mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-18 04:44:37 +00:00
* fix(router): stop skipping rate-limit checks when token counting fails Router._pre_call_checks() caught any token_counter() exception and returned the full, unfiltered deployment list, bypassing the RPM, region, and invalid-params checks for every deployment - not just the context-window check that actually depends on the count. A content block token_counter() cannot parse (e.g. an invalid image_url detail value) silently disabled rate-limit enforcement for that request, with only a log line and no error surfaced to the caller. Now a counting failure only disables the context-window check; the other checks still run for every deployment as before. Input-token resolution is pulled into its own method and computed once before the loop, so nothing needs to track "did counting fail" with a mutable flag across iterations. * test(router): cover the async region check when token counting fails Replace the unit test of the private input-token resolver with a regression test through async_get_healthy_deployments, the path the proxy uses: an EU-pinned request whose content cannot be token-counted must not get the US deployment * test(router): cover the param filter when token counting fails and satisfy router coverage * fix(router): answer 400 when no deployment takes the requested param, honor allowlists and drop_params The pre-call param filter now reuses the router's param-acceptance helper, so allowed_openai_params, drop_params, and additional_drop_params keep a deployment instead of filtering it out. When every available deployment lacks a requested param, the router answers 400 UnsupportedParamsError right away instead of a retried 429. Deployment context windows are resolved once per call. * fix(router): keep the retried 429 when a cooling-down deployment takes the param * fix(router): steer to a deployment that takes the param before dropping it, keep context-window fallbacks * fix(router): steer by deployment capability, apply drop settings after region and RPM checks A deployment-level allowed_openai_params no longer counts as taking the param when another deployment does, the drop and allow settings now apply only when no deployment that passed the region, RPM, and context-window checks takes it, and an authenticating provider is checked by its declared model instead of always passing * test(router): make the github_copilot pre-call filter test hermetic in CI * test(router): cover pre-call region and response_format filters through the proxy * fix(router): judge the 400 against the request's own deployments, not a name lookup * fix(router): judge the 400 against the passed deployments when the group name lists none * fix(router): let a request's drop and allow settings replace the deployment's in the param check --------- Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>
170 lines
9.7 KiB
Python
170 lines
9.7 KiB
Python
import ast
|
|
import os
|
|
|
|
|
|
def get_function_names_from_file(file_path):
|
|
"""
|
|
Extracts all function names from a given Python file.
|
|
"""
|
|
with open(file_path, "r", encoding="utf-8") as file:
|
|
tree = ast.parse(file.read())
|
|
|
|
function_names = []
|
|
|
|
for node in tree.body:
|
|
if isinstance(node, (ast.FunctionDef, ast.AsyncFunctionDef)):
|
|
# Top-level functions
|
|
function_names.append(node.name)
|
|
elif isinstance(node, ast.ClassDef):
|
|
# Functions inside classes
|
|
for class_node in node.body:
|
|
if isinstance(class_node, (ast.FunctionDef, ast.AsyncFunctionDef)):
|
|
function_names.append(class_node.name)
|
|
|
|
return function_names
|
|
|
|
|
|
def get_all_functions_called_in_tests(base_dir):
|
|
"""
|
|
Returns a set of function names that are called in test functions
|
|
inside the unit test directories,
|
|
specifically in files containing the word 'router'.
|
|
"""
|
|
called_functions = set()
|
|
test_dirs = ["unit"]
|
|
|
|
for test_dir in test_dirs:
|
|
dir_path = os.path.join(base_dir, test_dir)
|
|
if not os.path.exists(dir_path):
|
|
print(f"Warning: Directory {dir_path} does not exist.")
|
|
continue
|
|
|
|
print("dir_path: ", dir_path)
|
|
for root, _, files in os.walk(dir_path):
|
|
for file in files:
|
|
if file.endswith(".py") and ("router" in file.lower() or test_dir == "unit"):
|
|
print("file: ", file)
|
|
file_path = os.path.join(root, file)
|
|
with open(file_path, "r", encoding="utf-8") as f:
|
|
try:
|
|
tree = ast.parse(f.read())
|
|
except SyntaxError:
|
|
print(f"Warning: Syntax error in file {file_path}")
|
|
continue
|
|
if file == "test_router_validate_fallbacks.py":
|
|
print(f"tree: {tree}")
|
|
for node in ast.walk(tree):
|
|
if isinstance(node, ast.Call) and isinstance(node.func, ast.Name):
|
|
called_functions.add(node.func.id)
|
|
elif isinstance(node, ast.Call) and isinstance(node.func, ast.Attribute):
|
|
called_functions.add(node.func.attr)
|
|
|
|
return called_functions
|
|
|
|
|
|
def get_functions_from_router(file_path):
|
|
"""
|
|
Extracts all functions defined in router.py.
|
|
"""
|
|
return get_function_names_from_file(file_path)
|
|
|
|
|
|
ignored_function_names = [
|
|
"_acancel_batch",
|
|
"_acreate_batch",
|
|
"_acreate_file",
|
|
"__init__",
|
|
"avector_store_create", # Tested via proxy vector_store_endpoints (files lack "router" in name)
|
|
"_override_vector_store_methods_for_router", # No-op placeholder, called during Router init
|
|
"_merge_tools_from_deployment", # Tested indirectly via _update_kwargs_with_deployment (test files lack "router" in name)
|
|
"_invalidate_access_groups_cache", # Tested indirectly via set_model_list, upsert_model etc. (test files lack "router" in name)
|
|
"has_buffered_provider_output", # Property, so its reads in test_router.py are never an ast.Call
|
|
"chunks", # Property on FallbackAwareAnthropicMessagesStream, so its reads in tests are never an ast.Call
|
|
"messages", # Property on FallbackAwareAnthropicMessagesStream, so its reads in tests are never an ast.Call
|
|
"model", # Property on FallbackAwareAnthropicMessagesStream, so its reads in tests are never an ast.Call
|
|
"_routing_groups", # Property getter and setter reads are never ast.Call nodes
|
|
"_request_header", # Tested through Claude Code session routing in test_router.py
|
|
"_claude_code_session_router_cache_key", # Tested through Claude Code session routing in test_router.py
|
|
"_delete_claude_code_session_router_binding", # Tested through Redis cleanup failure in test_router.py
|
|
"_resolve_claude_code_session_router", # Tested through Claude Code session routing in test_router.py
|
|
"_get_claude_code_session_router_binding", # Tested through the two-worker session routing test in test_router.py
|
|
"_apply_updated_routing_strategy_args", # Tested via update_settings in test_lowest_latency.py (file lacks "router" in name)
|
|
"_wait_for_scheduler_turn", # Tested through prioritized acompletion and atext_completion in test_router.py
|
|
"arm_routing_read_prefetch", # Tested in tests/unit/caching/test_request_redis_batch_pre_call.py (file lacks "router" in name)
|
|
"_configured_model_info", # Tested through get_configured_service_tiers in test_router.py
|
|
"_routable_deployments", # Tested through get_configured_service_tiers and get_routable_upstream_model in test_router.py
|
|
"_async_get_available_deployment", # Body of the `route {model}` phase wrapper, exercised through async_get_available_deployment in test_router.py
|
|
"_async_get_available_deployment_for_pass_through", # Same, through async_get_available_deployment_for_pass_through in test_router.py
|
|
"_embedding",
|
|
"_aembedding",
|
|
"_stream_pre_content_error", # Tested through the non-retriable retry error tests in test_router.py
|
|
"_deployment_num_retries", # Tested through the deployment num_retries mid-stream budget test in test_router.py
|
|
"_request_fallback_list", # Tested through every mid-stream retry test in test_router.py
|
|
"_request_model_group", # Tested through test_anthropic_messages_retry_budget_precedence_direct_call
|
|
"_mid_stream_retry_trigger", # Tested through the retry policy mid-stream budget test in test_router.py
|
|
"_anthropic_messages_group_retry_policy", # Tested through the retry budget precedence test in test_router.py
|
|
"_anthropic_messages_resolved_retry_policy", # Tested through the malformed retry policy tests in test_router.py
|
|
"_anthropic_messages_plain_retry_budget", # Tested through the retry budget precedence test in test_router.py
|
|
"_stream_should_retry", # Tested through every mid-stream retry test in test_router.py
|
|
"_astream_retry_same_group", # Tested through the dropped-before-content retry tests in test_router.py
|
|
"_aanthropic_messages_yield_recovered", # Tested through every mid-stream retry and fallback test in test_router.py
|
|
"_anthropic_messages_policy_retries", # Tested through the retry budget precedence test in test_router.py
|
|
"_get_wildcard_deployments", # Tested through the get_model_list_of_routed_group wildcard test in test_router.py
|
|
"_is_fallback_hop", # Tested through the order fallback hop tests in test_router_order_fallback.py
|
|
"_deployment_that_just_failed", # Tested through the same-boundary hop test in test_router_order_fallback.py
|
|
"_collect_ptu_share_teams", # Cached behind team_ptu_ceiling, tested through the live model list test in test_router.py
|
|
"_resolve_team_ptu_ceiling", # Cached behind team_ptu_ceiling, tested through the live model list test in test_router.py
|
|
"_resolve_pre_call_check_input_tokens", # Tested through the token-count failure pre-call check tests in test_router.py
|
|
"_params_neither_dropped_nor_allowed", # Tested through the allowed-or-dropped param pre-call check tests in test_router.py
|
|
"_deployment_rejects_params", # Tested through the unsupported-param pre-call check tests in test_router.py
|
|
"_deployment_lacks_params", # Tested through the unsupported-param pre-call check tests in test_router.py
|
|
"_pre_call_check_requested_chat_params", # Tested through the unsupported-param pre-call check tests in test_router.py
|
|
"_deployment_supports_param", # Tested through the deployment-capability pre-call check tests in test_router.py
|
|
"_capability_model_and_provider", # Tested through the authenticating-provider pre-call check test in test_router.py
|
|
"_exact_base_model_max_input_tokens", # Tested through get_router_model_info in test_router_model_cost_isolation.py
|
|
"_base_model_widened_window", # Tested through get_router_model_info in test_router_model_cost_isolation.py
|
|
"_provider_model_id", # Tested through get_router_model_info in test_router_model_cost_isolation.py
|
|
"_names_route_model", # Tested through get_router_model_info in test_router_model_cost_isolation.py
|
|
"_declared_base_model", # Tested through get_router_model_info in test_router_model_cost_isolation.py
|
|
]
|
|
|
|
|
|
def main():
|
|
router_file = [
|
|
"./litellm/router.py",
|
|
"./litellm/router_utils/batch_utils.py",
|
|
"./litellm/router_utils/pattern_match_deployments.py",
|
|
]
|
|
# router_file = [
|
|
# "../../litellm/router.py",
|
|
# "../../litellm/router_utils/pattern_match_deployments.py",
|
|
# "../../litellm/router_utils/batch_utils.py",
|
|
# ] ## LOCAL TESTING
|
|
tests_dir = "./tests/" # Update this path if your tests directory is located elsewhere
|
|
# tests_dir = "../../tests/" # LOCAL TESTING
|
|
|
|
router_functions = []
|
|
for file in router_file:
|
|
router_functions.extend(get_functions_from_router(file))
|
|
print("router_functions: ", router_functions)
|
|
called_functions_in_tests = get_all_functions_called_in_tests(tests_dir)
|
|
untested_functions = [fn for fn in router_functions if fn not in called_functions_in_tests]
|
|
|
|
if untested_functions:
|
|
all_untested_functions = []
|
|
for func in untested_functions:
|
|
if func not in ignored_function_names:
|
|
all_untested_functions.append(func)
|
|
untested_perc = (len(all_untested_functions)) / len(router_functions)
|
|
print("untested_perc: ", untested_perc)
|
|
if untested_perc > 0:
|
|
print("The following functions in router.py are not tested:")
|
|
raise Exception(
|
|
f"{untested_perc * 100:.2f}% of functions in router.py are not tested: {all_untested_functions}"
|
|
)
|
|
else:
|
|
print("All functions in router.py are covered by tests.")
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|