Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-36079

Null-based filter estimates should always be non-negative

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • 3.2.0
    • 3.2.0, 3.1.3, 3.0.4
    • SQL
    • None

    Description

      It's possible for a column's statistics to have a higher `nullCount` than the table's `rowCount`. In this case, the filter estimates come back outside of the reasonable range (between 0 and 1).

      Attachments

        Activity

          People

            karenfeng Karen Feng
            karenfeng Karen Feng
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: