Skip to content

fix: Enterprise WeChat docking sub application cannot output thinking process #2489

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
shaohuzhang1 merged 1 commit into from
Mar 4, 2025
Merged

fix: Enterprise WeChat docking sub application cannot output thinking process #2489

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions apps/application/flow/step_node/application_node/impl/base_application_node.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -115,6 +115,10 @@ def write_context(node_variable: Dict, workflow_variable: Dict, node: INode, wor
'prompt_tokens': response.get('prompt_tokens')}}
answer = response.get('content', '') or "抱歉，没有查找到相关内容，请重新描述您的问题或提供更多信息。"
reasoning_content = response.get('reasoning_content', '')
answer_list = response.get('answer_list', [])
node_variable['application_node_dict'] = {answer.get('real_node_id'): {**answer, 'index': index} for answer, index
in
zip(answer_list, range(len(answer_list)))}
_write_context(node_variable, workflow_variable, node, workflow, answer, reasoning_content)


Copy link
Copy Markdown
Contributor Author

@shaohuzhang1 shaohuzhang1 Mar 4, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Here are the identified issues and potential improvements:

Issues:

  1. Variable Access:

    • The line node_variable['application_node_dict'] is accessing a variable named response, which does not exist at that stage of the function execution. This will likely raise an error.

  2. Answer List Handling:

    • The line ${{answer.get('real_node_id') contains interpolation syntax, but it's missing the opening brace {.
    • The logic to update node_variable['application_node_dict'] seems incorrect because both keys ("real_node_id" and "index") should be included in each entry.

  3. Function Parameters:

    • The last parameter passed to _write_context() is workflow, but there is no such parameter defined in the original code snippet you provided.

Potential Improvements:

  1. Correct Variable Name:

    • Ensure that the response variable is correctly referenced if needed or removed if unnecessary.

  2. Fix Template Syntax:

    • Replace ${{answer.get('real_node_id')}} with simply answer.get('real_node_id').

  3. Re-evaluate Data Structure:

    • Review what data structure is intended for node_variable['application_node_dict']. If you mean to store different types of data under 'actual_node_id', adjust the dictionary accordingly.

  4. Parameter Correctness:

    • Double-check that all parameters passed to _write_context() match those expected, and add any necessary parameters that might have been omitted from the previous version.

  5. Error Handling:

    • Consider adding error handling to manage cases where certain fields like 'prompt_tokens', 'content', ..., ' might not always be available in the response.

Without additional context on how these variables interact, some further refinement may be required.

Expand Down
2 changes: 1 addition & 1 deletion apps/common/handle/impl/response/openai_to_response.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@ def to_block_response(self, chat_id, chat_record_id, content, is_end, completion
other_params = {}
data = ChatCompletion(id=chat_record_id, choices=[
BlockChoice(finish_reason='stop', index=0, chat_id=chat_id,
reasoning_content=other_params.get('reasoning_content', ""),
answer_list=other_params.get('answer_list', ""),
message=ChatCompletionMessage(role='assistant', content=content))],
created=datetime.datetime.now().second, model='', object='chat.completion',
usage=CompletionUsage(completion_tokens=completion_tokens,
Copy link
Copy Markdown
Contributor Author

@shaohuzhang1 shaohuzhang1 Mar 4, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The provided code snippet seems mostly correct, but there are a few minor adjustments and clarifications that can be made to ensure it works as intended:

  1. Use of get method correctly: In Python dictionaries, using the get() method with a default parameter (other_params.get('reasoning_content', "")) will handle cases where 'reasoning_content' is not present in other_params. This prevents a potential KeyError.

  2. Correct import statement for datetime module: Ensure you have imported the datetime module at the beginning of your file if it's not already imported. The current line imports from both pandas and pyarrow, so I assume datetime is used elsewhere; however, here we just need to make sure it's available.

  3. Consistency in handling None values: When accessing dictionary keys, always check for their existence before proceeding to access them to avoid NoneType errors.

Here’s the code with these points addressed:

import datetime

def to_block_response(self, chat_id, chat_record_id, content, is_end, completion_tokens, other_params):
    # Using get() to safely retrieve reasoning_content from other_params
    reasoning_content = other_params.get('reasoning_content', "")
    
    # Similarly, use get() to safely retrieve answer_list from other_params
    answer_list = other_params.get('answer_list', [])
    
    data = ChatCompletion(
        id=chat_record_id,
        choices=[
            BlockChoice(
                finish_reason='stop',
                index=0,
                chat_id=chat_id,
                reasoning_content=reasoning_content,
-                         answer_list answer_list,
                 message=ChatCompletionMessage(role='assistant', content=content)
            )
        ],
        created=datetime.datetime.now().timestamp(),
        model='',
        object='chat.completion',
        usage=CompletionUsage(completion_tokens=completion_tokens)
    )

# Example usage with an empty list for completeness
example_other_params = {}
to_block_response("id_123", "record_1", "Hello!", True, 20, example_other_params)

This should prevent any runtime errors related to missing keys in the other_params dictionary and maintain proper consistency in value fetching.

Expand Down