Apache Drill
  1. Apache Drill
  2. DRILL-303

RecordBatchLoader.load always creates new schema

    Details

    • Type: Bug Bug
    • Status: Resolved
    • Priority: Major Major
    • Resolution: Fixed
    • Affects Version/s: None
    • Fix Version/s: 0.4.0
    • Component/s: None
    • Labels:
      None

      Description

      Every single batch coming off the wire is treated as having a new schema, even if it isn't.

      The problem is that oldVectors is Map<MaterializedField,ValueVector>, but we are calling oldFields.remove(fieldDef), where fieldDef is a type FieldDef. Java doesn't do a check in this instance, and this call simply returns null.

      This bug actually masked another problem. If it did not return null, the returned vector is added to container, but it should be added to newVectors. We also don't load the vector with data from the incoming ByteBuf.

      1. DRILL-303.patch
        6 kB
        Steven Phillips

        Activity

        Hide
        Steven Phillips added a comment -
        Show
        Steven Phillips added a comment - Created reviewboard https://reviews.apache.org/r/15766/
        Hide
        Jacques Nadeau added a comment -

        Fixed in b12c0b1

        Show
        Jacques Nadeau added a comment - Fixed in b12c0b1

          People

          • Assignee:
            Steven Phillips
            Reporter:
            Steven Phillips
          • Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Development