use of org.apache.hadoop.hive.serde2.objectinspector.primitive.DoubleObjectInspector in project hive by apache.
the class DataWritableWriter method createWriter.
/**
* Creates a writer for the specific object inspector. The returned writer will be used
* to call Parquet API for the specific data type.
* @param inspector The object inspector used to get the correct value type.
* @param type Type that contains information about the type schema.
* @return A ParquetWriter object used to call the Parquet API fo the specific data type.
*/
private DataWriter createWriter(ObjectInspector inspector, Type type) {
if (type.isPrimitive()) {
checkInspectorCategory(inspector, ObjectInspector.Category.PRIMITIVE);
PrimitiveObjectInspector primitiveInspector = (PrimitiveObjectInspector) inspector;
switch(primitiveInspector.getPrimitiveCategory()) {
case BOOLEAN:
return new BooleanDataWriter((BooleanObjectInspector) inspector);
case BYTE:
return new ByteDataWriter((ByteObjectInspector) inspector);
case SHORT:
return new ShortDataWriter((ShortObjectInspector) inspector);
case INT:
return new IntDataWriter((IntObjectInspector) inspector);
case LONG:
return new LongDataWriter((LongObjectInspector) inspector);
case FLOAT:
return new FloatDataWriter((FloatObjectInspector) inspector);
case DOUBLE:
return new DoubleDataWriter((DoubleObjectInspector) inspector);
case STRING:
return new StringDataWriter((StringObjectInspector) inspector);
case CHAR:
return new CharDataWriter((HiveCharObjectInspector) inspector);
case VARCHAR:
return new VarcharDataWriter((HiveVarcharObjectInspector) inspector);
case BINARY:
return new BinaryDataWriter((BinaryObjectInspector) inspector);
case TIMESTAMP:
return new TimestampDataWriter((TimestampObjectInspector) inspector);
case DECIMAL:
return new DecimalDataWriter((HiveDecimalObjectInspector) inspector);
case DATE:
return new DateDataWriter((DateObjectInspector) inspector);
default:
throw new IllegalArgumentException("Unsupported primitive data type: " + primitiveInspector.getPrimitiveCategory());
}
} else {
GroupType groupType = type.asGroupType();
OriginalType originalType = type.getOriginalType();
if (originalType != null && originalType.equals(OriginalType.LIST)) {
checkInspectorCategory(inspector, ObjectInspector.Category.LIST);
return new ListDataWriter((ListObjectInspector) inspector, groupType);
} else if (originalType != null && originalType.equals(OriginalType.MAP)) {
checkInspectorCategory(inspector, ObjectInspector.Category.MAP);
return new MapDataWriter((MapObjectInspector) inspector, groupType);
} else {
checkInspectorCategory(inspector, ObjectInspector.Category.STRUCT);
return new StructDataWriter((StructObjectInspector) inspector, groupType);
}
}
}
use of org.apache.hadoop.hive.serde2.objectinspector.primitive.DoubleObjectInspector in project hive by apache.
the class DruidSerDe method serialize.
@Override
public Writable serialize(Object o, ObjectInspector objectInspector) throws SerDeException {
if (objectInspector.getCategory() != ObjectInspector.Category.STRUCT) {
throw new SerDeException(getClass().toString() + " can only serialize struct types, but we got: " + objectInspector.getTypeName());
}
// Prepare the field ObjectInspectors
StructObjectInspector soi = (StructObjectInspector) objectInspector;
List<? extends StructField> fields = soi.getAllStructFieldRefs();
List<Object> values = soi.getStructFieldsDataAsList(o);
// We deserialize the result
final Map<String, Object> value = new HashMap<>();
for (int i = 0; i < columns.length; i++) {
if (values.get(i) == null) {
// null, we just add it
value.put(columns[i], null);
continue;
}
final Object res;
switch(types[i].getPrimitiveCategory()) {
case TIMESTAMP:
res = ((TimestampObjectInspector) fields.get(i).getFieldObjectInspector()).getPrimitiveJavaObject(values.get(i)).getTime();
break;
case TIMESTAMPLOCALTZ:
res = ((TimestampLocalTZObjectInspector) fields.get(i).getFieldObjectInspector()).getPrimitiveJavaObject(values.get(i)).getZonedDateTime().toInstant().toEpochMilli();
break;
case BYTE:
res = ((ByteObjectInspector) fields.get(i).getFieldObjectInspector()).get(values.get(i));
break;
case SHORT:
res = ((ShortObjectInspector) fields.get(i).getFieldObjectInspector()).get(values.get(i));
break;
case INT:
res = ((IntObjectInspector) fields.get(i).getFieldObjectInspector()).get(values.get(i));
break;
case LONG:
res = ((LongObjectInspector) fields.get(i).getFieldObjectInspector()).get(values.get(i));
break;
case FLOAT:
res = ((FloatObjectInspector) fields.get(i).getFieldObjectInspector()).get(values.get(i));
break;
case DOUBLE:
res = ((DoubleObjectInspector) fields.get(i).getFieldObjectInspector()).get(values.get(i));
break;
case DECIMAL:
res = ((HiveDecimalObjectInspector) fields.get(i).getFieldObjectInspector()).getPrimitiveJavaObject(values.get(i)).doubleValue();
break;
case CHAR:
res = ((HiveCharObjectInspector) fields.get(i).getFieldObjectInspector()).getPrimitiveJavaObject(values.get(i)).getValue();
break;
case VARCHAR:
res = ((HiveVarcharObjectInspector) fields.get(i).getFieldObjectInspector()).getPrimitiveJavaObject(values.get(i)).getValue();
break;
case STRING:
res = ((StringObjectInspector) fields.get(i).getFieldObjectInspector()).getPrimitiveJavaObject(values.get(i));
break;
case BOOLEAN:
res = ((BooleanObjectInspector) fields.get(i).getFieldObjectInspector()).get(values.get(i));
break;
default:
throw new SerDeException("Unknown type: " + types[i].getPrimitiveCategory());
}
value.put(columns[i], res);
}
// Extract the partitions keys segments granularity and partition key if any
// First Segment Granularity has to be here.
final int granularityFieldIndex = columns.length;
assert values.size() > granularityFieldIndex;
Preconditions.checkArgument(fields.get(granularityFieldIndex).getFieldName().equals(Constants.DRUID_TIMESTAMP_GRANULARITY_COL_NAME));
value.put(Constants.DRUID_TIMESTAMP_GRANULARITY_COL_NAME, ((TimestampObjectInspector) fields.get(granularityFieldIndex).getFieldObjectInspector()).getPrimitiveJavaObject(values.get(granularityFieldIndex)).getTime());
if (values.size() == columns.length + 2) {
// Then partition number if any.
final int partitionNumPos = granularityFieldIndex + 1;
Preconditions.checkArgument(fields.get(partitionNumPos).getFieldName().equals(Constants.DRUID_SHARD_KEY_COL_NAME), String.format("expecting to encounter %s but was %s", Constants.DRUID_SHARD_KEY_COL_NAME, fields.get(partitionNumPos).getFieldName()));
value.put(Constants.DRUID_SHARD_KEY_COL_NAME, ((LongObjectInspector) fields.get(partitionNumPos).getFieldObjectInspector()).get(values.get(partitionNumPos)));
}
return new DruidWritable(value);
}
use of org.apache.hadoop.hive.serde2.objectinspector.primitive.DoubleObjectInspector in project hive by apache.
the class ColumnStatisticsObjTranslator method unpackDoubleStats.
private static void unpackDoubleStats(ObjectInspector oi, Object o, String fName, ColumnStatisticsObj statsObj) throws UnsupportedDoubleException {
if (fName.equals("countnulls")) {
long v = ((LongObjectInspector) oi).get(o);
statsObj.getStatsData().getDoubleStats().setNumNulls(v);
} else if (fName.equals("numdistinctvalues")) {
long v = ((LongObjectInspector) oi).get(o);
statsObj.getStatsData().getDoubleStats().setNumDVs(v);
} else if (fName.equals("max")) {
double d = ((DoubleObjectInspector) oi).get(o);
if (Double.isInfinite(d) || Double.isNaN(d)) {
throw new UnsupportedDoubleException();
}
statsObj.getStatsData().getDoubleStats().setHighValue(d);
} else if (fName.equals("min")) {
double d = ((DoubleObjectInspector) oi).get(o);
if (Double.isInfinite(d) || Double.isNaN(d)) {
throw new UnsupportedDoubleException();
}
statsObj.getStatsData().getDoubleStats().setLowValue(d);
} else if (fName.equals("ndvbitvector")) {
PrimitiveObjectInspector poi = (PrimitiveObjectInspector) oi;
byte[] buf = ((BinaryObjectInspector) poi).getPrimitiveJavaObject(o);
statsObj.getStatsData().getDoubleStats().setBitVectors(buf);
;
}
}
use of org.apache.hadoop.hive.serde2.objectinspector.primitive.DoubleObjectInspector in project hive by apache.
the class GenericUDFFormatNumber method evaluate.
@Override
public Object evaluate(DeferredObject[] arguments) throws HiveException {
Object arg0;
Object arg1;
if ((arg0 = arguments[0].get()) == null || (arg1 = arguments[1].get()) == null) {
return null;
}
if (!dType.equals(PrimitiveCategory.STRING)) {
int dValue = ((IntObjectInspector) argumentOIs[1]).get(arg1);
if (dValue < 0) {
throw new HiveException("Argument 2 of function FORMAT_NUMBER must be >= 0, but \"" + dValue + "\" was found");
}
if (dValue != lastDValue) {
// construct a new DecimalFormat only if a new dValue
pattern.delete(0, pattern.length());
pattern.append("#,###,###,###,###,###,##0");
// decimal place
if (dValue > 0) {
pattern.append(".");
for (int i = 0; i < dValue; i++) {
pattern.append("0");
}
}
DecimalFormat dFormat = new DecimalFormat(pattern.toString());
lastDValue = dValue;
numberFormat.applyPattern(dFormat.toPattern());
}
}
double xDoubleValue = 0.0;
float xFloatValue = 0.0f;
HiveDecimal xDecimalValue = null;
int xIntValue = 0;
long xLongValue = 0L;
PrimitiveObjectInspector xObjectInspector = (PrimitiveObjectInspector) argumentOIs[0];
switch(xObjectInspector.getPrimitiveCategory()) {
case VOID:
case DOUBLE:
xDoubleValue = ((DoubleObjectInspector) argumentOIs[0]).get(arg0);
resultText.set(numberFormat.format(xDoubleValue));
break;
case FLOAT:
xFloatValue = ((FloatObjectInspector) argumentOIs[0]).get(arg0);
resultText.set(numberFormat.format(xFloatValue));
break;
case DECIMAL:
xDecimalValue = ((HiveDecimalObjectInspector) argumentOIs[0]).getPrimitiveJavaObject(arg0);
resultText.set(numberFormat.format(xDecimalValue.bigDecimalValue()));
break;
case BYTE:
case SHORT:
case INT:
xIntValue = ((IntObjectInspector) argumentOIs[0]).get(arg0);
resultText.set(numberFormat.format(xIntValue));
break;
case LONG:
xLongValue = ((LongObjectInspector) argumentOIs[0]).get(arg0);
resultText.set(numberFormat.format(xLongValue));
break;
default:
throw new HiveException("Argument 1 of function FORMAT_NUMBER must be " + serdeConstants.TINYINT_TYPE_NAME + "\"" + " or \"" + serdeConstants.SMALLINT_TYPE_NAME + "\"" + " or \"" + serdeConstants.INT_TYPE_NAME + "\"" + " or \"" + serdeConstants.BIGINT_TYPE_NAME + "\"" + " or \"" + serdeConstants.DOUBLE_TYPE_NAME + "\"" + " or \"" + serdeConstants.FLOAT_TYPE_NAME + "\"" + " or \"" + serdeConstants.DECIMAL_TYPE_NAME + "\", but \"" + argumentOIs[0].getTypeName() + "\" was found.");
}
return resultText;
}
use of org.apache.hadoop.hive.serde2.objectinspector.primitive.DoubleObjectInspector in project hive by apache.
the class BinarySortableSerDe method serialize.
static void serialize(ByteStream.Output buffer, Object o, ObjectInspector oi, boolean invert, byte nullMarker, byte notNullMarker) throws SerDeException {
// Is this field a null?
if (o == null) {
writeByte(buffer, nullMarker, invert);
return;
}
// This field is not a null.
writeByte(buffer, notNullMarker, invert);
switch(oi.getCategory()) {
case PRIMITIVE:
{
PrimitiveObjectInspector poi = (PrimitiveObjectInspector) oi;
switch(poi.getPrimitiveCategory()) {
case VOID:
{
return;
}
case BOOLEAN:
{
boolean v = ((BooleanObjectInspector) poi).get(o);
writeByte(buffer, (byte) (v ? 2 : 1), invert);
return;
}
case BYTE:
{
ByteObjectInspector boi = (ByteObjectInspector) poi;
byte v = boi.get(o);
writeByte(buffer, (byte) (v ^ 0x80), invert);
return;
}
case SHORT:
{
ShortObjectInspector spoi = (ShortObjectInspector) poi;
short v = spoi.get(o);
serializeShort(buffer, v, invert);
return;
}
case INT:
{
IntObjectInspector ioi = (IntObjectInspector) poi;
int v = ioi.get(o);
serializeInt(buffer, v, invert);
return;
}
case LONG:
{
LongObjectInspector loi = (LongObjectInspector) poi;
long v = loi.get(o);
serializeLong(buffer, v, invert);
return;
}
case FLOAT:
{
FloatObjectInspector foi = (FloatObjectInspector) poi;
serializeFloat(buffer, foi.get(o), invert);
return;
}
case DOUBLE:
{
DoubleObjectInspector doi = (DoubleObjectInspector) poi;
serializeDouble(buffer, doi.get(o), invert);
return;
}
case STRING:
{
StringObjectInspector soi = (StringObjectInspector) poi;
Text t = soi.getPrimitiveWritableObject(o);
serializeBytes(buffer, t.getBytes(), t.getLength(), invert);
return;
}
case CHAR:
{
HiveCharObjectInspector hcoi = (HiveCharObjectInspector) poi;
HiveCharWritable hc = hcoi.getPrimitiveWritableObject(o);
// Trailing space should ignored for char comparisons.
// So write stripped values for this SerDe.
Text t = hc.getStrippedValue();
serializeBytes(buffer, t.getBytes(), t.getLength(), invert);
return;
}
case VARCHAR:
{
HiveVarcharObjectInspector hcoi = (HiveVarcharObjectInspector) poi;
HiveVarcharWritable hc = hcoi.getPrimitiveWritableObject(o);
// use varchar's text field directly
Text t = hc.getTextValue();
serializeBytes(buffer, t.getBytes(), t.getLength(), invert);
return;
}
case BINARY:
{
BinaryObjectInspector baoi = (BinaryObjectInspector) poi;
BytesWritable ba = baoi.getPrimitiveWritableObject(o);
byte[] toSer = new byte[ba.getLength()];
System.arraycopy(ba.getBytes(), 0, toSer, 0, ba.getLength());
serializeBytes(buffer, toSer, ba.getLength(), invert);
return;
}
case DATE:
{
DateObjectInspector doi = (DateObjectInspector) poi;
int v = doi.getPrimitiveWritableObject(o).getDays();
serializeInt(buffer, v, invert);
return;
}
case TIMESTAMP:
{
TimestampObjectInspector toi = (TimestampObjectInspector) poi;
TimestampWritable t = toi.getPrimitiveWritableObject(o);
serializeTimestampWritable(buffer, t, invert);
return;
}
case TIMESTAMPLOCALTZ:
{
TimestampLocalTZObjectInspector toi = (TimestampLocalTZObjectInspector) poi;
TimestampLocalTZWritable t = toi.getPrimitiveWritableObject(o);
serializeTimestampTZWritable(buffer, t, invert);
return;
}
case INTERVAL_YEAR_MONTH:
{
HiveIntervalYearMonthObjectInspector ioi = (HiveIntervalYearMonthObjectInspector) poi;
HiveIntervalYearMonth intervalYearMonth = ioi.getPrimitiveJavaObject(o);
serializeHiveIntervalYearMonth(buffer, intervalYearMonth, invert);
return;
}
case INTERVAL_DAY_TIME:
{
HiveIntervalDayTimeObjectInspector ioi = (HiveIntervalDayTimeObjectInspector) poi;
HiveIntervalDayTime intervalDayTime = ioi.getPrimitiveJavaObject(o);
serializeHiveIntervalDayTime(buffer, intervalDayTime, invert);
return;
}
case DECIMAL:
{
HiveDecimalObjectInspector boi = (HiveDecimalObjectInspector) poi;
HiveDecimal dec = boi.getPrimitiveJavaObject(o);
serializeHiveDecimal(buffer, dec, invert);
return;
}
default:
{
throw new RuntimeException("Unrecognized type: " + poi.getPrimitiveCategory());
}
}
}
case LIST:
{
ListObjectInspector loi = (ListObjectInspector) oi;
ObjectInspector eoi = loi.getListElementObjectInspector();
// \1 followed by each element
int size = loi.getListLength(o);
for (int eid = 0; eid < size; eid++) {
writeByte(buffer, (byte) 1, invert);
serialize(buffer, loi.getListElement(o, eid), eoi, invert, nullMarker, notNullMarker);
}
// and \0 to terminate
writeByte(buffer, (byte) 0, invert);
return;
}
case MAP:
{
MapObjectInspector moi = (MapObjectInspector) oi;
ObjectInspector koi = moi.getMapKeyObjectInspector();
ObjectInspector voi = moi.getMapValueObjectInspector();
// \1 followed by each key and then each value
Map<?, ?> map = moi.getMap(o);
for (Map.Entry<?, ?> entry : map.entrySet()) {
writeByte(buffer, (byte) 1, invert);
serialize(buffer, entry.getKey(), koi, invert, nullMarker, notNullMarker);
serialize(buffer, entry.getValue(), voi, invert, nullMarker, notNullMarker);
}
// and \0 to terminate
writeByte(buffer, (byte) 0, invert);
return;
}
case STRUCT:
{
StructObjectInspector soi = (StructObjectInspector) oi;
List<? extends StructField> fields = soi.getAllStructFieldRefs();
for (int i = 0; i < fields.size(); i++) {
serialize(buffer, soi.getStructFieldData(o, fields.get(i)), fields.get(i).getFieldObjectInspector(), invert, nullMarker, notNullMarker);
}
return;
}
case UNION:
{
UnionObjectInspector uoi = (UnionObjectInspector) oi;
byte tag = uoi.getTag(o);
writeByte(buffer, tag, invert);
serialize(buffer, uoi.getField(o), uoi.getObjectInspectors().get(tag), invert, nullMarker, notNullMarker);
return;
}
default:
{
throw new RuntimeException("Unrecognized type: " + oi.getCategory());
}
}
}
Aggregations