Tests are fully mocked but require WATSONX_URL and WATSONX_PROJECT_ID
to pass provider validation before reaching the mocked HTTP client.
Add an autouse fixture with dummy values and clear WATSONX_ZENAPIKEY/
WATSONX_TOKEN to prevent env leakage from dotenv.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The router_code_coverage.py script only scans test files with "router" in the filename.
test_health_check_routing.py was invisible to this check, causing _async_filter_health_check_unhealthy_deployments
and _filter_health_check_unhealthy_deployments to appear untested.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add reasoning_items field to ChatCompletionAssistantMessage TypedDict
and extract a typed _get_reasoning_items helper instead of using cast.
Also widen _reasoning_item_to_response_input to accept ChatCompletionReasoningItem.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The DeepInfra tests were making real API calls and failing with AuthenticationError.
Mock the HTTP layer to verify request shape instead.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The hanging-response constructor was fixed but the sibling failed-response
constructor at line 104 was still missing this field.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add missing `user_api_key_project_alias` key to SpendLogsMetadata and
PagerDutyInternalEvent constructors, and cast `reasoning_items` to list
for safe iteration in responses transformation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Ruff flagged `process_output_streaming_response` (A2A) and
`process_output_response` (Anthropic) for exceeding 50 statements.
Extract inline logic into private helpers to bring both under the limit.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The test only needs a bad key to verify AuthenticationError handling.
No other tests use this env var, and it's not provisioned in CI.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Removes 10 CircleCI jobs that are fully duplicated by GHA workflows:
- caching_unit_tests → test-unit-caching-redis.yml
- litellm_security_tests → test-unit-security.yml
- litellm_proxy_unit_testing_{key_generation,part1,part2} → test-unit-proxy-db.yml + test-unit-proxy-legacy.yml
- litellm_mapped_tests_{llms,core,litellm_core_utils,mcps,integrations} → test-unit-*.yml workflows
Also cleans up upload-coverage requires and workflow entries.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace Tremor Text with native <span> elements (with block class where
stacking is needed) and Tremor Badge with antd Tag. All Tailwind classes
preserved. Gray badge default mapped to undefined (antd Tag default).
Substring matching for user_id and key_alias on /key/list is now gated
by caller role (admin only) rather than by whether user_id was auto-set.
This closes the gap where a non-admin explicitly passing their own
user_id still got contains-based filtering, which could surface keys
belonging to other users with overlapping IDs.
Adds three unit tests covering the substring, exact-match, and default
code paths in _build_key_filter_conditions. Restores the existing
member-team test to assert exact matching as a security regression guard.
