Skip to content

Commit

Permalink
Fixed alternating message role bug in Anthropic client (microsoft#2992)
Browse files Browse the repository at this point in the history 
* Fixed alternating message role bug

* Fix bug
  • Loading branch information
@marklysze
marklysze authored Jun 21, 2024
1 parent b483721 commit e3b2560
Showing 1 changed file with 84 additions and 66 deletions.
150 changes: 84 additions & 66 deletions autogen/oai/anthropic.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -242,86 +242,104 @@ def oai_messages_to_anthropic_messages(params: Dict[str, Any]) -> list[dict[str,

# Convert messages to Anthropic compliant format
processed_messages = []

# Used to interweave user messages to ensure user/assistant alternating
user_continue_message = {"content": "Please continue.", "role": "user"}
assistant_continue_message = {"content": "Please continue.", "role": "assistant"}

tool_use_messages = 0
tool_result_messages = 0
last_tool_use_index = -1
for message in params["messages"]:
if message["role"] == "system":
params["system"] = message["content"]
elif "tool_calls" in message:
# Map the tool call options to Anthropic's ToolUseBlock
tool_uses = []
tool_names = []
for tool_call in message["tool_calls"]:
tool_uses.append(
ToolUseBlock(
type="tool_use",
id=tool_call["id"],
name=tool_call["function"]["name"],
input=json.loads(tool_call["function"]["arguments"]),
else:
# New messages will be added here, manage role alternations
expected_role = "user" if len(processed_messages) % 2 == 0 else "assistant"

if "tool_calls" in message:
# Map the tool call options to Anthropic's ToolUseBlock
tool_uses = []
tool_names = []
for tool_call in message["tool_calls"]:
tool_uses.append(
ToolUseBlock(
type="tool_use",
id=tool_call["id"],
name=tool_call["function"]["name"],
input=json.loads(tool_call["function"]["arguments"]),
)
)
)
tool_names.append(tool_call["function"]["name"])

if has_tools:
processed_messages.append({"role": "assistant", "content": tool_uses})
tool_use_messages += 1
last_tool_use_index = len(processed_messages) - 1
else:
# Not using tools, so put in a plain text message
processed_messages.append(
{
"role": "assistant",
"content": f"Some internal function(s) that could be used: [{', '.join(tool_names)}]",
}
)
elif "tool_call_id" in message:
if has_tools:
# Map the tool usage call to tool_result for Anthropic
processed_messages.append(
{
"role": "user",
"content": [
{
"type": "tool_result",
"tool_use_id": message["tool_call_id"],
"content": message["content"],
}
],
}
)
tool_result_messages += 1
if has_tools:
tool_use_messages += 1
tool_names.append(tool_call["function"]["name"])

if expected_role == "user":
# Insert an extra user message as we will append an assistant message
processed_messages.append(user_continue_message)

if has_tools:
processed_messages.append({"role": "assistant", "content": tool_uses})
last_tool_use_index = len(processed_messages) - 1
else:
# Not using tools, so put in a plain text message
processed_messages.append(
{
"role": "assistant",
"content": f"Some internal function(s) that could be used: [{', '.join(tool_names)}]",
}
)
elif "tool_call_id" in message:

if expected_role == "assistant":
# Insert an extra assistant message as we will append a user message
processed_messages.append(assistant_continue_message)

if has_tools:
# Map the tool usage call to tool_result for Anthropic
processed_messages.append(
{
"role": "user",
"content": [
{
"type": "tool_result",
"tool_use_id": message["tool_call_id"],
"content": message["content"],
}
],
}
)
tool_result_messages += 1
else:
# Not using tools, so put in a plain text message
processed_messages.append(
{"role": "user", "content": f"Running the function returned: {message['content']}"}
)
elif message["content"] == "":
# Ignoring empty messages
pass
else:
# Not using tools, so put in a plain text message
processed_messages.append(
{"role": "user", "content": f"Running the function returned: {message['content']}"}
)
elif message["content"] == "":
message["content"] = (
"I'm done. Please send TERMINATE" # TODO: Determine why we would be getting a blank response. Typically this is because 'assistant' is the last message role.
)
processed_messages.append(message)
else:
processed_messages.append(message)
if expected_role != message["role"]:
# Inserting the alternating continue message
processed_messages.append(
user_continue_message if expected_role == "user" else assistant_continue_message
)

# We'll drop the last tool_use if there's no tool_result (occurs if we finish the conversation before running the function)
if tool_use_messages != tool_result_messages:
# Too many tool_use messages, drop the last one as we haven't run it.
processed_messages.pop(last_tool_use_index)
processed_messages.append(message)

# Check for interleaving roles and correct, for Anthropic must be: user, assistant, user, etc.
for i, message in enumerate(processed_messages):
if message["role"] is not ("user" if i % 2 == 0 else "assistant"):
message["role"] = "user" if i % 2 == 0 else "assistant"
# We'll replace the last tool_use if there's no tool_result (occurs if we finish the conversation before running the function)
if has_tools and tool_use_messages != tool_result_messages:
processed_messages[last_tool_use_index] = assistant_continue_message

# Also remove name key from message as it is not supported
message.pop("name", None)
# name is not a valid field on messages
for message in processed_messages:
if "name" in message:
message.pop("name", None)

# Note: When using reflection_with_llm we may end up with an "assistant" message as the last message and that may cause a blank response
# So, if the last role is not user, add a 'user' continue message at the end
if processed_messages[-1]["role"] != "user":
# If the last role is not user, add a continue message at the end
continue_message = {"content": "continue", "role": "user"}
processed_messages.append(continue_message)
processed_messages.append(user_continue_message)

return processed_messages

Expand Down

0 comments on commit e3b2560

Please sign in to comment.