[Evals] Fix function calling support (#19658)

Current implementation is overzealous in validating chat datasets

Fixes
[#langsmith-sdk:557](https://github.com/langchain-ai/langsmith-sdk/issues/557)

libs/langchain/langchain/smith/evaluation/runner_utils.py

@@ -280,7 +280,11 @@ def _get_prompt(inputs: Dict[str, Any]) -> str:
         )
 
 
-def _get_messages(inputs: Dict[str, Any]) -> List[BaseMessage]:
+class ChatModelInput(TypedDict):
+    messages: List[BaseMessage]
+
+
+def _get_messages(inputs: Dict[str, Any]) -> dict:
     """Get Chat Messages from inputs.
 
     Args:
@@ -293,35 +297,29 @@ def _get_messages(inputs: Dict[str, Any]) -> List[BaseMessage]:
     """
     if not inputs:
         raise InputFormatError("Inputs should not be empty.")
-
+    input_copy = inputs.copy()
     if "messages" in inputs:
-        single_input = inputs["messages"]
+        input_copy["input"] = input_copy.pop("messages")
     elif len(inputs) == 1:
-        single_input = next(iter(inputs.values()))
+        input_copy["input"] = next(iter(inputs.values()))
-    else:
+    if "input" in input_copy:
-        raise InputFormatError(
+        raw_messages = input_copy["input"]
-            f"Chat Run expects 'messages' in inputs when example has multiple"
+        if isinstance(raw_messages, list) and all(
-            f" input keys. Got {inputs}"
+            isinstance(i, dict) for i in raw_messages
-        )
+        ):
-    if isinstance(single_input, list) and all(
+            raw_messages = [raw_messages]
-        isinstance(i, dict) for i in single_input
+        if len(raw_messages) == 1:
-    ):
+            input_copy["input"] = messages_from_dict(raw_messages[0])
-        raw_messages = [single_input]
+        else:
-    elif isinstance(single_input, list) and all(
+            raise InputFormatError(
-        isinstance(i, list) for i in single_input
+                "Batch messages not supported. Please provide a"
-    ):
+                " single list of messages."
-        raw_messages = single_input
+            )
-    else:
+        return input_copy
-        raise InputFormatError(
-            f"Chat Run expects List[dict] or List[List[dict]] values for"
-            f" 'messages' key input. Got {inputs}"
-        )
-    if len(raw_messages) == 1:
-        return messages_from_dict(raw_messages[0])
     else:
         raise InputFormatError(
             f"Chat Run expects single List[dict] or List[List[dict]] 'messages'"
-            f" input. Got {len(raw_messages)} messages from inputs {inputs}"
+            f" input. Got {inputs}"
         )
 
 
@@ -711,9 +709,9 @@ async def _arun_llm(
                 ),
             )
         except InputFormatError:
-            messages = _get_messages(inputs)
+            llm_inputs = _get_messages(inputs)
             llm_output = await llm.ainvoke(
-                messages,
+                **llm_inputs,
                 config=RunnableConfig(
                     callbacks=callbacks, tags=tags or [], metadata=metadata or {}
                 ),
@@ -864,9 +862,9 @@ def _run_llm(
                 ),
             )
         except InputFormatError:
-            llm_messages = _get_messages(inputs)
+            llm_inputs = _get_messages(inputs)
             llm_output = llm.invoke(
-                llm_messages,
+                **llm_inputs,
                 config=RunnableConfig(callbacks=callbacks, metadata=metadata or {}),
             )
     return llm_output