/**
* Get a map of field names to default values for an Avro schema.
*
* @param avroRecordSchema The schema to get the map of field names to values.
* @return The map.
*/
public static Map<String, Object> getDefaultValueMap(Schema avroRecordSchema) {
List<Field> defaultFields = new ArrayList<Field>();
for (Field f : avroRecordSchema.getFields()) {
if (f.defaultValue() != null) {
// Need to create a new Field here or we will get
// org.apache.avro.AvroRuntimeException: Field already used:
// schemaVersion
defaultFields.add(new Field(f.name(), f.schema(), f.doc(), f.defaultValue(), f.order()));
}
}
Schema defaultSchema = Schema.createRecord(defaultFields);
Schema emptyRecordSchema = Schema.createRecord(new ArrayList<Field>());
DatumWriter<GenericRecord> writer = new GenericDatumWriter<GenericRecord>(emptyRecordSchema);
DatumReader<GenericRecord> reader =
new GenericDatumReader<GenericRecord>(emptyRecordSchema, defaultSchema);
GenericRecord emptyRecord = new GenericData.Record(emptyRecordSchema);
GenericRecord defaultRecord =
AvroUtils.readAvroEntity(AvroUtils.writeAvroEntity(emptyRecord, writer), reader);
Map<String, Object> defaultValueMap = new HashMap<String, Object>();
for (Field f : defaultFields) {
defaultValueMap.put(f.name(), defaultRecord.get(f.name()));
}
return defaultValueMap;
}
private static int hashCode(HashData data, Schema schema) throws IOException {
Decoder decoder = data.decoder;
switch (schema.getType()) {
case RECORD:
{
int hashCode = 1;
for (Field field : schema.getFields()) {
if (field.order() == Field.Order.IGNORE) {
GenericDatumReader.skip(field.schema(), decoder);
continue;
}
hashCode = hashCode * 31 + hashCode(data, field.schema());
}
return hashCode;
}
case ENUM:
case INT:
return decoder.readInt();
case FLOAT:
return Float.floatToIntBits(decoder.readFloat());
case LONG:
{
long l = decoder.readLong();
return (int) (l ^ (l >>> 32));
}
case DOUBLE:
{
long l = Double.doubleToLongBits(decoder.readDouble());
return (int) (l ^ (l >>> 32));
}
case ARRAY:
{
Schema elementType = schema.getElementType();
int hashCode = 1;
for (long l = decoder.readArrayStart(); l != 0; l = decoder.arrayNext())
for (long i = 0; i < l; i++) hashCode = hashCode * 31 + hashCode(data, elementType);
return hashCode;
}
case MAP:
throw new AvroRuntimeException("Can't hashCode maps!");
case UNION:
return hashCode(data, schema.getTypes().get(decoder.readInt()));
case FIXED:
return hashBytes(1, data, schema.getFixedSize(), false);
case STRING:
return hashBytes(0, data, decoder.readInt(), false);
case BYTES:
return hashBytes(1, data, decoder.readInt(), true);
case BOOLEAN:
return decoder.readBoolean() ? 1231 : 1237;
case NULL:
return 0;
default:
throw new AvroRuntimeException("Unexpected schema to hashCode!");
}
}
/**
* If equal, return the number of bytes consumed. If greater than, return GT, if less than, return
* LT.
*/
private static int compare(Decoders d, Schema schema) throws IOException {
Decoder d1 = d.d1;
Decoder d2 = d.d2;
switch (schema.getType()) {
case RECORD:
{
for (Field field : schema.getFields()) {
if (field.order() == Field.Order.IGNORE) {
GenericDatumReader.skip(field.schema(), d1);
GenericDatumReader.skip(field.schema(), d2);
continue;
}
int c = compare(d, field.schema());
if (c != 0) return (field.order() != Field.Order.DESCENDING) ? c : -c;
}
return 0;
}
case ENUM:
case INT:
{
int i1 = d1.readInt();
int i2 = d2.readInt();
return i1 == i2 ? 0 : (i1 > i2 ? 1 : -1);
}
case LONG:
{
long l1 = d1.readLong();
long l2 = d2.readLong();
return l1 == l2 ? 0 : (l1 > l2 ? 1 : -1);
}
case ARRAY:
{
long i = 0; // position in array
long r1 = 0, r2 = 0; // remaining in current block
long l1 = 0, l2 = 0; // total array length
while (true) {
if (r1 == 0) { // refill blocks(s)
r1 = d1.readLong();
if (r1 < 0) {
r1 = -r1;
d1.readLong();
}
l1 += r1;
}
if (r2 == 0) {
r2 = d2.readLong();
if (r2 < 0) {
r2 = -r2;
d2.readLong();
}
l2 += r2;
}
if (r1 == 0 || r2 == 0) // empty block: done
return (l1 == l2) ? 0 : ((l1 > l2) ? 1 : -1);
long l = Math.min(l1, l2);
while (i < l) { // compare to end of block
int c = compare(d, schema.getElementType());
if (c != 0) return c;
i++;
r1--;
r2--;
}
}
}
case MAP:
throw new AvroRuntimeException("Can't compare maps!");
case UNION:
{
int i1 = d1.readInt();
int i2 = d2.readInt();
if (i1 == i2) {
return compare(d, schema.getTypes().get(i1));
} else {
return i1 - i2;
}
}
case FIXED:
{
int size = schema.getFixedSize();
int c =
compareBytes(d.d1.getBuf(), d.d1.getPos(), size, d.d2.getBuf(), d.d2.getPos(), size);
d.d1.skipFixed(size);
d.d2.skipFixed(size);
return c;
}
case STRING:
case BYTES:
{
int l1 = d1.readInt();
int l2 = d2.readInt();
int c = compareBytes(d.d1.getBuf(), d.d1.getPos(), l1, d.d2.getBuf(), d.d2.getPos(), l2);
d.d1.skipFixed(l1);
d.d2.skipFixed(l2);
return c;
}
case FLOAT:
{
float f1 = d1.readFloat();
float f2 = d2.readFloat();
return (f1 == f2) ? 0 : ((f1 > f2) ? 1 : -1);
}
case DOUBLE:
{
double f1 = d1.readDouble();
double f2 = d2.readDouble();
return (f1 == f2) ? 0 : ((f1 > f2) ? 1 : -1);
}
case BOOLEAN:
boolean b1 = d1.readBoolean();
boolean b2 = d2.readBoolean();
return (b1 == b2) ? 0 : (b1 ? 1 : -1);
case NULL:
return 0;
default:
throw new AvroRuntimeException("Unexpected schema to compare!");
}
}