scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-01 02:02:20 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 92

test: add ProviderConfigManager coverage for hosted_vllm branch

Browse source 
Exercises the real (unmocked) get_provider_anthropic_messages_config
with HOSTED_VLLM provider to cover the 5 lines that codecov/patch flagged.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
Denis Khachyan 2026-05-28 11:49:21 +03:00
parent cdcf64725b
commit 57085d958b
1 changed files with 69 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

69
tests/litellm/llms/hosted_vllm/test_anthropic_passthrough.py
View file

@ -305,3 +305,72 @@ class TestHandlerRouting:
)
mock_translate.assert_called_once()
# ---------------------------------------------------------------------------
# ProviderConfigManager integration — real (unmocked) path
# ---------------------------------------------------------------------------
class TestProviderConfigManager:
"""Exercises the real ProviderConfigManager.get_provider_anthropic_messages_config
to cover the hosted_vllm branch that CI codecov flags as uncovered."""
def test_hosted_vllm_with_flag_returns_config(self):
from litellm.utils import ProviderConfigManager
params = GenericLiteLLMParams(disable_anthropic_translation=True)
with patch.dict(os.environ, {}, clear=False):
os.environ.pop("DISABLE_HOSTED_VLLM_ANTHROPIC_TRANSLATION", None)
config = ProviderConfigManager.get_provider_anthropic_messages_config(
model="qwen36-27b",
provider=litellm.LlmProviders.HOSTED_VLLM,
litellm_params=params,
)
assert isinstance(config, HostedVLLMAnthropicMessagesConfig)
def test_hosted_vllm_without_flag_returns_none(self):
from litellm.utils import ProviderConfigManager
params = GenericLiteLLMParams()
with patch.dict(os.environ, {}, clear=False):
os.environ.pop("DISABLE_HOSTED_VLLM_ANTHROPIC_TRANSLATION", None)
config = ProviderConfigManager.get_provider_anthropic_messages_config(
model="qwen36-27b",
provider=litellm.LlmProviders.HOSTED_VLLM,
litellm_params=params,
)
assert config is None
def test_hosted_vllm_no_litellm_params_returns_none(self):
from litellm.utils import ProviderConfigManager
config = ProviderConfigManager.get_provider_anthropic_messages_config(
model="qwen36-27b",
provider=litellm.LlmProviders.HOSTED_VLLM,
litellm_params=None,
)
assert config is None
def test_hosted_vllm_env_var_returns_config(self):
from litellm.utils import ProviderConfigManager
params = GenericLiteLLMParams()
with patch.dict(os.environ, {"DISABLE_HOSTED_VLLM_ANTHROPIC_TRANSLATION": "true"}):
config = ProviderConfigManager.get_provider_anthropic_messages_config(
model="qwen36-27b",
provider=litellm.LlmProviders.HOSTED_VLLM,
litellm_params=params,
)
assert isinstance(config, HostedVLLMAnthropicMessagesConfig)
def test_non_hosted_vllm_delegates_to_cached(self):
from litellm.utils import ProviderConfigManager
config = ProviderConfigManager.get_provider_anthropic_messages_config(
model="claude-3",
provider=litellm.LlmProviders.ANTHROPIC,
litellm_params=None,
)
assert config is not None
assert not isinstance(config, HostedVLLMAnthropicMessagesConfig)