litellm/tests/test_litellm/proxy/db
tin-berri 4e48d74455
feat(shadow_eval): measure both arms' cost so a job reports what the router would have saved (#38631)
The attempt row now prices the real arm (the payload's response_cost plus its own
routing classifier when it routed) beside the shadow arm (completion plus the
classifier cost the routing decision writes back), and flags turns litellm's
response cache served. A per-leg funnel table counts the eligible requests that
produced no row (lost the sampling dice, unjudgeable shape, concurrency shed),
so results can weigh judged rows against the traffic they stand for. Job results
gain per-slice and overall arm spends plus the coverage counts, the budget gates
charge the shadow arm's classifier spend against max_budget, and the dashboard
shows the measured cost comparison beside the win rate

Resolves LIT-6358

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-28 15:13:19 -07:00
..
db_transaction_queue feat(spend): report prompt caching savings as total and gateway-attributed (#38134) 2026-08-28 00:19:06 -07:00
mcp_server refactor(repositories): type prisma table access with one generic protocol 2026-08-25 12:14:17 +00:00
conftest.py feat(proxy): authenticate to Azure Postgres with Microsoft Entra ID tokens 2026-08-20 11:50:16 -07:00
test_autorouter_session_rollup.py feat(shadow_eval): measure both arms' cost so a job reports what the router would have saved (#38631) 2026-08-28 15:13:19 -07:00
test_check_migration.py test: drop the cwd-relative sys.path.insert calls from the test suite (#37802) 2026-08-22 09:25:58 -07:00
test_create_views.py fix(proxy): guard every view creation, not just the first and last 2026-08-13 14:41:59 -07:00
test_daily_spend_bulk_upsert.py feat(spend): report prompt caching savings as total and gateway-attributed (#38134) 2026-08-28 00:19:06 -07:00
test_db_spend_update_writer.py feat(spend): report prompt caching savings as total and gateway-attributed (#38134) 2026-08-28 00:19:06 -07:00
test_db_url_settings.py fix(proxy): honor DATABASE_DISABLE_PREPARED_STATEMENTS in componentized entrypoints (#38363) 2026-08-26 10:39:04 -07:00
test_exception_handler.py test: drop the cwd-relative sys.path.insert calls from the test suite (#37802) 2026-08-22 09:25:58 -07:00
test_exception_handler_reconnect_retry.py test: drop the cwd-relative sys.path.insert calls from the test suite (#37802) 2026-08-22 09:25:58 -07:00
test_gateway_request_tracking.py feat(sgr): make the gateway middleware the source of truth for successful requests (#35717) 2026-08-05 12:40:47 -07:00
test_prisma_client.py fix: keep schema reconciliation from fighting a partitioned LiteLLM_SpendLogs (#38452) 2026-08-27 12:52:52 -07:00
test_prisma_planned_engine_restart.py test: drop the cwd-relative sys.path.insert calls from the test suite (#37802) 2026-08-22 09:25:58 -07:00
test_prisma_self_heal.py test: drop the cwd-relative sys.path.insert calls from the test suite (#37802) 2026-08-22 09:25:58 -07:00
test_proxy_worker_heartbeat.py fix(proxy): count worker heartbeats on the primary so replica lag cannot undercount 2026-08-14 17:23:06 -07:00
test_query_engine_reaper.py fix(proxy_cli): reap orphaned prisma query-engine processes when a worker dies (#33424) 2026-07-16 13:16:41 -07:00
test_rds_iam_token_expiry.py fix(tests): stop DATABASE_URL env pollution from read-replica tests breaking DB e2e tests (#32653) 2026-07-09 14:37:49 -07:00
test_replica_identity.py feat(db): opt-in REPLICA IDENTITY FULL after prisma migrations (#35267) 2026-07-30 15:45:40 -07:00
test_routing_prisma_wrapper.py test: drop the cwd-relative sys.path.insert calls from the test suite (#37802) 2026-08-22 09:25:58 -07:00
test_shadow_eval_funnel.py feat(shadow_eval): measure both arms' cost so a job reports what the router would have saved (#38631) 2026-08-28 15:13:19 -07:00
test_spend_log_batching.py fix(spend): bound each spend-log write statement by row count as well as bytes (#37758) 2026-08-21 09:49:51 -07:00
test_spend_log_tool_index.py test: require a match= on broad pytest.raises, and drop duplicate parametrize cases (#37769) 2026-08-20 20:24:49 -07:00
test_token_auth.py fix(proxy): keep token-auth URLs, toggles, and refresh sleeps safe 2026-08-20 12:59:58 -07:00
test_tool_registry_writer.py test: drop the cwd-relative sys.path.insert calls from the test suite (#37802) 2026-08-22 09:25:58 -07:00