[Bug]: Multiple Custom TransformMessages not chained

[WebsheetPlugin]

Describe the bug

transform_messages.TransformMessages(transforms=[RedactA(), RedactB()])

I would expect RedactB to be based on messages returned from RedactA. But this seems not to be the case.

Steps to reproduce

from autogen.agentchat.contrib.capabilities import transform_messages, transforms

class RedactA:
    def __init__(self):
        pass

    def apply_transform(self, messages: List[Dict]) -> List[Dict]:
        temp_messages = copy.deepcopy(messages)
        temp_messages[0]['content'] = "What is the capital of Poland?"
        return temp_messages
    
    def get_logs(self, pre_transform_messages: List[Dict], post_transform_messages: List[Dict]) -> Tuple[str, bool]:
        return "transormed emails", True

class RedactB:
    def __init__(self):
        pass

    def apply_transform(self, messages: List[Dict]) -> List[Dict]:
        temp_messages = copy.deepcopy(messages)
        temp_messages[0]['content'].replace("Poland", "France")
        return temp_messages
    
    def get_logs(self, pre_transform_messages: List[Dict], post_transform_messages: List[Dict]) -> Tuple[str, bool]:
        return "transormed picker", True

REDACT_PICKER_HANDLING = transform_messages.TransformMessages(transforms=[RedactA(), RedactB()])


TWO_REDACT = transform_messages.TransformMessages(transforms=[RedactA(), RedactB()])

---

Create two custom redact functions according to the tutorial for TransformMessages and apply them like the above. After asking a user question I would expect to get from the chat answer Paris, but because they seem to be not chained it returns Warsaw.

Model Used

No response

Expected Behavior

No response

Screenshots and logs

No response

Additional Information

v27.

[WebsheetPlugin]

Sorry, my mistake. All seems to work.

[WaelKarkoub]

@WebsheetPlugin that's actually a great test to add, thank you :)