NVIDIA / NVIDIA/cudf

[FEA] Propagate `validate_schema` from `register_io_source` into the `PythonScan` IR

Open
#23,043 0 comments 0 reactions 1 assignee Claimed by @Matt711 View on GitHub
cudf-polars feature request
Dominant language
C++
Stars
9.8k
Forks
1.1k
Avg merge
3d 6m
Merged PRs (30d)
278

Description

**Is your feature request related to a problem? Please describe.**

`register_io_source` accepts a `validate_schema` flag. When `False`, Polars skips checking that the source output matches the declared schema, which is useful for sources that deliberately emit a different dtype for coercion. This flag is not propagated into the `PythonScan` IR node, so cudf-polars always validates unconditionally. A source that uses `validate_schema=False` and intentionally emits a mismatched dtype will get a hard `SchemaError` on the GPU engine instead of the expected coercion behaviour.

**Describe the solution you'd like**

Expose `validate_schema` in the `PythonScan` IR so cudf-polars can honour the flag and match Polars' behaviour.

**Describe alternatives you've considered**

No workaround.

**Additional context**

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.