Skip to content

Fallback Hudi incremental queries from native scan conversion #2273

@weimingdiit

Description

@weimingdiit

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

Native Hudi scan conversion currently handles time travel fallback explicitly, but Hudi incremental queries also depend on Hudi timeline and instant filtering semantics. A plain native Parquet/ORC file scan should not take over these queries unless incremental semantics are explicitly supported.

Describe the solution you'd like

Detect Hudi incremental query options during native scan support checks and fallback to Spark/Hudi scan when they are present.

The fallback should cover:

  • hoodie.datasource.query.type=incremental
  • hoodie.datasource.read.begin.instanttime
  • hoodie.datasource.read.end.instanttime

The option matching should remain case-insensitive.

Describe alternatives you've considered

Implementing native incremental scan support directly would require modeling Hudi timeline and instant filtering semantics in Auron. This is larger in scope and should be handled separately.

Additional context

This change is a safety guard. Snapshot and read-optimized paths keep their existing behavior.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions