prestodb/presto · error · PrestoException

DECODER_CONVERSION_NOT_SUPPORTED

DECODER_CONVERSION_NOT_SUPPORTED

Error message

cannot decode object of '%s' as '%s' for column '%s'

What it means

Presto's Avro record decoder throws DECODER_CONVERSION_NOT_SUPPORTED when an Avro field value cannot be converted to the declared Presto column type. The getDouble() accessor only accepts Avro values that are Double or Float; anything else (e.g. String, Integer, GenericRecord) is rejected. This signals a mismatch between the Avro schema/actual record data and the Presto table column type mapping.

Source

Thrown at presto-record-decoder/src/main/java/com/facebook/presto/decoder/avro/AvroColumnDecoder.java:168

        {
            this.value = value;
            this.columnType = columnType;
            this.columnName = columnName;
        }

        @Override
        public boolean isNull()
        {
            return value == null;
        }

        @Override
        public double getDouble()
        {
            if (value instanceof Double || value instanceof Float) {
                return ((Number) value).doubleValue();
            }
            throw new PrestoException(DECODER_CONVERSION_NOT_SUPPORTED, format("cannot decode object of '%s' as '%s' for column '%s'", value.getClass(), columnType, columnName));
        }

        @Override
        public boolean getBoolean()
        {
            if (value instanceof Boolean) {
                return (Boolean) value;
            }
            throw new PrestoException(DECODER_CONVERSION_NOT_SUPPORTED, format("cannot decode object of '%s' as '%s' for column '%s'", value.getClass(), columnType, columnName));
        }

        @Override
        public long getLong()
        {
            if (value instanceof Long || value instanceof Integer) {
                return ((Number) value).longValue();
            }
            throw new PrestoException(DECODER_CONVERSION_NOT_SUPPORTED, format("cannot decode object of '%s' as '%s' for column '%s'", value.getClass(), columnType, columnName));

View on GitHub (pinned to 55bb57d202)

Solutions

1. Change the Presto column type (or the Avro field mapping) to match the actual Avro field type, e.g. use BIGINT/INTEGER for int/long fields
2. Ensure the correct writer/reader Avro schema is configured so the field deserializes as float/double
3. Add an explicit conversion in the decoder (custom decoder or view with CAST) instead of relying on implicit coercion

Example fix

// before (Presto column: DOUBLE, Avro field: "int")
CREATE TABLE t (metric DOUBLE) ...;
// after
CREATE TABLE t (metric BIGINT) ...; -- matches Avro int/long field

// Validate Avro schema field types against expected Presto column types before creating/decoding
Schema.Field field = schema.getField(columnName);
if (field.schema().getType() != Schema.Type.DOUBLE && field.schema().getType() != Schema.Type.FLOAT) {
    throw new IllegalArgumentException("Column '" + columnName + "' expects a double/float Avro field");
}

boolean isDecodableAsDouble(Object avroValue) {
    return avroValue instanceof Double || avroValue instanceof Float;
}

try {
    double d = block.getDouble(field);
} catch (PrestoException e) {
    if (e.getErrorCode().getName().contains("DECODER_CONVERSION_NOT_SUPPORTED")) {
        // fall back to reading as string / log schema mismatch
    } else {
        throw e;
    }
}

When it happens

Trigger: Decoding an Avro record whose field value is not Double/Float while the mapped column requires double (e.g. DOUBLE-typed column whose Avro field is a string or int).

Common situations: Table column declared DOUBLE in Presto but the Avro schema field is int/long/string; schema evolution changed the field type; wrong Avro schema supplied in the Kafka topic decoder config so values deserialize as Strings or GenericFixed.

Related errors

• ELASTICSEARCH_TYPE_MISMATCH
• DECODER_CONVERSION_NOT_SUPPORTED
• BIGQUERY_UNSUPPORTED_TYPE_FOR_VARBINARY
• Invalid array block:
• Invalid map block:

AI-assisted analysis of prestodb/presto@55bb57d202 (2026-09-04). Data as JSON: /api/errors/cd4f4e66c0bb6263. Report an issue: GitHub.