[BFCL] Fix silent drop of text responses when tool_calls is empty list - #1328
Open
shubhra wants to merge 1 commit into
Open
[BFCL] Fix silent drop of text responses when tool_calls is empty list#1328shubhra wants to merge 1 commit into
shubhra wants to merge 1 commit into
Conversation
When an OpenAI-compatible server (e.g. vLLM) returns `"tool_calls": []` instead of `"tool_calls": null` for a text-only response, `_parse_query_response_FC` iterated over the empty list successfully, producing `model_responses = []` and silently discarding the actual text in `message.content`. This broke all agentic and multi-turn tasks where the model needs to return a final text answer after using tools. Replace the fragile `try/except` (which relied on `TypeError` from iterating over `None`) with an explicit truthiness check. Both `None` and `[]` are falsy, so the `else` branch now correctly falls back to `message.content` in either case. A populated tool_calls list is truthy and is processed as before — no regressions for the normal path. Made-with: Cursor
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
When an OpenAI-compatible server (e.g. vLLM) returns
"tool_calls": []instead of"tool_calls": nullfor a text-only response,_parse_query_response_FCiterated over the empty list successfully, producingmodel_responses = []and silently discarding the actual text inmessage.content. This broke all agentic and multi-turn tasks where the model needs to return a final text answer after using tools.Replace the
try/except(which relied onTypeErrorfrom iterating overNone) with an explicit truthiness check. BothNoneand[]producefalse, so theelsebranch now correctly falls back tomessage.contentin either case. A populated tool_calls list is processed as before - no regressions for the normal path.