_cost: null, + monthly_input_cost: null, + monthly_output_cost: null, + monthly_margin_cost: null, + input_cost_per_token: 0.00003, + output_cost_per_token: 0.00004, + provider: "openai", + ...overrides, + }; +} + +describe("useMultiCostEstimate", () => { + beforeEach(() => { + vi.clearAllMocks(); + vi.useFakeTimers(); + }); + + afterEach(() => { + vi.useRealTimers(); + }); + + describe("debouncedFetchForEntry", () => { + it("should not fetch when access token is null", async () => { + const fetchSpy = vi.spyOn(global, "fetch"); + const { result } = renderHook(() => useMultiCostEstimate(null)); + + await act(async () => { + result.current.debouncedFetchForEntry(makeEntry()); + await vi.runAllTimersAsync(); + }); + + expect(fetchSpy).not.toHaveBeenCalled(); + }); + + it("should not fetch when the model field is empty", async () => { + const fetchSpy = vi.spyOn(global, "fetch"); + const { result } = renderHook(() => useMultiCostEstimate("token123")); + + await act(async () => { + result.current.debouncedFetchForEntry(makeEntry({ model: "" })); + await vi.runAllTimersAsync(); + }); + + expect(fetchSpy).not.toHaveBeenCalled(); + }); + + it("should not fetch immediately — only after the debounce delay", async () => { + const fetchSpy = vi.spyOn(global, "fetch").mockResolvedValue({ + ok: true, + json: async () => makeApiResponse(), + } as Response); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + + act(() => { + result.current.debouncedFetchForEntry(makeEntry()); + }); + + expect(fetchSpy).not.toHaveBeenCalled(); + + await act(async () => { + await vi.runAllTimersAsync(); + }); + + expect(fetchSpy).toHaveBeenCalledTimes(1); + }); + + it("should cancel an in-flight debounce when called again for the same entry", async () => { + const fetchSpy = vi.spyOn(global, "fetch").mockResolvedValue({ + ok: true, + json: async () => makeApiResponse(), + } as Response); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + + await act(async () => { + result.current.debouncedFetchForEntry(makeEntry()); + vi.advanceTimersByTime(200); + result.current.debouncedFetchForEntry(makeEntry()); + vi.advanceTimersByTime(200); + result.current.debouncedFetchForEntry(makeEntry()); + await vi.runAllTimersAsync(); + }); + + expect(fetchSpy).toHaveBeenCalledTimes(1); + }); + + it("should store the API result after a successful fetch", async () => { + vi.spyOn(global, "fetch").mockResolvedValue({ + ok: true, + json: async () => makeApiResponse(), + } as Response); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + const entry = makeEntry(); + + await act(async () => { + result.current.debouncedFetchForEntry(entry); + await vi.runAllTimersAsync(); + }); + + const multiResult = result.current.getMultiModelResult([entry]); + expect(multiResult.entries[0].result).not.toBeNull(); + expect(multiResult.entries[0].result?.cost_per_request).toBe(0.05); + expect(multiResult.entries[0].loading).toBe(false); + expect(multiResult.entries[0].error).toBeNull(); + }); + + it("should set an error message when the API returns a non-ok response", async () => { + vi.spyOn(global, "fetch").mockResolvedValue({ + ok: false, + json: async () => ({ detail: { error: "Model not found" } }), + } as Response); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + const entry = makeEntry(); + + await act(async () => { + result.current.debouncedFetchForEntry(entry); + await vi.runAllTimersAsync(); + }); + + const multiResult = result.current.getMultiModelResult([entry]); + expect(multiResult.entries[0].result).toBeNull(); + expect(multiResult.entries[0].error).toBe("Model not found"); + }); + + it("should fall back to detail string when error has no nested error field", async () => { + vi.spyOn(global, "fetch").mockResolvedValue({ + ok: false, + json: async () => ({ detail: "Bad request" }), + } as Response); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + const entry = makeEntry(); + + await act(async () => { + result.current.debouncedFetchForEntry(entry); + await vi.runAllTimersAsync(); + }); + + const multiResult = result.current.getMultiModelResult([entry]); + expect(multiResult.entries[0].error).toBe("Bad request"); + }); + + it("should set 'Network error' when fetch throws", async () => { + vi.spyOn(global, "fetch").mockRejectedValue(new Error("connection refused")); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + const entry = makeEntry(); + + await act(async () => { + result.current.debouncedFetchForEntry(entry); + await vi.runAllTimersAsync(); + }); + + const multiResult = result.current.getMultiModelResult([entry]); + expect(multiResult.entries[0].error).toBe("Network error"); + expect(multiResult.entries[0].result).toBeNull(); + }); + }); + + describe("removeEntry", () => { + it("should remove an entry's cached result", async () => { + vi.spyOn(global, "fetch").mockResolvedValue({ + ok: true, + json: async () => makeApiResponse(), + } as Response); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + const entry = makeEntry(); + + await act(async () => { + result.current.debouncedFetchForEntry(entry); + await vi.runAllTimersAsync(); + }); + + // Confirm result was stored + expect(result.current.getMultiModelResult([entry]).entries[0].result).not.toBeNull(); + + act(() => { + result.current.removeEntry(entry.id); + }); + + // After removal, the entry should return as if it never fetched + const multiResult = result.current.getMultiModelResult([entry]); + expect(multiResult.entries[0].result).toBeNull(); + }); + + it("should cancel a pending debounce for the removed entry", async () => { + const fetchSpy = vi.spyOn(global, "fetch").mockResolvedValue({ + ok: true, + json: async () => makeApiResponse(), + } as Response); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + const entry = makeEntry(); + + act(() => { + result.current.debouncedFetchForEntry(entry); + result.current.removeEntry(entry.id); + }); + + await act(async () => { + await vi.runAllTimersAsync(); + }); + + expect(fetchSpy).not.toHaveBeenCalled(); + }); + }); + + describe("getMultiModelResult", () => { + it("should return zero totals when no entries have results", () => { + const { result } = renderHook(() => useMultiCostEstimate("token123")); + const multiResult = result.current.getMultiModelResult([makeEntry()]); + + expect(multiResult.totals.cost_per_request).toBe(0); + expect(multiResult.totals.margin_per_request).toBe(0); + expect(multiResult.totals.daily_cost).toBeNull(); + expect(multiResult.totals.monthly_cost).toBeNull(); + }); + + it("should return an empty entries array for an empty input list", () => { + const { result } = renderHook(() => useMultiCostEstimate("token123")); + const multiResult = result.current.getMultiModelResult([]); + + expect(multiResult.entries).toHaveLength(0); + expect(multiResult.totals.daily_cost).toBeNull(); + expect(multiResult.totals.monthly_cost).toBeNull(); + }); + + it("should sum cost_per_request across multiple loaded entries", async () => { + const entry1 = makeEntry({ id: "e1", model: "gpt-4" }); + const entry2 = makeEntry({ id: "e2", model: "claude-3" }); + + let callIndex = 0; + const responses = [ + makeApiResponse({ cost_per_request: 0.05, margin_cost_per_request: 0 }), + makeApiResponse({ model: "claude-3", cost_per_request: 0.10, margin_cost_per_request: 0 }), + ]; + + vi.spyOn(global, "fetch").mockImplementation(async () => ({ + ok: true, + json: async () => responses[callIndex++], + } as Response)); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + + await act(async () => { + result.current.debouncedFetchForEntry(entry1); + result.current.debouncedFetchForEntry(entry2); + await vi.runAllTimersAsync(); + }); + + const multiResult = result.current.getMultiModelResult([entry1, entry2]); + expect(multiResult.totals.cost_per_request).toBeCloseTo(0.15); + }); + + it("should accumulate daily cost only when entries have a daily cost", async () => { + const entry1 = makeEntry({ id: "e1", model: "gpt-4" }); + const entry2 = makeEntry({ id: "e2", model: "claude-3" }); + + let callIndex = 0; + const responses = [ + makeApiResponse({ daily_cost: 5.0, daily_margin_cost: 0, monthly_cost: null, monthly_margin_cost: null }), + makeApiResponse({ model: "claude-3", daily_cost: 10.0, daily_margin_cost: 0, monthly_cost: null, monthly_margin_cost: null }), + ]; + + vi.spyOn(global, "fetch").mockImplementation(async () => ({ + ok: true, + json: async () => responses[callIndex++], + } as Response)); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + + await act(async () => { + result.current.debouncedFetchForEntry(entry1); + result.current.debouncedFetchForEntry(entry2); + await vi.runAllTimersAsync(); + }); + + const multiResult = result.current.getMultiModelResult([entry1, entry2]); + expect(multiResult.totals.daily_cost).toBeCloseTo(15.0); + expect(multiResult.totals.monthly_cost).toBeNull(); + }); + + it("should mark each entry's loading and error state from cached data", async () => { + vi.spyOn(global, "fetch").mockResolvedValue({ + ok: false, + json: async () => ({ detail: "Not found" }), + } as Response); + + const { result } = renderHook(() => useMultiCostEstimate("token123")); + const entry = makeEntry(); + + await act(async () => { + result.current.debouncedFetchForEntry(entry); + await vi.runAllTimersAsync(); + }); + + const multiResult = result.current.getMultiModelResult([entry]); + expect(multiResult.entries[0].error).toBe("Not found"); + expect(multiResult.entries[0].loading).toBe(false); + }); + }); +}); diff --git a/ui/litellm-dashboard/src/components/page_metadata.ts b/ui/litellm-dashboard/src/components/page_metadata.ts index 2459410d7a3..5e2c0ccc0b9 100644 --- a/ui/litellm-dashboard/src/components/page_metadata.ts +++ b/ui/litellm-dashboard/src/components/page_metadata.ts @@ -13,6 +13,7 @@ export const pageDescriptions: Record = { guardrails: "Set up content moderation and safety guardrails", policies: "Define access control and usage policies", "search-tools": "Configure RAG search and retrieval tools", + "tool-policies": "Configure tool use policies and permissions", "vector-stores": "Manage vector databases for embeddings", new_usage: "View usage analytics and metrics", logs: "Access request and response logs", From 5662228e207c5b5f49a727594949118f4a6524c4 Mon Sep 17 00:00:00 2001 From: Krish Dholakia Date: Wed, 25 Feb 2026 10:37:45 -0800 Subject: [PATCH 08/15] feat(ui): add user filtering to usage page (#22059) * feat(ui): add user filtering to usage page Adds "User Usage" as a new view option in the usage page dropdown, allowing admins to view and filter usage data by individual users via the existing /user/daily/activity backend endpoint. Co-Authored-By: Claude Opus 4.6 * feat(ui/): working usage filtering * fix(ui): use single-select for user filter and add tests The user entity type's backend endpoint only accepts a single user_id, so the filter now uses single-select mode instead of multi-select. Added tests for the new user entity type in EntityUsage and UsageViewSelect. Updated CLAUDE.md and AGENTS.md with guidance on UI/backend contract consistency and test coverage for new entity types. Co-Authored-By: Claude Opus 4.6 * revert: remove unintended package-lock.json changes Co-Authored-By: Claude Opus 4.6 * revert: restore package-lock.json to merge base state Co-Authored-By: Claude Opus 4.6 --------- Co-authored-by: Claude Opus 4.6 --- AGENTS.md | 2 + CLAUDE.md | 4 + .../EntityUsageExport/UsageExportHeader.tsx | 14 +++- .../src/components/EntityUsageExport/types.ts | 2 +- .../EntityUsage/EntityUsage.test.tsx | 19 +++++ .../components/EntityUsage/EntityUsage.tsx | 11 +++ .../UsagePage/components/UsagePageView.tsx | 79 ++++++++++--------- .../UsageViewSelect/UsageViewSelect.test.tsx | 1 + .../UsageViewSelect/UsageViewSelect.tsx | 10 ++- ui/litellm-dashboard/tsconfig.json | 2 +- 10 files changed, 102 insertions(+), 42 deletions(-) diff --git a/AGENTS.md b/AGENTS.md index 5a48049ef45..bfd44304d55 100644 --- a/AGENTS.md +++ b/AGENTS.md @@ -174,6 +174,8 @@ When opening issues or pull requests, follow these templates: 3. **Rate Limits**: Respect provider rate limits in tests 4. **Memory Usage**: Be mindful of memory usage in streaming scenarios 5. **Dependencies**: Keep dependencies minimal and well-justified +6. **UI/Backend Contract Mismatch**: When adding a new entity type to the UI, always check whether the backend endpoint accepts a single value or an array. Match the UI control accordingly (single-select vs. multi-select) to avoid silently dropping user selections +7. **Missing Tests for New Entity Types**: When adding a new entity type (e.g., in `EntityUsage`, `UsageViewSelect`), always add corresponding tests in the existing test files and update any icon/component mocks ## HELPFUL RESOURCES diff --git a/CLAUDE.md b/CLAUDE.md index 3cb67908076..3b597fb8a90 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -97,6 +97,10 @@ LiteLLM is a unified interface for 100+ LLM providers with two main components: - Integration tests for each provider in `tests/llm_translation/` - Proxy tests in `tests/proxy_unit_tests/` - Load tests in `tests/load_tests/` +- **Always add tests when adding new entity types or features** — if the existing test file covers other entity types, add corresponding tests for the new one + +### UI / Backend Consistency +- When wiring a new UI entity type to an existing backend endpoint, verify the backend API contract (single value vs. array, required vs. optional params) and ensure the UI controls match — e.g., use a single-select dropdown when the backend accepts a single value, not a multi-select ### Database Migrations - Prisma handles schema migrations diff --git a/ui/litellm-dashboard/src/components/EntityUsageExport/UsageExportHeader.tsx b/ui/litellm-dashboard/src/components/EntityUsageExport/UsageExportHeader.tsx index 989ff8703e8..3c9b695a7dd 100644 --- a/ui/litellm-dashboard/src/components/EntityUsageExport/UsageExportHeader.tsx +++ b/ui/litellm-dashboard/src/components/EntityUsageExport/UsageExportHeader.tsx @@ -17,6 +17,7 @@ interface UsageExportHeaderProps { selectedFilters?: string[]; onFiltersChange?: (filters: string[]) => void; filterOptions?: Array<{ label: string; value: string }>; + filterMode?: "multiple" | "single"; customTitle?: string; compactLayout?: boolean; teams?: Team[]; @@ -32,6 +33,7 @@ const UsageExportHeader: React.FC = ({ selectedFilters = [], onFiltersChange, filterOptions = [], + filterMode = "multiple", customTitle, compactLayout = false, teams = [], @@ -59,11 +61,17 @@ const UsageExportHeader: React.FC = ({
{filterLabel && {filterLabel}} setSelectedUserId(value ?? null)} + filterOption={false} + onSearch={handleUserSearchChange} + searchValue={userSearchInput} + onPopupScroll={handleUserPopupScroll} + loading={isLoadingUsers} + notFoundContent={isLoadingUsers ? : "No users found"} + options={userOptions} + popupRender={(menu) => ( + <> + {menu} + {isFetchingNextUsersPage && ( +
+ +
+ )} + + )} + /> +
+ )}
@@ -560,41 +590,6 @@ const UsagePage: React.FC = ({ teams, organizations }) => { )} - {isAdmin && ( -
- - + + + prev.allowed_mcp_servers_and_groups !== curr.allowed_mcp_servers_and_groups || + prev.mcp_tool_permissions !== curr.mcp_tool_permissions + } + > + {() => ( +
+ ) => form.setFieldsValue({ mcp_tool_permissions: toolPerms })} + /> +
+ )} + +
+ ); + const handleAgentTypeChange = (value: string) => { setAgentType(value); form.resetFields(); @@ -412,10 +512,16 @@ const AddAgentForm: React.FC = ({