The scalar UDF typing currently binds the callable return type to pyarrow.DataType. That makes type checkers expect UDF implementations to return a data type object, even though scalar UDF callables return Arrow arrays containing values of the declared return type.
This shows up with the existing examples/python-udf.py pattern, where a function annotated as returning pa.Array is rejected by mypy.
What changes are included in this PR?
This PR updates the scalar UDF type hints in python/datafusion/user_defined.py so that:
the UDF callable return type is bounded to pa.Array
ScalarUDF.__init__ accepts a pa.Field for the resolved return field
the decorator helper accepts the public pa.DataType | pa.Field return-field input
Runtime behavior is unchanged.
Are these changes tested?
Yes. I ran the following local checks:
uv tool run ruff@0.15.1 check --config pyproject.toml python/datafusion/user_defined.py examples/python-udf.py
Result: All checks passed!
uv tool run ruff@0.15.1 format --check --config pyproject.toml python/datafusion/user_defined.py examples/python-udf.py
Result: 2 files already formatted
git diff --check
Result: passed with no whitespace errors
Are there any user-facing changes?
Yes, for static typing only. User-defined scalar functions that return Arrow arrays should type-check more accurately. There is no runtime API change.
LLM-generated code disclosure
This type-hint update was prepared with assistance from OpenAI Codex and manually reviewed before submission.
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Which issue does this PR close?
Closes #1516.
Rationale for this change
The scalar UDF typing currently binds the callable return type to pyarrow.DataType. That makes type checkers expect UDF implementations to return a data type object, even though scalar UDF callables return Arrow arrays containing values of the declared return type.
This shows up with the existing examples/python-udf.py pattern, where a function annotated as returning pa.Array is rejected by mypy.
What changes are included in this PR?
This PR updates the scalar UDF type hints in python/datafusion/user_defined.py so that:
Runtime behavior is unchanged.
Are these changes tested?
Yes. I ran the following local checks:
Are there any user-facing changes?
Yes, for static typing only. User-defined scalar functions that return Arrow arrays should type-check more accurately. There is no runtime API change.
LLM-generated code disclosure
This type-hint update was prepared with assistance from OpenAI Codex and manually reviewed before submission.