FFI: FFI_AggregateUDF silently drops producer overrides of defaulted trait methods #22331

Gap

FFI_AggregateUDF in datafusion/ffi/src/udaf/mod.rs does not plumb several defaulted methods of AggregateUDFImpl. Producer overrides are silently lost on the consumer side.

Missing methods

• display_name
• schema_name
• human_display
• window_function_schema_name
• window_function_display_name
• simplify
• simplify_expr_op_literal
• reverse_udf / reverse_expr
• is_descending
• value_from_stats
• default_value
• supports_null_handling_clause
• supports_within_group_clause
• set_monotonicity
• documentation

Why it matters

Severity: critical. value_from_stats enables statistics-driven shortcuts (e.g. MIN/MAX from precomputed stats) — silent loss forces full re-aggregation across the FFI boundary. default_value affects empty-group correctness. supports_null_handling_clause / supports_within_group_clause change accepted SQL surface area. simplify / simplify_expr_op_literal / reverse_* are optimizer hooks. SQL output naming wrong without display_name / schema_name.

Implementation notes

• Plumb each as a plain unsafe extern \"C\" fn; wrapper body calls the trait method on inner Arc<dyn AggregateUDFImpl> and dispatch handles override-or-default.
• Methods that ship Expr (simplify, simplify_expr_op_literal, reverse_expr) require the embedded FFI_LogicalExtensionCodec.
• Layout change → api change label, target main only, no back-port to branch-<major>.
• Add unit tests (local-bypass + mock_foreign_marker_id forced-foreign) and integration tests under datafusion/ffi/tests/ for any method shipping non-trivial FFI types.

Generated from datafusion-ffi skill audit. See .ai/skills/datafusion-ffi/SKILL.md §"Method coverage" and §"Known gaps to close" (originated in PR #22327). If a PR addressing this finds any item to be a false positive (e.g., a method intentionally omitted for a documented reason), please also propose an update to the datafusion-ffi skill so future audits do not re-flag it.