Add response_schema parameter to Docs's aquery response generation

[HANJionghao]

Corresponds to #700, #446

Note

Add optional response_schema to Docs.aquery and pass it to llm_model.call_single as response_format for answer generation.

^{Written by Cursor Bugbot for commit de6751f. Configure here.}

[dosubot]

Related Documentation

Checked 1 published document(s) in 1 knowledge base(s). No updates required.

^{How did I do? Any feedback?}  

[cursor]

This PR is being reviewed by Cursor Bugbot

Details

Your team is on the Bugbot Free tier. On this plan, Bugbot will review limited PRs each billing cycle for each member of your team.

To receive Bugbot reviews on all of your PRs, visit the Cursor dashboard to activate Pro and start your 14-day free trial.

Bug: Synchronous method lacks schema feature parity.

The synchronous query method lacks the new response_schema parameter that was added to aquery, breaking the wrapper pattern where both methods should accept identical parameters. This prevents users of the synchronous method from using structured response schemas and causes a parameter mismatch between the wrapper and the underlying async implementation.

src/paperqa/docs.py#L744-L769

paper-qa/src/paperqa/docs.py

Lines 744 to 769 in de6751f

	def query(
	self,
	query: PQASession | str,
	settings: MaybeSettings = None,
	callbacks: Sequence[Callable] | None = None,
	llm_model: LLMModel | None = None,
	summary_llm_model: LLMModel | None = None,
	embedding_model: EmbeddingModel | None = None,
	partitioning_fn: Callable[[Embeddable], int] | None = None,
	) -> PQASession:
	warnings.warn(
	"The synchronous `query` method is being deprecated in favor of the"
	" asynchronous `aquery` method, this deprecation will conclude in"
	" version 6.",
	category=DeprecationWarning,
	stacklevel=2,
	)
	return get_loop().run_until_complete(
	self.aquery(
	query,
	settings=settings,
	callbacks=callbacks,
	llm_model=llm_model,
	summary_llm_model=summary_llm_model,
	embedding_model=embedding_model,

 

---

Comment @cursor review or bugbot run to trigger another review on this PR

[jamesbraza]

I think this should be output_type: type[BaseModel] | TypeAdapter | JSONSchema | None = None, are you sure it works?

Can you add or expand a unit test in tests/test_paperqa.py confirming this?