merge: base on litellm_integration_messages_folder

.github/scripts/assert_ci_coverage.py

@@ -516,7 +516,7 @@ def _integration_ownership(repo_root: pathlib.Path = REPO_ROOT) -> tuple[frozens
         str(path.relative_to(repo_root))
         for folders in groups.values()
         for folder in folders
-        for path in (integration_root / folder).glob("test_*.py")
+        for path in (integration_root / folder).rglob("test_*.py")
     )
     browser_manifest: Final = repo_root / "tests/e2e/ui/tests/integrationCritical/expected.json"
     browser_nodes: Final = json.loads(browser_manifest.read_text()) if browser_manifest.exists() else ()

tests/integration/README.md

@@ -28,6 +28,8 @@ Provider contracts exercise actual TCP requests with synthetic credentials and l
 
 Streaming checks send real HTTP transfer chunks, including one-byte partitions, fragmented tools, incomplete transfers and a cancellation barrier. They assert meaningful text, tool arguments, final usage and persisted cost. The Redis recovery case owns a separate database and Redis process, uses the supported one-second circuit-breaker recovery setting, waits for the real subscriber and verifies response data in Redis after restart. CircleCI reuses its existing Redis image for that extra process; it never pulls an image during tests
 
+The `messages_endpoint/` directory holds `/v1/messages` endpoint contracts: native-provider backends under `providers/` (`anthropic`, `bedrock`, `gemini`) and the translation bridges (`responses_bridge`, `chat_bridge`) at the top level. It runs in the providers shard; `run.py` selects test files recursively under each scheduled directory
+
 The sdk shard exercises the SDK's own HTTP clients against local protocol peers with no gateway in the path, so a case here fails only when the client library or its wire behavior changes. The HTTP/2 case runs a hypercorn TLS peer offering h2 and http/1.1 over ALPN, drives the sync and async httpx handlers at it with `LITELLM_HTTP2` off and on, and asserts the version both the client and the peer observed on the wire. Put a test here only when it needs no proxy, database or Redis; a case that reaches the gateway belongs in one of the other shards
 
 The extensions shard uses the built-in generic callback and guardrail transports. It checks callback correlation and credential exclusion, guardrail rewriting and denial, retained OpenAI consumers and A2A wire versions

tests/integration/_support/manifest.py

@@ -10,6 +10,7 @@ OWNED_DIRECTORIES: Final = frozenset(
         "routing",
         "providers",
         "streaming",
+        "messages_endpoint",
         "configuration",
         "mcp",
         "observability",

tests/integration/run.py

@@ -14,7 +14,7 @@ GROUPS: Final = MappingProxyType(
         "management": ("management", "authorization", "configuration"),
         "accounting": ("pricing", "spend"),
         "database": ("database",),
-        "providers": ("providers", "routing", "streaming"),
+        "providers": ("providers", "routing", "streaming", "messages_endpoint"),
         "extensions": ("observability", "compatibility"),
         "mcp": ("mcp",),
         "sdk": ("sdk",),
@@ -35,7 +35,7 @@ def main() -> int:
     selected: Final = tuple(
         str(path.relative_to(root))
         for folder in GROUPS[options.group]
-        for path in sorted((root / "tests/integration" / folder).glob("test_*.py"))
+        for path in sorted((root / "tests/integration" / folder).rglob("test_*.py"))
     )
     if not selected:
         parser.error(f"No integration test files selected for {options.group}")

tests/local_testing/test_get_llm_provider.py

@@ -133,46 +133,6 @@ def test_get_llm_provider_azure_o1():
     assert model == "o1-mini"
 
 
-def test_default_api_base():
-    from litellm.litellm_core_utils.get_llm_provider_logic import (
-        _get_openai_compatible_provider_info,
-    )
-    from litellm.types.utils import LlmProviders
-
-    # Patch environment variable to remove API base if it's set
-    with patch.dict(os.environ, {}, clear=True):
-        for provider in litellm.openai_compatible_providers:
-            # Get the API base for the given provider
-            if provider == "github_copilot":
-                continue
-            # Skip chatgpt as it requires OAuth authentication
-            if provider == "chatgpt":
-                continue
-            # Skip ragflow as it requires specific model format: ragflow/chat/{id}/{model} or ragflow/agent/{id}/{model}
-            if provider == "ragflow":
-                continue
-            _, _, _, api_base = _get_openai_compatible_provider_info(
-                model=f"{provider}/*", api_base=None, api_key=None, dynamic_api_key=None
-            )
-            if api_base is None:
-                continue
-
-            for other_provider in LlmProviders:
-                if other_provider.value != provider and provider != "{}_chat".format(
-                    other_provider.value
-                ):
-                    if provider == "codestral" and other_provider.value == "mistral":
-                        continue
-                    elif provider == "github" and other_provider.value == "azure":
-                        continue
-                    elif (
-                        provider in ("qwencloud", "qwen_ai_platform")
-                        and other_provider.value == "dashscope"
-                    ):
-                        continue
-                    assert other_provider.value not in api_base.replace("/openai", "")
-
-
 def test_hosted_vllm_default_api_key():
     from litellm.litellm_core_utils.get_llm_provider_logic import (
         _get_openai_compatible_provider_info,