Skip to content

feat(scan): support data-level stats pruning in TableScan #188

@QuakeWang

Description

@QuakeWang

Search before asking

  • I searched in the issues and found nothing similar.

Description

This is part of #173. The current scan path already supports partition pruning, but after that, TableScan still plans all candidate data files even when file-level stats already show that a filter cannot match.

This task is to add safe data-level stats pruning in the core scan layer based on DataFileMeta stats, so the planner can skip impossible files earlier.

Initial scope:

  • use file-level min/max values and null counts
  • support primitive columns
  • support =, !=, >, >=, <, <=, IS NULL, IS NOT NULL, and AND
  • fail open when stats are missing, unsupported, corrupted, or inconclusive

Willingness to contribute

  • I'm willing to submit a PR!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No fields configured for Task.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions