org.apache.iceberg.expressions.(Max/Min)Aggregate Don't handle null DataFile.(upper/lower)Bounds()

[jkolash]

Apache Iceberg version

1.5.0

Query engine

Spark

Please describe the bug 🐞

If you use the snapshot table command and run

select max(intValue) from table this can be reproduced.

The spec says this is optional
https://iceberg.apache.org/spec/#manifests

Let me know if you want me to provide a test case/fix as it seems like a pretty simple issue to test/fix for.

[jkolash]

relevant code
https://github.com/apache/iceberg/blob/main/api/src/main/java/org/apache/iceberg/expressions/MaxAggregate.java#L42

  @Override
  protected boolean hasValue(DataFile file) {
    boolean hasBound = file.upperBounds().containsKey(fieldId); //NPE here
    Long valueCount = safeGet(file.valueCounts(), fieldId);
    Long nullCount = safeGet(file.nullValueCounts(), fieldId);
    boolean boundAllNull =
        valueCount != null
            && valueCount > 0
            && nullCount != null
            && nullCount.longValue() == valueCount.longValue();
    return hasBound || boundAllNull;
  }

[amogh-jahagirdar]

Let me know if you want me to provide a test case/fix as it seems like a pretty simple issue to test/fix for.

@jkolash if you're up for it, feel free to raise a PR! I'm happy to review.