Merge pull request #502 from zxrys/main

fix: OpenAI chat.completions.create API function calling error

runtime/node/agent/providers/openai_provider.py

@@ -1,6 +1,7 @@
 """OpenAI provider implementation."""
 
 import base64
+import hashlib
 
 import binascii
 import os
@@ -64,7 +65,7 @@ class OpenAIProvider(ModelProvider):
         is_chat = self._is_chat_completions_mode(client)
         
         if is_chat:
-            request_payload = self._build_chat_payload(timeline, tool_specs, kwargs)
+            request_payload = self._build_chat_payload(conversation, tool_specs, kwargs)
             response = client.chat.completions.create(**request_payload)
             self._track_token_usage(response)
             self._append_chat_response_output(timeline, response)
@@ -80,7 +81,7 @@ class OpenAIProvider(ModelProvider):
             message = self._deserialize_response(response)
             return ModelResponse(message=message, raw_response=response)
         except Exception as e:
-            new_request_payload = self._build_chat_payload(timeline, tool_specs, kwargs)
+            new_request_payload = self._build_chat_payload(conversation, tool_specs, kwargs)
             response = client.chat.completions.create(**new_request_payload)
             self._track_token_usage(response)
             self._append_chat_response_output(timeline, response)
@@ -226,7 +227,7 @@ class OpenAIProvider(ModelProvider):
 
     def _build_chat_payload(
         self,
-        timeline: List[Any],
+        conversation: List[Message],
         tool_specs: Optional[List[ToolSpec]],
         raw_params: Dict[str, Any],
     ) -> Dict[str, Any]:
@@ -238,8 +239,8 @@ class OpenAIProvider(ModelProvider):
             max_tokens = max_output_tokens
 
         messages: List[Any] = []
-        for item in timeline:
-            serialized = self._serialize_timeline_item_for_chat(item)
+        for item in conversation:
+            serialized = self._serialize_message_for_chat(item)
             if serialized is not None:
                 messages.append(serialized)
 
@@ -366,12 +367,19 @@ class OpenAIProvider(ModelProvider):
         tool_calls: List[ToolCallPayload] = []
         tc_data = self._get_attr(msg, "tool_calls")
         if tc_data:
-            for tc in tc_data:
+            for idx, tc in enumerate(tc_data):
                 f_data = self._get_attr(tc, "function") or {}
+                function_name = self._get_attr(f_data, "name") or ""
+                arguments = self._get_attr(f_data, "arguments") or ""
+                if not isinstance(arguments, str):
+                    arguments = str(arguments)
+                call_id = self._get_attr(tc, "id")
+                if not call_id:
+                    call_id = self._build_tool_call_id(function_name, arguments, fallback_prefix=f"tool_call_{idx}")
                 tool_calls.append(ToolCallPayload(
-                    id=self._get_attr(tc, "id"),
-                    function_name=self._get_attr(f_data, "name"),
-                    arguments=self._get_attr(f_data, "arguments"),
+                    id=call_id,
+                    function_name=function_name,
+                    arguments=arguments,
                     type="function"
                 ))
         
@@ -391,13 +399,18 @@ class OpenAIProvider(ModelProvider):
 
         if getattr(msg, "tool_calls", None):
             assistant_msg["tool_calls"] = []
-            for tc in msg.tool_calls:
+            for idx, tc in enumerate(msg.tool_calls):
+                function_name = tc.function.name
+                arguments = tc.function.arguments or ""
+                if not isinstance(arguments, str):
+                    arguments = str(arguments)
+                call_id = tc.id or self._build_tool_call_id(function_name, arguments, fallback_prefix=f"tool_call_{idx}")
                 assistant_msg["tool_calls"].append({
-                    "id": tc.id,
+                    "id": call_id,
                     "type": "function",
                     "function": {
-                        "name": tc.function.name,
-                        "arguments": tc.function.arguments,
+                        "name": function_name,
+                        "arguments": arguments,
                     },
                 })
 
@@ -691,7 +704,6 @@ class OpenAIProvider(ModelProvider):
         )
 
     def _parse_tool_call(self, payload: Any) -> Optional[ToolCallPayload]:
-        call_id = self._get_attr(payload, "call_id") or self._get_attr(payload, "id") or ""
         function_payload = self._get_attr(payload, "function") or {}
         function_name = self._get_attr(function_payload, "name") or self._get_attr(payload, "name") or ""
         arguments = self._get_attr(function_payload, "arguments") or self._get_attr(payload, "arguments") or ""
@@ -706,6 +718,9 @@ class OpenAIProvider(ModelProvider):
                 arguments_str = str(arguments)
         else:
             arguments_str = str(arguments)
+        call_id = self._get_attr(payload, "call_id") or self._get_attr(payload, "id") or ""
+        if not call_id:
+            call_id = self._build_tool_call_id(function_name, arguments_str)
         return ToolCallPayload(
             id=call_id,
             function_name=function_name,
@@ -713,6 +728,12 @@ class OpenAIProvider(ModelProvider):
             type="function",
         )
 
+    def _build_tool_call_id(self, function_name: str, arguments: str, *, fallback_prefix: str = "tool_call") -> str:
+        base = function_name or fallback_prefix
+        payload = f"{base}:{arguments or ''}".encode("utf-8")
+        digest = hashlib.md5(payload).hexdigest()[:8]
+        return f"{base}_{digest}"
+
     def _get_attr(self, payload: Any, key: str) -> Any:
         if hasattr(payload, key):
             return getattr(payload, key)