apache/iceberg · error · UnsupportedOperationException

Cannot convert type to SQL:

Error message

Cannot convert type to SQL: 

What it means

Spark3Util.DescribeExpressionVisitor.primitive maps Iceberg primitive types to SQL type strings for DESCRIBE output. The switch covers boolean, integer types, float/double, date, timestamp, string, uuid, fixed, binary and decimal; anything else (or an unexpected Type) throws UnsupportedOperationException 'Cannot convert type to SQL'.

Solutions

1. Check the schema type reported in the message; map it to a supported primitive (e.g. use string/uuid/fixed/binary/decimal equivalents)
2. Upgrade the Iceberg Spark module so its type-to-SQL mapping covers the type
3. Patch Spark3Util to add a case for the new type in the switch

Example fix

// before
throw new UnsupportedOperationException("Cannot convert type to SQL: " + primitive);
// after
case TIMESTAMP_NANO:
  return "timestamp_ntz"; // add case before the throw
// or
return primitive.toString(); // degrade instead of throwing

Types.TypeID id = ((Types.PrimitiveType) type).typeId();
Set<Types.TypeID> known = Set.of(BOOLEAN, INTEGER, LONG, FLOAT, DOUBLE, DATE,
    TIMESTAMP, STRING, UUID, FIXED, BINARY, DECIMAL);
if (!known.contains(id)) throw new IllegalArgumentException("Type not renderable as SQL: " + type);

boolean isRenderable(Types.Type t) {
  return t instanceof Types.PrimitiveType && !t.equals(Types.TimestampNanoType.get());
}

try {
  String sql = primitive(type);
} catch (UnsupportedOperationException e) {
  return type.toString(); // degrade to Iceberg type name in DESCRIBE output
}

When it happens

Trigger: DESCRIBE TABLE on an Iceberg table whose schema contains a primitive type not handled by the visitor's switch — typically a newer Iceberg type added after this Spark v4.0 helper was written, or a nested/type dispatch bug passing a non-primitive into primitive().

Common situations: Schema written by a newer Iceberg version read by an older Spark connector; custom Type implementations; directly invoking Spark3Util.describeSchema on hand-built schemas.

Understand the failure class

Background: UnsupportedOperationException and "is not supported" errors: when a library deliberately refuses a call — this error's family across 30 libraries.

Related errors

• Cannot convert Spark geography with mixed SRID to Iceberg
• Cannot convert Spark geography with mixed SRID to Iceberg
• Cannot convert Spark geometry with mixed SRID to Iceberg
• Cannot convert Spark geometry with mixed SRID to Iceberg
• Cannot convert type to SQL

AI-assisted analysis of apache/iceberg@86d9c8fc54 (2026-09-12). Data as JSON: /api/errors/8caf0e51834365a0. Report an issue: GitHub.

Appendix: source

Thrown at spark/v4.0/spark/src/main/java/org/apache/iceberg/spark/Spark3Util.java:614

        case DOUBLE:
          return "double";
        case DATE:
          return "date";
        case TIME:
          return "time";
        case TIMESTAMP:
          return "timestamp";
        case STRING:
        case UUID:
          return "string";
        case FIXED:
        case BINARY:
          return "binary";
        case DECIMAL:
          Types.DecimalType decimal = (Types.DecimalType) primitive;
          return "decimal(" + decimal.precision() + "," + decimal.scale() + ")";
      }
      throw new UnsupportedOperationException("Cannot convert type to SQL: " + primitive);
    }
  }

  private static class DescribeExpressionVisitor
      extends ExpressionVisitors.ExpressionVisitor<String> {
    private static final DescribeExpressionVisitor INSTANCE = new DescribeExpressionVisitor();

    private DescribeExpressionVisitor() {}

    @Override
    public String alwaysTrue() {
      return "true";
    }

    @Override
    public String alwaysFalse() {
      return "false";
    }

View on GitHub (pinned to 86d9c8fc54)