feat: the tools are handled using a prompt or the native API mechanism depending on the native_tool_call parameter

The older code path is in fact unneeded (once you remove the tools, the new one ) but it's simpler an more tested so I leave it how it was.

backend/open_webui/main.py

@@ -445,7 +445,7 @@ async def handle_streaming_response(request: Request, response: Response,
     async def stream_wrapper(original_generator, data_items):
         for item in data_items:
             yield wrap_item(json.dumps(item))
-        
+
         citations = []
         body = json.loads(request._body)
         generator = original_generator
@@ -498,7 +498,7 @@ async def handle_streaming_response(request: Request, response: Response,
 
                     if not tool_call["function"]["arguments"]:
                         tool_function_params = {}
-                    else: 
+                    else:
                         tool_function_params = json.loads(tool_call["function"]["arguments"]) 
 
                     log.debug(f"calling {tool_function_name} with params {tool_function_params}")
@@ -559,7 +559,6 @@ async def handle_nonstreaming_response(request: Request, response: Response,
         is_ollama = True
     is_openai = not is_ollama
 
-
     while (is_ollama and "tool_calls" in response_dict.get("message", {})) or \
           (is_openai and "tool_calls" in response_dict.get("choices", [{}])[0].get("message",{}) ):
         if is_ollama:
@@ -801,6 +800,7 @@ class ChatCompletionMiddleware(BaseHTTPMiddleware):
             "session_id": body.pop("session_id", None),
             "tool_ids": body.get("tool_ids", None),
             "files": body.get("files", None),
+            "native_tool_call": body.pop("native_tool_call", False),
         }
         body["metadata"] = metadata
 
@@ -840,6 +840,21 @@ class ChatCompletionMiddleware(BaseHTTPMiddleware):
 
         body, tools = get_tools_body(body, user, extra_params)
 
+        if model["owned_by"] == "ollama" and \
+                body["metadata"]["native_tool_call"] and \
+                body.get("stream", False):
+            log.info("Ollama models don't support function calling in streaming yet. forcing native_tool_call to False")
+            body["metadata"]["native_tool_call"] = False
+
+        if not body["metadata"]["native_tool_call"]:
+            del body["tools"] # we won't use those
+            try:
+                body, flags = await chat_completion_tools_handler(body, user, extra_params)
+                contexts.extend(flags.get("contexts", []))
+                citations.extend(flags.get("citations", []))
+            except Exception as e:
+                log.exception(e)
+
         try:
             body, flags = await chat_completion_files_handler(body)
             contexts.extend(flags.get("contexts", []))
@@ -884,12 +899,36 @@ class ChatCompletionMiddleware(BaseHTTPMiddleware):
             data_items.append({"citations": citations})
 
         update_body_request(request, body)
-        first_response = await call_next(request)
+        response = await call_next(request)
 
-        if body.get("stream", False) is False:
+        if not body["metadata"]["native_tool_call"]:
-            return await handle_nonstreaming_response(request, first_response, tools, user, data_items)
 
-        return await handle_streaming_response(request, first_response, tools, data_items, call_next, user)
+            if not isinstance(response, StreamingResponse):
+                return response
+            content_type = response.headers["Content-Type"]
+            is_openai = "text/event-stream" in content_type
+            is_ollama = "application/x-ndjson" in content_type
+            if not is_openai and not is_ollama:
+               return response
+
+            def wrap_item(item):
+                return f"data: {item}\n\n" if is_openai else f"{item}\n"
+
+            async def stream_wrapper(original_generator, data_items):
+                for item in data_items:
+                    yield wrap_item(json.dumps(item))
+
+                async for data in original_generator:
+                    yield data
+
+            return StreamingResponse(
+                stream_wrapper(response.body_iterator, data_items),
+                headers=dict(response.headers),
+            )
+        else:
+            if not body.get("stream", False):
+                return await handle_nonstreaming_response(request, response, tools, user, data_items)
+            return await handle_streaming_response(request, response, tools, data_items, call_next, user)
 
     async def _receive(self, body: bytes):
         return {"type": "http.request", "body": body, "more_body": False}