fix: stream option is not working with tracing enabled (#2602)

* fix: cannot create 'generator' instances

* fix: cannot create 'generator' instances

* [autofix.ci] apply automated fixes

---------

Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>

src/backend/base/langflow/schema/message.py

@@ -86,24 +86,21 @@ class Message(Data):
         # they are: "text", "sender"
         if self.text is None or not self.sender:
             logger.warning("Missing required keys ('text', 'sender') in Message, defaulting to HumanMessage.")
+        if not isinstance(self.text, str):
+            text = ""
+        else:
+            text = self.text
 
         if self.sender == "User" or not self.sender:
             if self.files:
-                contents = [{"type": "text", "text": self.text}]
+                contents = [{"type": "text", "text": text}]
                 contents.extend(self.get_file_content_dicts())
                 human_message = HumanMessage(content=contents)  # type: ignore
             else:
-                if not isinstance(self.text, str):
-                    text = ""
-                else:
-                    text = self.text
-                human_message = HumanMessage(
-                    content=text,
-                )
-
+                human_message = HumanMessage(content=text)
             return human_message
 
-        return AIMessage(content=self.text)  # type: ignore
+        return AIMessage(content=text)  # type: ignore
 
     @classmethod
     def from_data(cls, data: "Data") -> "Message":

src/backend/base/langflow/services/tracing/service.py

@@ -1,6 +1,7 @@
 import asyncio
 import os
 import traceback
+import types
 from collections import defaultdict
 from contextlib import asynccontextmanager
 from datetime import datetime, timezone
@@ -214,11 +215,8 @@ class LangSmithTracer(BaseTracer):
     ):
         if not self._ready:
             return
-        raw_inputs = {}
         processed_inputs = {}
         if inputs:
-            raw_inputs = inputs.copy()
-            raw_inputs |= metadata or {}
             processed_inputs = self._convert_to_langchain_types(inputs)
         child = self._run_tree.create_child(
             name=trace_name,
@@ -226,7 +224,7 @@ class LangSmithTracer(BaseTracer):
             inputs=processed_inputs,
         )
         if metadata:
-            child.add_metadata(metadata)
+            child.add_metadata(self._convert_to_langchain_types(metadata))
         self._children[trace_name] = child
         self._child_link: dict[str, str] = {}
 
@@ -254,6 +252,9 @@ class LangSmithTracer(BaseTracer):
                 value = value.to_lc_document()
         elif isinstance(value, Data):
             value = value.to_lc_document()
+        elif isinstance(value, types.GeneratorType):
+            # generator is not serializable, also we can't consume it
+            value = str(value)
         return value
 
     def end_trace(
@@ -270,8 +271,8 @@ class LangSmithTracer(BaseTracer):
             raw_outputs = outputs
             processed_outputs = self._convert_to_langchain_types(outputs)
         if logs:
-            child.add_metadata({"logs": {log.get("name"): log for log in logs}})
-        child.add_metadata({"outputs": raw_outputs})
+            child.add_metadata(self._convert_to_langchain_types({"logs": {log.get("name"): log for log in logs}}))
+        child.add_metadata(self._convert_to_langchain_types({"outputs": raw_outputs}))
         child.end(outputs=processed_outputs, error=error)
         if error:
             child.patch()

tests/conftest.py

@@ -73,13 +73,6 @@ def get_text():
         assert path.exists(), f"File {path} does not exist. Available files: {list(data_path.iterdir())}"
 
 
-@pytest.fixture(autouse=True)
-def check_openai_api_key_in_environment_variables():
-    import os
-
-    assert os.environ.get("OPENAI_API_KEY") is not None, "OPENAI_API_KEY is not set in environment variables"
-
-
 @pytest.fixture()
 async def async_client() -> AsyncGenerator:
     from langflow.main import create_app

tests/unit/test_messages.py

@@ -7,6 +7,7 @@ from langflow.schema.message import Message
 from langflow.services.database.models.message import MessageCreate, MessageRead
 from langflow.services.database.models.message.model import MessageTable
 from langflow.services.deps import session_scope
+from langflow.services.tracing.utils import convert_to_langchain_type
 
 
 @pytest.fixture()
@@ -74,3 +75,28 @@ def test_store_message():
     stored_messages = store_message(message)
     assert len(stored_messages) == 1
     assert stored_messages[0].text == "Stored message"
+
+
+@pytest.mark.parametrize("method_name", ["message", "convert_to_langchain_type"])
+def test_convert_to_langchain(method_name):
+    def convert(value):
+        if method_name == "message":
+            return value.to_lc_message()
+        elif method_name == "convert_to_langchain_type":
+            return convert_to_langchain_type(value)
+        else:
+            raise ValueError(f"Invalid method: {method_name}")
+
+    lc_message = convert(Message(text="Test message 1", sender="User", sender_name="User", session_id="session_id2"))
+    assert lc_message.content == "Test message 1"
+    assert lc_message.type == "human"
+
+    lc_message = convert(Message(text="Test message 2", sender="AI", session_id="session_id2"))
+    assert lc_message.content == "Test message 2"
+    assert lc_message.type == "ai"
+
+    iterator = iter(["stream", "message"])
+    lc_message = convert(Message(text=iterator, sender="AI", session_id="session_id2"))
+    assert lc_message.content == ""
+    assert lc_message.type == "ai"
+    assert len(list(iterator)) == 2