Search in sources :

Example 21 with TimestampObjectInspector

use of org.apache.hadoop.hive.serde2.objectinspector.primitive.TimestampObjectInspector in project hive by apache.

the class AvroSerializer method serializePrimitive.

private Object serializePrimitive(TypeInfo typeInfo, PrimitiveObjectInspector fieldOI, Object structFieldData, Schema schema) throws AvroSerdeException {
    switch(fieldOI.getPrimitiveCategory()) {
        case BINARY:
            if (schema.getType() == Type.BYTES) {
                return AvroSerdeUtils.getBufferFromBytes((byte[]) fieldOI.getPrimitiveJavaObject(structFieldData));
            } else if (schema.getType() == Type.FIXED) {
                Fixed fixed = new GenericData.Fixed(schema, (byte[]) fieldOI.getPrimitiveJavaObject(structFieldData));
                return fixed;
            } else {
                throw new AvroSerdeException("Unexpected Avro schema for Binary TypeInfo: " + schema.getType());
            }
        case DECIMAL:
            HiveDecimal dec = (HiveDecimal) fieldOI.getPrimitiveJavaObject(structFieldData);
            return AvroSerdeUtils.getBufferFromDecimal(dec, ((DecimalTypeInfo) typeInfo).scale());
        case CHAR:
            HiveChar ch = (HiveChar) fieldOI.getPrimitiveJavaObject(structFieldData);
            return ch.getStrippedValue();
        case VARCHAR:
            HiveVarchar vc = (HiveVarchar) fieldOI.getPrimitiveJavaObject(structFieldData);
            return vc.getValue();
        case DATE:
            Date date = ((DateObjectInspector) fieldOI).getPrimitiveJavaObject(structFieldData);
            return defaultProleptic ? date.toEpochDay() : CalendarUtils.convertDateToHybrid(date.toEpochDay());
        case TIMESTAMP:
            Timestamp timestamp = ((TimestampObjectInspector) fieldOI).getPrimitiveJavaObject(structFieldData);
            long millis = defaultProleptic ? timestamp.toEpochMilli() : CalendarUtils.convertTimeToHybrid(timestamp.toEpochMilli());
            timestamp = TimestampTZUtil.convertTimestampToZone(Timestamp.ofEpochMilli(millis), TimeZone.getDefault().toZoneId(), ZoneOffset.UTC, legacyConversion);
            return timestamp.toEpochMilli();
        case UNKNOWN:
            throw new AvroSerdeException("Received UNKNOWN primitive category.");
        case VOID:
            return null;
        default:
            // All other primitive types are simple
            return fieldOI.getPrimitiveJavaObject(structFieldData);
    }
}
Also used : Fixed(org.apache.avro.generic.GenericData.Fixed) DateObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.DateObjectInspector) TimestampObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.TimestampObjectInspector) HiveDecimal(org.apache.hadoop.hive.common.type.HiveDecimal) HiveChar(org.apache.hadoop.hive.common.type.HiveChar) HiveVarchar(org.apache.hadoop.hive.common.type.HiveVarchar) GenericData(org.apache.avro.generic.GenericData) Timestamp(org.apache.hadoop.hive.common.type.Timestamp) Date(org.apache.hadoop.hive.common.type.Date) Fixed(org.apache.avro.generic.GenericData.Fixed)

Example 22 with TimestampObjectInspector

use of org.apache.hadoop.hive.serde2.objectinspector.primitive.TimestampObjectInspector in project hive by apache.

the class TestOrcFile method testReadFormat_0_11.

@Test
public void testReadFormat_0_11() throws Exception {
    Path oldFilePath = new Path(HiveTestUtils.getFileFromClasspath("orc-file-11-format.orc"));
    Reader reader = OrcFile.createReader(oldFilePath, OrcFile.readerOptions(conf).filesystem(fs));
    int stripeCount = 0;
    int rowCount = 0;
    long currentOffset = -1;
    for (StripeInformation stripe : reader.getStripes()) {
        stripeCount += 1;
        rowCount += stripe.getNumberOfRows();
        if (currentOffset < 0) {
            currentOffset = stripe.getOffset() + stripe.getIndexLength() + stripe.getDataLength() + stripe.getFooterLength();
        } else {
            assertEquals(currentOffset, stripe.getOffset());
            currentOffset += stripe.getIndexLength() + stripe.getDataLength() + stripe.getFooterLength();
        }
    }
    assertEquals(reader.getNumberOfRows(), rowCount);
    assertEquals(2, stripeCount);
    // check the stats
    ColumnStatistics[] stats = reader.getStatistics();
    assertEquals(7500, stats[1].getNumberOfValues());
    assertEquals(3750, ((BooleanColumnStatistics) stats[1]).getFalseCount());
    assertEquals(3750, ((BooleanColumnStatistics) stats[1]).getTrueCount());
    assertEquals("count: 7500 hasNull: true true: 3750", stats[1].toString());
    assertEquals(2048, ((IntegerColumnStatistics) stats[3]).getMaximum());
    assertEquals(1024, ((IntegerColumnStatistics) stats[3]).getMinimum());
    assertEquals(true, ((IntegerColumnStatistics) stats[3]).isSumDefined());
    assertEquals(11520000, ((IntegerColumnStatistics) stats[3]).getSum());
    assertEquals("count: 7500 hasNull: true min: 1024 max: 2048 sum: 11520000", stats[3].toString());
    assertEquals(Long.MAX_VALUE, ((IntegerColumnStatistics) stats[5]).getMaximum());
    assertEquals(Long.MAX_VALUE, ((IntegerColumnStatistics) stats[5]).getMinimum());
    assertEquals(false, ((IntegerColumnStatistics) stats[5]).isSumDefined());
    assertEquals("count: 7500 hasNull: true min: 9223372036854775807 max: 9223372036854775807", stats[5].toString());
    assertEquals(-15.0, ((DoubleColumnStatistics) stats[7]).getMinimum());
    assertEquals(-5.0, ((DoubleColumnStatistics) stats[7]).getMaximum());
    assertEquals(-75000.0, ((DoubleColumnStatistics) stats[7]).getSum(), 0.00001);
    assertEquals("count: 7500 hasNull: true min: -15.0 max: -5.0 sum: -75000.0", stats[7].toString());
    assertEquals("count: 7500 hasNull: true min: bye max: hi sum: 0", stats[9].toString());
    // check the inspectors
    StructObjectInspector readerInspector = (StructObjectInspector) reader.getObjectInspector();
    assertEquals(ObjectInspector.Category.STRUCT, readerInspector.getCategory());
    assertEquals("struct<boolean1:boolean,byte1:tinyint,short1:smallint," + "int1:int,long1:bigint,float1:float,double1:double,bytes1:" + "binary,string1:string,middle:struct<list:array<struct<int1:int," + "string1:string>>>,list:array<struct<int1:int,string1:string>>," + "map:map<string,struct<int1:int,string1:string>>,ts:timestamp," + "decimal1:decimal(38,10)>", readerInspector.getTypeName());
    List<? extends StructField> fields = readerInspector.getAllStructFieldRefs();
    BooleanObjectInspector bo = (BooleanObjectInspector) readerInspector.getStructFieldRef("boolean1").getFieldObjectInspector();
    ByteObjectInspector by = (ByteObjectInspector) readerInspector.getStructFieldRef("byte1").getFieldObjectInspector();
    ShortObjectInspector sh = (ShortObjectInspector) readerInspector.getStructFieldRef("short1").getFieldObjectInspector();
    IntObjectInspector in = (IntObjectInspector) readerInspector.getStructFieldRef("int1").getFieldObjectInspector();
    LongObjectInspector lo = (LongObjectInspector) readerInspector.getStructFieldRef("long1").getFieldObjectInspector();
    FloatObjectInspector fl = (FloatObjectInspector) readerInspector.getStructFieldRef("float1").getFieldObjectInspector();
    DoubleObjectInspector dbl = (DoubleObjectInspector) readerInspector.getStructFieldRef("double1").getFieldObjectInspector();
    BinaryObjectInspector bi = (BinaryObjectInspector) readerInspector.getStructFieldRef("bytes1").getFieldObjectInspector();
    StringObjectInspector st = (StringObjectInspector) readerInspector.getStructFieldRef("string1").getFieldObjectInspector();
    StructObjectInspector mid = (StructObjectInspector) readerInspector.getStructFieldRef("middle").getFieldObjectInspector();
    List<? extends StructField> midFields = mid.getAllStructFieldRefs();
    ListObjectInspector midli = (ListObjectInspector) midFields.get(0).getFieldObjectInspector();
    StructObjectInspector inner = (StructObjectInspector) midli.getListElementObjectInspector();
    List<? extends StructField> inFields = inner.getAllStructFieldRefs();
    ListObjectInspector li = (ListObjectInspector) readerInspector.getStructFieldRef("list").getFieldObjectInspector();
    MapObjectInspector ma = (MapObjectInspector) readerInspector.getStructFieldRef("map").getFieldObjectInspector();
    TimestampObjectInspector tso = (TimestampObjectInspector) readerInspector.getStructFieldRef("ts").getFieldObjectInspector();
    HiveDecimalObjectInspector dco = (HiveDecimalObjectInspector) readerInspector.getStructFieldRef("decimal1").getFieldObjectInspector();
    StringObjectInspector mk = (StringObjectInspector) ma.getMapKeyObjectInspector();
    RecordReader rows = reader.rows();
    Object row = rows.next(null);
    assertNotNull(row);
    // check the contents of the first row
    assertEquals(false, bo.get(readerInspector.getStructFieldData(row, fields.get(0))));
    assertEquals(1, by.get(readerInspector.getStructFieldData(row, fields.get(1))));
    assertEquals(1024, sh.get(readerInspector.getStructFieldData(row, fields.get(2))));
    assertEquals(65536, in.get(readerInspector.getStructFieldData(row, fields.get(3))));
    assertEquals(Long.MAX_VALUE, lo.get(readerInspector.getStructFieldData(row, fields.get(4))));
    assertEquals(1.0, fl.get(readerInspector.getStructFieldData(row, fields.get(5))), 0.00001);
    assertEquals(-15.0, dbl.get(readerInspector.getStructFieldData(row, fields.get(6))), 0.00001);
    assertEquals(bytes(0, 1, 2, 3, 4), bi.getPrimitiveWritableObject(readerInspector.getStructFieldData(row, fields.get(7))));
    assertEquals("hi", st.getPrimitiveJavaObject(readerInspector.getStructFieldData(row, fields.get(8))));
    List<?> midRow = midli.getList(mid.getStructFieldData(readerInspector.getStructFieldData(row, fields.get(9)), midFields.get(0)));
    assertNotNull(midRow);
    assertEquals(2, midRow.size());
    assertEquals(1, in.get(inner.getStructFieldData(midRow.get(0), inFields.get(0))));
    assertEquals("bye", st.getPrimitiveJavaObject(inner.getStructFieldData(midRow.get(0), inFields.get(1))));
    assertEquals(2, in.get(inner.getStructFieldData(midRow.get(1), inFields.get(0))));
    assertEquals("sigh", st.getPrimitiveJavaObject(inner.getStructFieldData(midRow.get(1), inFields.get(1))));
    List<?> list = li.getList(readerInspector.getStructFieldData(row, fields.get(10)));
    assertEquals(2, list.size());
    assertEquals(3, in.get(inner.getStructFieldData(list.get(0), inFields.get(0))));
    assertEquals("good", st.getPrimitiveJavaObject(inner.getStructFieldData(list.get(0), inFields.get(1))));
    assertEquals(4, in.get(inner.getStructFieldData(list.get(1), inFields.get(0))));
    assertEquals("bad", st.getPrimitiveJavaObject(inner.getStructFieldData(list.get(1), inFields.get(1))));
    Map<?, ?> map = ma.getMap(readerInspector.getStructFieldData(row, fields.get(11)));
    assertEquals(0, map.size());
    assertEquals(Timestamp.valueOf("2000-03-12 15:00:00"), tso.getPrimitiveJavaObject(readerInspector.getStructFieldData(row, fields.get(12))));
    assertEquals(HiveDecimal.create("12345678.6547456"), dco.getPrimitiveJavaObject(readerInspector.getStructFieldData(row, fields.get(13))));
    // check the contents of second row
    assertEquals(true, rows.hasNext());
    rows.seekToRow(7499);
    row = rows.next(null);
    assertEquals(true, bo.get(readerInspector.getStructFieldData(row, fields.get(0))));
    assertEquals(100, by.get(readerInspector.getStructFieldData(row, fields.get(1))));
    assertEquals(2048, sh.get(readerInspector.getStructFieldData(row, fields.get(2))));
    assertEquals(65536, in.get(readerInspector.getStructFieldData(row, fields.get(3))));
    assertEquals(Long.MAX_VALUE, lo.get(readerInspector.getStructFieldData(row, fields.get(4))));
    assertEquals(2.0, fl.get(readerInspector.getStructFieldData(row, fields.get(5))), 0.00001);
    assertEquals(-5.0, dbl.get(readerInspector.getStructFieldData(row, fields.get(6))), 0.00001);
    assertEquals(bytes(), bi.getPrimitiveWritableObject(readerInspector.getStructFieldData(row, fields.get(7))));
    assertEquals("bye", st.getPrimitiveJavaObject(readerInspector.getStructFieldData(row, fields.get(8))));
    midRow = midli.getList(mid.getStructFieldData(readerInspector.getStructFieldData(row, fields.get(9)), midFields.get(0)));
    assertNotNull(midRow);
    assertEquals(2, midRow.size());
    assertEquals(1, in.get(inner.getStructFieldData(midRow.get(0), inFields.get(0))));
    assertEquals("bye", st.getPrimitiveJavaObject(inner.getStructFieldData(midRow.get(0), inFields.get(1))));
    assertEquals(2, in.get(inner.getStructFieldData(midRow.get(1), inFields.get(0))));
    assertEquals("sigh", st.getPrimitiveJavaObject(inner.getStructFieldData(midRow.get(1), inFields.get(1))));
    list = li.getList(readerInspector.getStructFieldData(row, fields.get(10)));
    assertEquals(3, list.size());
    assertEquals(100000000, in.get(inner.getStructFieldData(list.get(0), inFields.get(0))));
    assertEquals("cat", st.getPrimitiveJavaObject(inner.getStructFieldData(list.get(0), inFields.get(1))));
    assertEquals(-100000, in.get(inner.getStructFieldData(list.get(1), inFields.get(0))));
    assertEquals("in", st.getPrimitiveJavaObject(inner.getStructFieldData(list.get(1), inFields.get(1))));
    assertEquals(1234, in.get(inner.getStructFieldData(list.get(2), inFields.get(0))));
    assertEquals("hat", st.getPrimitiveJavaObject(inner.getStructFieldData(list.get(2), inFields.get(1))));
    map = ma.getMap(readerInspector.getStructFieldData(row, fields.get(11)));
    assertEquals(2, map.size());
    boolean[] found = new boolean[2];
    for (Object key : map.keySet()) {
        String str = mk.getPrimitiveJavaObject(key);
        if (str.equals("chani")) {
            assertEquals(false, found[0]);
            assertEquals(5, in.get(inner.getStructFieldData(map.get(key), inFields.get(0))));
            assertEquals(str, st.getPrimitiveJavaObject(inner.getStructFieldData(map.get(key), inFields.get(1))));
            found[0] = true;
        } else if (str.equals("mauddib")) {
            assertEquals(false, found[1]);
            assertEquals(1, in.get(inner.getStructFieldData(map.get(key), inFields.get(0))));
            assertEquals(str, st.getPrimitiveJavaObject(inner.getStructFieldData(map.get(key), inFields.get(1))));
            found[1] = true;
        } else {
            throw new IllegalArgumentException("Unknown key " + str);
        }
    }
    assertEquals(true, found[0]);
    assertEquals(true, found[1]);
    assertEquals(Timestamp.valueOf("2000-03-12 15:00:01"), tso.getPrimitiveJavaObject(readerInspector.getStructFieldData(row, fields.get(12))));
    assertEquals(HiveDecimal.create("12345678.6547457"), dco.getPrimitiveJavaObject(readerInspector.getStructFieldData(row, fields.get(13))));
    // handle the close up
    assertEquals(false, rows.hasNext());
    rows.close();
}
Also used : LongObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.LongObjectInspector) IntObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.IntObjectInspector) BinaryObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.BinaryObjectInspector) StringObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.StringObjectInspector) FloatObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.FloatObjectInspector) ByteObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.ByteObjectInspector) TimestampObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.TimestampObjectInspector) ShortObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.ShortObjectInspector) MapObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.MapObjectInspector) ListObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.ListObjectInspector) HiveDecimalObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.HiveDecimalObjectInspector) Path(org.apache.hadoop.fs.Path) DecimalColumnStatistics(org.apache.orc.DecimalColumnStatistics) BooleanColumnStatistics(org.apache.orc.BooleanColumnStatistics) StringColumnStatistics(org.apache.orc.StringColumnStatistics) DoubleColumnStatistics(org.apache.orc.DoubleColumnStatistics) IntegerColumnStatistics(org.apache.orc.IntegerColumnStatistics) ColumnStatistics(org.apache.orc.ColumnStatistics) BinaryColumnStatistics(org.apache.orc.BinaryColumnStatistics) DoubleObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.DoubleObjectInspector) BooleanObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.BooleanObjectInspector) StripeInformation(org.apache.orc.StripeInformation) StructObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector) Test(org.junit.Test)

Aggregations

TimestampObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.TimestampObjectInspector)20 DateObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.DateObjectInspector)16 BinaryObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.BinaryObjectInspector)15 HiveCharObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.HiveCharObjectInspector)14 HiveDecimalObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.HiveDecimalObjectInspector)14 LongObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.LongObjectInspector)14 ByteObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.ByteObjectInspector)13 DoubleObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.DoubleObjectInspector)13 FloatObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.FloatObjectInspector)13 HiveVarcharObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.HiveVarcharObjectInspector)13 IntObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.IntObjectInspector)13 ShortObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.ShortObjectInspector)13 StringObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.StringObjectInspector)13 BooleanObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.primitive.BooleanObjectInspector)12 PrimitiveObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.PrimitiveObjectInspector)11 StructObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector)11 Map (java.util.Map)9 MapObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.MapObjectInspector)9 ObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector)9 Text (org.apache.hadoop.io.Text)9