test: gate the test suite on F601, B023, B025 and F632

Four more ruff rules for code the test suite runs but never checks. F601 is the
one that paid: the duplicate key it flagged in a get_form_data fixture was the
mock reproducing the production bug fixed in the previous commit.

B025 removed two unreachable handlers, one of them a pytest.skip shadowed by an
earlier `pass`, so an upstream Vertex flake reported green having asserted
nothing. F632 turned an `is ""` identity check, which passes only on CPython
interning, into the `== ""` it meant. B023 fixed three closures over loop
variables, all latent today but one iteration-order change away from checking the
last case N times.

ruff-tests.toml

@@ -36,6 +36,17 @@
 #         `re.search`, so a `.` copied out of an error message is a wildcard and the block
 #         accepts messages the author never meant to accept. Mark a real regex raw, wrap a
 #         literal message in `re.escape`, and the pattern says which one it is
+# F601    the same key literal twice in one dict. Python keeps the last value, so the
+#         first is dropped before the test ever runs, and a fixture that looks like it
+#         covers two cases covers one
+# B023    a closure over a loop variable. Every closure sees the last iteration's value,
+#         so a per-case callback built in a loop checks the last case N times. Bind the
+#         value as a parameter instead
+# B025    an `except` for a type an earlier `except` already catches. The second handler
+#         is unreachable, so the recovery or skip written there never happens
+# F632    `is` against a literal. It compares identity, so it passes only where CPython
+#         happens to intern the value and stops meaning what it says the moment the
+#         value is built at runtime
 #
 # No target-version here on purpose: it resolves from requires-python (>=3.10), so
 # 3.11-only builtins like BaseExceptionGroup are correctly flagged in a tree that
@@ -58,4 +69,8 @@ lint.select = [
     "PLR0133",
     "PLW0127",
     "RUF043",
+    "F601",
+    "B023",
+    "B025",
+    "F632",
 ]

tests/code_coverage_tests/bedrock_pricing.py

@@ -95,7 +95,7 @@ def get_bedrock_pricing(url, providers):
         else:
             # General logic for other providers
             section = soup.find(
-                "h2", text=lambda t: t and provider.lower() in t.lower()
+                "h2", text=lambda t, needle=provider.lower(): t and needle in t.lower()
             )
             if not section:
                 pricing_data[provider] = "Provider section not found"

tests/load_tests/test_langsmith_load_test.py

@@ -66,11 +66,6 @@ def test_langsmith_logging_async():
     except Exception as e:
         pytest.fail(f"An exception occurred - {e}")
 
-    except litellm.Timeout as e:
-        pass
-    except Exception as e:
-        pytest.fail(f"An exception occurred - {e}")
-
 
 async def make_async_calls(metadata=None, **completion_kwargs):
     total_tasks = 300

tests/local_testing/test_amazing_vertex_completion.py

@@ -4207,13 +4207,7 @@ def test_gemini_google_maps_tool_simple():
             )
         print(f"Response: {response.model_dump_json(indent=4)}")
         assert response.choices[0].message.content is not None
-    except (litellm.RateLimitError, litellm.InternalServerError):
-        # Transient Vertex-side failures (rate limiting, 500 INTERNAL from the
-        # Google Maps grounding backend) are not LiteLLM bugs — don't fail CI.
-        pass
-    except litellm.InternalServerError:
-        pytest.skip(
-            "Google Maps Platform returned a transient 500 (upstream flake); skipping."
-        )
+    except (litellm.RateLimitError, litellm.InternalServerError) as e:
+        pytest.skip(f"Transient Vertex-side failure, not a LiteLLM bug: {e}")
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")

tests/logging_callback_tests/test_spend_logs.py

@@ -148,7 +148,6 @@ def test_spend_logs_payload(model_id: Optional[str]):
             "completion_start_time": datetime.datetime(2024, 6, 7, 12, 43, 30, 954146),
             "max_tokens": 10,
             "extra_body": {},
-            "custom_llm_provider": "azure",
             "input": [
                 {"role": "system", "content": "you are a helpful assistant.\n"},
                 {"role": "user", "content": "bom dia"},

tests/proxy_unit_tests/test_key_generate_prisma.py

@@ -3327,16 +3327,17 @@ async def test_team_access_groups(prisma_client):
 
     request._url = URL(url="/chat/completions")
 
+    def body_reader(requested_model: str):
+        async def return_body() -> bytes:
+            return f'{{"model": "{requested_model}"}}'.encode()
+
+        return return_body
+
     for model in ["gpt-4o", "gemini-pro-vision"]:
         # Expect these to pass
-        async def return_body():
-            return_string = f'{{"model": "{model}"}}'
-            # return string as bytes
-            return return_string.encode()
-
         request = Request(scope={"type": "http"})
         request._url = URL(url="/chat/completions")
-        request.body = return_body
+        request.body = body_reader(model)
 
         # use generated key to auth in
         print(
@@ -3346,14 +3347,9 @@ async def test_team_access_groups(prisma_client):
 
     for model in ["gpt-4", "gpt-4o-mini", "gemini-experimental"]:
         # Expect these to fail
-        async def return_body_2():
-            return_string = f'{{"model": "{model}"}}'
-            # return string as bytes
-            return return_string.encode()
-
         request = Request(scope={"type": "http"})
         request._url = URL(url="/chat/completions")
-        request.body = return_body_2
+        request.body = body_reader(model)
 
         # use generated key to auth in
         print(

tests/test_litellm/llms/ollama/test_ollama_completion_transformation.py

@@ -480,7 +480,7 @@ class TestOllamaTextCompletionResponseIterator:
         assert isinstance(result, ModelResponseStream)
         assert result.choices and result.choices[0].delta is not None
         assert result.choices[0].delta.content == None
-        assert getattr(result.choices[0].delta, "reasoning_content", None) is ""
+        assert getattr(result.choices[0].delta, "reasoning_content", None) == ""
 
     def test_chunk_parser_done_chunk(self):
         """Test that done chunks work correctly."""

tests/test_litellm/llms/watsonx/test_watsonx_common_utils.py

@@ -129,11 +129,7 @@ class TestGenerateIAMToken:
             mock_client.reset_mock()
             mock_cache.reset_mock()
 
-            # Configure mock to return values based on env_keys
-            def get_secret_side_effect(key):
-                return env_keys.get(key)
-
-            mock_get_secret_str.side_effect = get_secret_side_effect
+            mock_get_secret_str.side_effect = env_keys.get
 
             mock_response = MagicMock()
             mock_response.json.return_value = {

tests/test_litellm/test_utils.py

@@ -864,7 +864,6 @@ def test_aaamodel_prices_and_context_window_json_is_valid():
                 "input_cost_per_character_above_128k_tokens": {"type": "number"},
                 "input_cost_per_image": {"type": "number"},
                 "input_cost_per_image_above_128k_tokens": {"type": "number"},
-                "input_cost_per_image_token": {"type": "number"},
                 "input_cost_per_video_token": {"type": "number"},
                 "input_cost_per_token_above_200k_tokens": {"type": "number"},
                 "input_cost_per_token_above_256k_tokens": {"type": "number"},
@@ -1008,7 +1007,6 @@ def test_aaamodel_prices_and_context_window_json_is_valid():
                 },
                 "bedrock_converse_supports_strict_tools": {"type": "boolean"},
                 "tpm": {"type": "number"},
-                "provider_specific_entry": {"type": "object"},
                 "supported_endpoints": {
                     "type": "array",
                     "items": {