Search in sources :

Example 11 with LazyFactory.createLazyObject

use of org.apache.hadoop.hive.serde2.lazy.LazyFactory.createLazyObject in project hive by apache.

the class TestLazyArrayMapStruct method testNestedinArrayAtLevel.

/**
 * @param nestingLevel
 * @param dtype
 * @param tableProp
 * @throws SerDeException
 */
private void testNestedinArrayAtLevel(int nestingLevel, ObjectInspector.Category dtype, Properties tableProp) throws SerDeException {
    // create type with nestingLevel levels of nesting
    // set inner schema for dtype
    String inSchema = null;
    switch(dtype) {
        case LIST:
            inSchema = "array<tinyint>";
            break;
        case MAP:
            inSchema = "map<string,int>";
            break;
        case STRUCT:
            inSchema = "struct<s:string,i:tinyint>";
            break;
        case UNION:
            inSchema = "uniontype<string,tinyint>";
            break;
        default:
            fail("type not supported by test case");
    }
    StringBuilder schema = new StringBuilder(inSchema);
    for (int i = 0; i < nestingLevel - 1; i++) {
        schema.insert(0, "array<");
        schema.append(">");
    }
    System.err.println("Testing nesting level " + nestingLevel + ". Using schema " + schema);
    // Create the SerDe
    LazySimpleSerDe serDe = new LazySimpleSerDe();
    Configuration conf = new Configuration();
    tableProp.setProperty("columns", "narray");
    tableProp.setProperty("columns.types", schema.toString());
    SerDeUtils.initializeSerDe(serDe, conf, tableProp, null);
    LazySerDeParameters serdeParams = new LazySerDeParameters(conf, tableProp, LazySimpleSerDe.class.getName());
    // create the serialized string for type
    byte[] separators = serdeParams.getSeparators();
    System.err.println("Using separator " + (char) separators[nestingLevel]);
    byte[] serializedRow = null;
    switch(dtype) {
        case LIST:
            serializedRow = new byte[] { '8', separators[nestingLevel], '9' };
            break;
        case MAP:
            byte kvSep = separators[nestingLevel + 1];
            byte kvPairSep = separators[nestingLevel];
            serializedRow = new byte[] { '1', kvSep, '1', kvPairSep, '2', kvSep, '2' };
            break;
        case STRUCT:
            serializedRow = new byte[] { '8', separators[nestingLevel], '9' };
            break;
        case UNION:
            serializedRow = new byte[] { '0', separators[nestingLevel], '9' };
            break;
        default:
            fail("type not supported by test case");
    }
    // create LazyStruct with serialized string with expected separators
    StructObjectInspector oi = (StructObjectInspector) serDe.getObjectInspector();
    LazyStruct struct = (LazyStruct) LazyFactory.createLazyObject(oi);
    TestLazyPrimitive.initLazyObject(struct, serializedRow, 0, serializedRow.length);
    // Get fields out of the lazy struct and check if they match expected
    // results
    // Get first level array
    LazyArray array = (LazyArray) struct.getField(0);
    // Peel off the n-1 levels to get to the underlying array
    for (int i = 0; i < nestingLevel - 2; i++) {
        array = (LazyArray) array.getListElementObject(0);
    }
    // verify the serialized format for dtype
    switch(dtype) {
        case LIST:
            LazyArray array1 = (LazyArray) array.getListElementObject(0);
            // check elements of the innermost array
            assertEquals(2, array1.getListLength());
            assertEquals(new ByteWritable((byte) 8), ((LazyByte) array1.getListElementObject(0)).getWritableObject());
            assertEquals(new ByteWritable((byte) 9), ((LazyByte) array1.getListElementObject(1)).getWritableObject());
            break;
        case MAP:
            LazyMap lazyMap = (LazyMap) array.getListElementObject(0);
            Map map = lazyMap.getMap();
            System.err.println(map);
            assertEquals(2, map.size());
            Iterator<Map.Entry<LazyString, LazyInteger>> it = map.entrySet().iterator();
            Entry<LazyString, LazyInteger> e1 = it.next();
            assertEquals(e1.getKey().getWritableObject(), new Text(new byte[] { '1' }));
            assertEquals(e1.getValue().getWritableObject(), new IntWritable(1));
            Entry<LazyString, LazyInteger> e2 = it.next();
            assertEquals(e2.getKey().getWritableObject(), new Text(new byte[] { '2' }));
            assertEquals(e2.getValue().getWritableObject(), new IntWritable(2));
            break;
        case STRUCT:
            LazyStruct innerStruct = (LazyStruct) array.getListElementObject(0);
            // check elements of the innermost struct
            assertEquals(2, innerStruct.getFieldsAsList().size());
            assertEquals(new Text(new byte[] { '8' }), ((LazyString) innerStruct.getField(0)).getWritableObject());
            assertEquals(new ByteWritable((byte) 9), ((LazyByte) innerStruct.getField(1)).getWritableObject());
            break;
        case UNION:
            LazyUnion lazyUnion = (LazyUnion) array.getListElementObject(0);
            // check elements of the innermost union
            assertEquals(new Text(new byte[] { '9' }), ((LazyString) lazyUnion.getField()).getWritableObject());
            break;
        default:
            fail("type not supported by test case");
    }
    // test serialization
    Text serializedText = (Text) serDe.serialize(struct.getObject(), serDe.getObjectInspector());
    org.junit.Assert.assertArrayEquals(serializedRow, serializedText.getBytes());
}
Also used : Configuration(org.apache.hadoop.conf.Configuration) Text(org.apache.hadoop.io.Text) Entry(java.util.Map.Entry) ByteWritable(org.apache.hadoop.hive.serde2.io.ByteWritable) LazyBinaryMap(org.apache.hadoop.hive.serde2.lazybinary.LazyBinaryMap) Map(java.util.Map) IntWritable(org.apache.hadoop.io.IntWritable) StructObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector)

Example 12 with LazyFactory.createLazyObject

use of org.apache.hadoop.hive.serde2.lazy.LazyFactory.createLazyObject in project hive by apache.

the class TestLazyArrayMapStruct method testLazyArray.

/**
 * Test the LazyArray class.
 */
public void testLazyArray() throws Throwable {
    try {
        // Array of Byte
        Text nullSequence = new Text("\\N");
        ObjectInspector oi = LazyFactory.createLazyObjectInspector(TypeInfoUtils.getTypeInfosFromTypeString("array<tinyint>").get(0), new byte[] { (byte) 1 }, 0, nullSequence, false, (byte) 0);
        LazyArray b = (LazyArray) LazyFactory.createLazyObject(oi);
        byte[] data = new byte[] { '-', '1', 1, '\\', 'N', 1, '8' };
        TestLazyPrimitive.initLazyObject(b, data, 0, data.length);
        assertNull(b.getListElementObject(-1));
        assertEquals(new ByteWritable((byte) -1), ((LazyByte) b.getListElementObject(0)).getWritableObject());
        assertEquals(new ByteWritable((byte) -1), ((LazyByte) b.getList().get(0)).getWritableObject());
        assertNull(b.getListElementObject(1));
        assertNull(b.getList().get(1));
        assertEquals(new ByteWritable((byte) 8), ((LazyByte) b.getListElementObject(2)).getWritableObject());
        assertEquals(new ByteWritable((byte) 8), ((LazyByte) b.getList().get(2)).getWritableObject());
        assertNull(b.getListElementObject(3));
        assertEquals(3, b.getList().size());
        // Array of String
        oi = LazyFactory.createLazyObjectInspector(TypeInfoUtils.getTypeInfosFromTypeString("array<string>").get(0), new byte[] { (byte) '\t' }, 0, nullSequence, false, (byte) 0);
        b = (LazyArray) LazyFactory.createLazyObject(oi);
        data = new byte[] { 'a', 'b', '\t', 'c', '\t', '\\', 'N', '\t', '\t', 'd' };
        // Note: the first and last element of the byte[] are NOT used
        TestLazyPrimitive.initLazyObject(b, data, 1, data.length - 2);
        assertNull(b.getListElementObject(-1));
        assertEquals(new Text("b"), ((LazyString) b.getListElementObject(0)).getWritableObject());
        assertEquals(new Text("b"), ((LazyString) b.getList().get(0)).getWritableObject());
        assertEquals(new Text("c"), ((LazyString) b.getListElementObject(1)).getWritableObject());
        assertEquals(new Text("c"), ((LazyString) b.getList().get(1)).getWritableObject());
        assertNull((b.getListElementObject(2)));
        assertNull((b.getList().get(2)));
        assertEquals(new Text(""), ((LazyString) b.getListElementObject(3)).getWritableObject());
        assertEquals(new Text(""), ((LazyString) b.getList().get(3)).getWritableObject());
        assertEquals(new Text(""), ((LazyString) b.getListElementObject(4)).getWritableObject());
        assertEquals(new Text(""), ((LazyString) b.getList().get(4)).getWritableObject());
        assertNull((b.getListElementObject(5)));
        assertEquals(5, b.getList().size());
        // -- HIVE-4149
        b = (LazyArray) LazyFactory.createLazyObject(oi);
        data = new byte[] { 'a', '\t', '\\', 'N' };
        TestLazyPrimitive.initLazyObject(b, data, 0, data.length);
        assertEquals(new Text("a"), ((LazyString) b.getListElementObject(0)).getWritableObject());
        assertNull(b.getListElementObject(1));
        data = new byte[] { '\\', 'N', '\t', 'a' };
        TestLazyPrimitive.initLazyObject(b, data, 0, data.length);
        assertNull(b.getListElementObject(0));
        // twice (returns not cleaned cache)
        assertNull(b.getListElementObject(0));
        assertEquals(new Text("a"), ((LazyString) b.getListElementObject(1)).getWritableObject());
    } catch (Throwable e) {
        e.printStackTrace();
        throw e;
    }
}
Also used : StructObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector) ObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector) Text(org.apache.hadoop.io.Text) ByteWritable(org.apache.hadoop.hive.serde2.io.ByteWritable)

Example 13 with LazyFactory.createLazyObject

use of org.apache.hadoop.hive.serde2.lazy.LazyFactory.createLazyObject in project hive by apache.

the class TestLazyBinarySerDe method testLazyBinaryObjectInspector.

/**
 * Test to see if byte[] with correct contents is generated by
 * LazyBinaryObjectInspector from input BytesWritable
 * @throws Throwable
 */
public void testLazyBinaryObjectInspector() throws Throwable {
    // create input ByteArrayRef
    ByteArrayRef inpBARef = new ByteArrayRef();
    inpBARef.setData(inpBArray);
    AbstractPrimitiveLazyObjectInspector<?> binInspector = LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(TypeInfoFactory.binaryTypeInfo, false, (byte) 0);
    // create LazyBinary initialed with inputBA
    LazyBinary lazyBin = (LazyBinary) LazyFactory.createLazyObject(binInspector);
    lazyBin.init(inpBARef, 0, inpBArray.length);
    // use inspector to get a byte[] out of LazyBinary
    byte[] outBARef = (byte[]) binInspector.getPrimitiveJavaObject(lazyBin);
    assertTrue("compare input and output BAs", Arrays.equals(inpBArray, outBARef));
}
Also used : ByteArrayRef(org.apache.hadoop.hive.serde2.lazy.ByteArrayRef) LazyBinary(org.apache.hadoop.hive.serde2.lazy.LazyBinary)

Example 14 with LazyFactory.createLazyObject

use of org.apache.hadoop.hive.serde2.lazy.LazyFactory.createLazyObject in project hive by apache.

the class TestHiveAccumuloTableOutputFormat method testWriteMap.

@Test
public void testWriteMap() throws Exception {
    Instance inst = new MockInstance(test.getMethodName());
    Connector conn = inst.getConnector("root", new PasswordToken(""));
    HiveAccumuloTableOutputFormat outputFormat = new HiveAccumuloTableOutputFormat();
    String table = test.getMethodName();
    conn.tableOperations().create(table);
    JobConf conf = new JobConf();
    conf.set(AccumuloConnectionParameters.INSTANCE_NAME, inst.getInstanceName());
    conf.set(AccumuloConnectionParameters.USER_NAME, "root");
    conf.set(AccumuloConnectionParameters.USER_PASS, "");
    conf.setBoolean(AccumuloConnectionParameters.USE_MOCK_INSTANCE, true);
    conf.set(AccumuloConnectionParameters.TABLE_NAME, test.getMethodName());
    FileSystem local = FileSystem.getLocal(conf);
    outputFormat.checkOutputSpecs(local, conf);
    RecordWriter<Text, Mutation> recordWriter = outputFormat.getRecordWriter(local, conf, null, null);
    List<String> names = Arrays.asList("row", "col1");
    List<TypeInfo> types = Arrays.<TypeInfo>asList(TypeInfoFactory.stringTypeInfo, TypeInfoFactory.stringTypeInfo);
    Properties tableProperties = new Properties();
    tableProperties.setProperty(AccumuloSerDeParameters.COLUMN_MAPPINGS, ":rowID,cf:*");
    tableProperties.setProperty(serdeConstants.FIELD_DELIM, " ");
    tableProperties.setProperty(serdeConstants.LIST_COLUMNS, Joiner.on(',').join(names));
    tableProperties.setProperty(serdeConstants.LIST_COLUMN_TYPES, Joiner.on(',').join(types));
    AccumuloSerDeParameters accumuloSerDeParams = new AccumuloSerDeParameters(new Configuration(), tableProperties, AccumuloSerDe.class.getSimpleName());
    LazySerDeParameters serDeParams = accumuloSerDeParams.getSerDeParameters();
    AccumuloRowSerializer serializer = new AccumuloRowSerializer(0, serDeParams, accumuloSerDeParams.getColumnMappings(), AccumuloSerDeParameters.DEFAULT_VISIBILITY_LABEL, accumuloSerDeParams.getRowIdFactory());
    TypeInfo stringTypeInfo = TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.STRING_TYPE_NAME);
    LazyStringObjectInspector stringOI = (LazyStringObjectInspector) LazyFactory.createLazyObjectInspector(stringTypeInfo, new byte[] { 0 }, 0, serDeParams.getNullSequence(), serDeParams.isEscaped(), serDeParams.getEscapeChar());
    LazyMapObjectInspector mapOI = LazyObjectInspectorFactory.getLazySimpleMapObjectInspector(stringOI, stringOI, (byte) ',', (byte) ':', serDeParams.getNullSequence(), serDeParams.isEscaped(), serDeParams.getEscapeChar());
    LazySimpleStructObjectInspector structOI = (LazySimpleStructObjectInspector) LazyObjectInspectorFactory.getLazySimpleStructObjectInspector(Arrays.asList("row", "data"), Arrays.asList(stringOI, mapOI), (byte) ' ', serDeParams.getNullSequence(), serDeParams.isLastColumnTakesRest(), serDeParams.isEscaped(), serDeParams.getEscapeChar());
    LazyStruct struct = (LazyStruct) LazyFactory.createLazyObject(structOI);
    ByteArrayRef bytes = new ByteArrayRef();
    bytes.setData("row cq1:value1,cq2:value2".getBytes());
    struct.init(bytes, 0, bytes.getData().length);
    // Serialize the struct into a mutation
    Mutation m = serializer.serialize(struct, structOI);
    // Write the mutation
    recordWriter.write(new Text(table), m);
    // Close the writer
    recordWriter.close(null);
    Iterator<Entry<Key, Value>> iter = conn.createScanner(table, new Authorizations()).iterator();
    Assert.assertTrue("Iterator did not have an element as expected", iter.hasNext());
    Entry<Key, Value> entry = iter.next();
    Key k = entry.getKey();
    Value v = entry.getValue();
    Assert.assertEquals("row", k.getRow().toString());
    Assert.assertEquals("cf", k.getColumnFamily().toString());
    Assert.assertEquals("cq1", k.getColumnQualifier().toString());
    Assert.assertEquals(AccumuloSerDeParameters.DEFAULT_VISIBILITY_LABEL, k.getColumnVisibilityParsed());
    Assert.assertEquals("value1", new String(v.get()));
    Assert.assertTrue("Iterator did not have an element as expected", iter.hasNext());
    entry = iter.next();
    k = entry.getKey();
    v = entry.getValue();
    Assert.assertEquals("row", k.getRow().toString());
    Assert.assertEquals("cf", k.getColumnFamily().toString());
    Assert.assertEquals("cq2", k.getColumnQualifier().toString());
    Assert.assertEquals(AccumuloSerDeParameters.DEFAULT_VISIBILITY_LABEL, k.getColumnVisibilityParsed());
    Assert.assertEquals("value2", new String(v.get()));
    Assert.assertFalse("Iterator unexpectedly had more data", iter.hasNext());
}
Also used : Connector(org.apache.accumulo.core.client.Connector) Configuration(org.apache.hadoop.conf.Configuration) MockInstance(org.apache.accumulo.core.client.mock.MockInstance) Instance(org.apache.accumulo.core.client.Instance) LazySerDeParameters(org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters) Properties(java.util.Properties) PasswordToken(org.apache.accumulo.core.client.security.tokens.PasswordToken) Entry(java.util.Map.Entry) MockInstance(org.apache.accumulo.core.client.mock.MockInstance) FileSystem(org.apache.hadoop.fs.FileSystem) JobConf(org.apache.hadoop.mapred.JobConf) AccumuloRowSerializer(org.apache.hadoop.hive.accumulo.serde.AccumuloRowSerializer) LazyStruct(org.apache.hadoop.hive.serde2.lazy.LazyStruct) Authorizations(org.apache.accumulo.core.security.Authorizations) LazyStringObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyStringObjectInspector) LazySimpleStructObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.LazySimpleStructObjectInspector) Text(org.apache.hadoop.io.Text) LazyMapObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.LazyMapObjectInspector) TypeInfo(org.apache.hadoop.hive.serde2.typeinfo.TypeInfo) AccumuloSerDe(org.apache.hadoop.hive.accumulo.serde.AccumuloSerDe) AccumuloSerDeParameters(org.apache.hadoop.hive.accumulo.serde.AccumuloSerDeParameters) ByteArrayRef(org.apache.hadoop.hive.serde2.lazy.ByteArrayRef) Value(org.apache.accumulo.core.data.Value) Mutation(org.apache.accumulo.core.data.Mutation) Key(org.apache.accumulo.core.data.Key) Test(org.junit.Test)

Example 15 with LazyFactory.createLazyObject

use of org.apache.hadoop.hive.serde2.lazy.LazyFactory.createLazyObject in project hive by apache.

the class TestHiveAccumuloTypes method testUtf8Types.

@Test
public void testUtf8Types() throws Exception {
    final String tableName = test.getMethodName(), user = "root", pass = "";
    MockInstance mockInstance = new MockInstance(test.getMethodName());
    Connector conn = mockInstance.getConnector(user, new PasswordToken(pass));
    HiveAccumuloTableInputFormat inputformat = new HiveAccumuloTableInputFormat();
    JobConf conf = new JobConf();
    conf.set(AccumuloSerDeParameters.TABLE_NAME, tableName);
    conf.set(AccumuloSerDeParameters.USE_MOCK_INSTANCE, "true");
    conf.set(AccumuloSerDeParameters.INSTANCE_NAME, test.getMethodName());
    conf.set(AccumuloSerDeParameters.USER_NAME, user);
    conf.set(AccumuloSerDeParameters.USER_PASS, pass);
    // not used for mock, but
    conf.set(AccumuloSerDeParameters.ZOOKEEPERS, "localhost:2181");
    // required by input format.
    conf.set(AccumuloSerDeParameters.COLUMN_MAPPINGS, AccumuloHiveConstants.ROWID + ",cf:string,cf:boolean,cf:tinyint,cf:smallint,cf:int,cf:bigint" + ",cf:float,cf:double,cf:decimal,cf:date,cf:timestamp,cf:char,cf:varchar");
    conf.set(serdeConstants.LIST_COLUMNS, "string,string,boolean,tinyint,smallint,int,bigint,float,double,decimal,date,timestamp,char(4),varchar(7)");
    conf.set(serdeConstants.LIST_COLUMN_TYPES, "string,string,boolean,tinyint,smallint,int,bigint,float,double,decimal,date,timestamp,char(4),varchar(7)");
    conn.tableOperations().create(tableName);
    BatchWriterConfig writerConf = new BatchWriterConfig();
    BatchWriter writer = conn.createBatchWriter(tableName, writerConf);
    ByteArrayOutputStream baos = new ByteArrayOutputStream();
    String cf = "cf";
    byte[] cfBytes = cf.getBytes();
    ByteArrayRef byteRef = new ByteArrayRef();
    Mutation m = new Mutation("row1");
    // string
    String stringValue = "string";
    baos.reset();
    JavaStringObjectInspector stringOI = (JavaStringObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.STRING_TYPE_NAME));
    LazyUtils.writePrimitiveUTF8(baos, stringOI.create(stringValue), stringOI, false, (byte) 0, null);
    m.put(cfBytes, "string".getBytes(), baos.toByteArray());
    // boolean
    boolean booleanValue = true;
    baos.reset();
    JavaBooleanObjectInspector booleanOI = (JavaBooleanObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.BOOLEAN_TYPE_NAME));
    LazyUtils.writePrimitiveUTF8(baos, booleanOI.create(booleanValue), booleanOI, false, (byte) 0, null);
    m.put(cfBytes, "boolean".getBytes(), baos.toByteArray());
    // tinyint
    byte tinyintValue = -127;
    baos.reset();
    JavaByteObjectInspector byteOI = (JavaByteObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.TINYINT_TYPE_NAME));
    LazyUtils.writePrimitiveUTF8(baos, tinyintValue, byteOI, false, (byte) 0, null);
    m.put(cfBytes, "tinyint".getBytes(), baos.toByteArray());
    // smallint
    short smallintValue = Short.MAX_VALUE;
    baos.reset();
    JavaShortObjectInspector shortOI = (JavaShortObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.SMALLINT_TYPE_NAME));
    LazyUtils.writePrimitiveUTF8(baos, smallintValue, shortOI, false, (byte) 0, null);
    m.put(cfBytes, "smallint".getBytes(), baos.toByteArray());
    // int
    int intValue = Integer.MAX_VALUE;
    baos.reset();
    JavaIntObjectInspector intOI = (JavaIntObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.INT_TYPE_NAME));
    LazyUtils.writePrimitiveUTF8(baos, intValue, intOI, false, (byte) 0, null);
    m.put(cfBytes, "int".getBytes(), baos.toByteArray());
    // bigint
    long bigintValue = Long.MAX_VALUE;
    baos.reset();
    JavaLongObjectInspector longOI = (JavaLongObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.BIGINT_TYPE_NAME));
    LazyUtils.writePrimitiveUTF8(baos, bigintValue, longOI, false, (byte) 0, null);
    m.put(cfBytes, "bigint".getBytes(), baos.toByteArray());
    // float
    float floatValue = Float.MAX_VALUE;
    baos.reset();
    JavaFloatObjectInspector floatOI = (JavaFloatObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.FLOAT_TYPE_NAME));
    LazyUtils.writePrimitiveUTF8(baos, floatValue, floatOI, false, (byte) 0, null);
    m.put(cfBytes, "float".getBytes(), baos.toByteArray());
    // double
    double doubleValue = Double.MAX_VALUE;
    baos.reset();
    JavaDoubleObjectInspector doubleOI = (JavaDoubleObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.DOUBLE_TYPE_NAME));
    LazyUtils.writePrimitiveUTF8(baos, doubleValue, doubleOI, false, (byte) 0, null);
    m.put(cfBytes, "double".getBytes(), baos.toByteArray());
    // decimal
    HiveDecimal decimalValue = HiveDecimal.create("1.23");
    baos.reset();
    JavaHiveDecimalObjectInspector decimalOI = (JavaHiveDecimalObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(new DecimalTypeInfo(5, 2));
    LazyUtils.writePrimitiveUTF8(baos, decimalOI.create(decimalValue), decimalOI, false, (byte) 0, null);
    m.put(cfBytes, "decimal".getBytes(), baos.toByteArray());
    // date
    Date now = new Date(System.currentTimeMillis());
    DateWritable dateWritable = new DateWritable(now);
    Date dateValue = dateWritable.get();
    baos.reset();
    JavaDateObjectInspector dateOI = (JavaDateObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.DATE_TYPE_NAME));
    LazyUtils.writePrimitiveUTF8(baos, dateOI.create(dateValue), dateOI, false, (byte) 0, null);
    m.put(cfBytes, "date".getBytes(), baos.toByteArray());
    // timestamp
    Timestamp timestampValue = new Timestamp(now.getTime());
    baos.reset();
    JavaTimestampObjectInspector timestampOI = (JavaTimestampObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.TIMESTAMP_TYPE_NAME));
    LazyUtils.writePrimitiveUTF8(baos, timestampOI.create(timestampValue), timestampOI, false, (byte) 0, null);
    m.put(cfBytes, "timestamp".getBytes(), baos.toByteArray());
    // char
    baos.reset();
    HiveChar charValue = new HiveChar("char", 4);
    JavaHiveCharObjectInspector charOI = (JavaHiveCharObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(new CharTypeInfo(4));
    LazyUtils.writePrimitiveUTF8(baos, charOI.create(charValue), charOI, false, (byte) 0, null);
    m.put(cfBytes, "char".getBytes(), baos.toByteArray());
    // varchar
    baos.reset();
    HiveVarchar varcharValue = new HiveVarchar("varchar", 7);
    JavaHiveVarcharObjectInspector varcharOI = (JavaHiveVarcharObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(new VarcharTypeInfo(7));
    LazyUtils.writePrimitiveUTF8(baos, varcharOI.create(varcharValue), varcharOI, false, (byte) 0, null);
    m.put(cfBytes, "varchar".getBytes(), baos.toByteArray());
    writer.addMutation(m);
    writer.close();
    for (Entry<Key, Value> e : conn.createScanner(tableName, new Authorizations())) {
        System.out.println(e);
    }
    // Create the RecordReader
    FileInputFormat.addInputPath(conf, new Path("unused"));
    InputSplit[] splits = inputformat.getSplits(conf, 0);
    assertEquals(splits.length, 1);
    RecordReader<Text, AccumuloHiveRow> reader = inputformat.getRecordReader(splits[0], conf, null);
    Text key = reader.createKey();
    AccumuloHiveRow value = reader.createValue();
    reader.next(key, value);
    Assert.assertEquals(13, value.getTuples().size());
    // string
    Text cfText = new Text(cf), cqHolder = new Text();
    cqHolder.set("string");
    byte[] valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyStringObjectInspector lazyStringOI = LazyPrimitiveObjectInspectorFactory.getLazyStringObjectInspector(false, (byte) 0);
    LazyString lazyString = (LazyString) LazyFactory.createLazyObject(lazyStringOI);
    lazyString.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(new Text(stringValue), lazyString.getWritableObject());
    // boolean
    cqHolder.set("boolean");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyBooleanObjectInspector lazyBooleanOI = (LazyBooleanObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.BOOLEAN_TYPE_NAME));
    LazyBoolean lazyBoolean = (LazyBoolean) LazyFactory.createLazyObject(lazyBooleanOI);
    lazyBoolean.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(booleanValue, lazyBoolean.getWritableObject().get());
    // tinyint
    cqHolder.set("tinyint");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyByteObjectInspector lazyByteOI = (LazyByteObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.TINYINT_TYPE_NAME));
    LazyByte lazyByte = (LazyByte) LazyFactory.createLazyObject(lazyByteOI);
    lazyByte.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(tinyintValue, lazyByte.getWritableObject().get());
    // smallint
    cqHolder.set("smallint");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyShortObjectInspector lazyShortOI = (LazyShortObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.SMALLINT_TYPE_NAME));
    LazyShort lazyShort = (LazyShort) LazyFactory.createLazyObject(lazyShortOI);
    lazyShort.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(smallintValue, lazyShort.getWritableObject().get());
    // int
    cqHolder.set("int");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyIntObjectInspector lazyIntOI = (LazyIntObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.INT_TYPE_NAME));
    LazyInteger lazyInt = (LazyInteger) LazyFactory.createLazyObject(lazyIntOI);
    lazyInt.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(intValue, lazyInt.getWritableObject().get());
    // bigint
    cqHolder.set("bigint");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyLongObjectInspector lazyLongOI = (LazyLongObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.BIGINT_TYPE_NAME));
    LazyLong lazyLong = (LazyLong) LazyFactory.createLazyObject(lazyLongOI);
    lazyLong.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(bigintValue, lazyLong.getWritableObject().get());
    // float
    cqHolder.set("float");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyFloatObjectInspector lazyFloatOI = (LazyFloatObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.FLOAT_TYPE_NAME));
    LazyFloat lazyFloat = (LazyFloat) LazyFactory.createLazyObject(lazyFloatOI);
    lazyFloat.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(floatValue, lazyFloat.getWritableObject().get(), 0);
    // double
    cqHolder.set("double");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyDoubleObjectInspector lazyDoubleOI = (LazyDoubleObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.DOUBLE_TYPE_NAME));
    LazyDouble lazyDouble = (LazyDouble) LazyFactory.createLazyObject(lazyDoubleOI);
    lazyDouble.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(doubleValue, lazyDouble.getWritableObject().get(), 0);
    // decimal
    cqHolder.set("decimal");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyHiveDecimalObjectInspector lazyDecimalOI = (LazyHiveDecimalObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(new DecimalTypeInfo(5, 2));
    LazyHiveDecimal lazyDecimal = (LazyHiveDecimal) LazyFactory.createLazyObject(lazyDecimalOI);
    lazyDecimal.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(decimalValue, lazyDecimal.getWritableObject().getHiveDecimal());
    // date
    cqHolder.set("date");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyDateObjectInspector lazyDateOI = (LazyDateObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.DATE_TYPE_NAME));
    LazyDate lazyDate = (LazyDate) LazyFactory.createLazyObject(lazyDateOI);
    lazyDate.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(dateValue, lazyDate.getWritableObject().get());
    // timestamp
    cqHolder.set("timestamp");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyTimestampObjectInspector lazyTimestampOI = (LazyTimestampObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.TIMESTAMP_TYPE_NAME));
    LazyTimestamp lazyTimestamp = (LazyTimestamp) LazyFactory.createLazyObject(lazyTimestampOI);
    lazyTimestamp.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(timestampValue, lazyTimestamp.getWritableObject().getTimestamp());
    // char
    cqHolder.set("char");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyHiveCharObjectInspector lazyCharOI = (LazyHiveCharObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(new CharTypeInfo(4));
    LazyHiveChar lazyChar = (LazyHiveChar) LazyFactory.createLazyObject(lazyCharOI);
    lazyChar.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(charValue, lazyChar.getWritableObject().getHiveChar());
    // varchar
    cqHolder.set("varchar");
    valueBytes = value.getValue(cfText, cqHolder);
    Assert.assertNotNull(valueBytes);
    byteRef.setData(valueBytes);
    LazyHiveVarcharObjectInspector lazyVarcharOI = (LazyHiveVarcharObjectInspector) LazyPrimitiveObjectInspectorFactory.getLazyObjectInspector(new VarcharTypeInfo(7));
    LazyHiveVarchar lazyVarchar = (LazyHiveVarchar) LazyFactory.createLazyObject(lazyVarcharOI);
    lazyVarchar.init(byteRef, 0, valueBytes.length);
    Assert.assertEquals(varcharValue.toString(), lazyVarchar.getWritableObject().getHiveVarchar().toString());
}
Also used : LazyHiveVarchar(org.apache.hadoop.hive.serde2.lazy.LazyHiveVarchar) VarcharTypeInfo(org.apache.hadoop.hive.serde2.typeinfo.VarcharTypeInfo) LazyHiveDecimal(org.apache.hadoop.hive.serde2.lazy.LazyHiveDecimal) LazyIntObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyIntObjectInspector) LazyDoubleObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyDoubleObjectInspector) LazyString(org.apache.hadoop.hive.serde2.lazy.LazyString) AccumuloHiveRow(org.apache.hadoop.hive.accumulo.AccumuloHiveRow) LazyHiveChar(org.apache.hadoop.hive.serde2.lazy.LazyHiveChar) PasswordToken(org.apache.accumulo.core.client.security.tokens.PasswordToken) BatchWriterConfig(org.apache.accumulo.core.client.BatchWriterConfig) JobConf(org.apache.hadoop.mapred.JobConf) JavaHiveCharObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaHiveCharObjectInspector) LazyShortObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyShortObjectInspector) Authorizations(org.apache.accumulo.core.security.Authorizations) LazyLong(org.apache.hadoop.hive.serde2.lazy.LazyLong) LazyStringObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyStringObjectInspector) JavaStringObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaStringObjectInspector) LazyDateObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyDateObjectInspector) CharTypeInfo(org.apache.hadoop.hive.serde2.typeinfo.CharTypeInfo) JavaLongObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaLongObjectInspector) LazyHiveVarcharObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyHiveVarcharObjectInspector) JavaTimestampObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaTimestampObjectInspector) LazyBoolean(org.apache.hadoop.hive.serde2.lazy.LazyBoolean) LazyLongObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyLongObjectInspector) DecimalTypeInfo(org.apache.hadoop.hive.serde2.typeinfo.DecimalTypeInfo) LazyByte(org.apache.hadoop.hive.serde2.lazy.LazyByte) JavaHiveVarcharObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaHiveVarcharObjectInspector) ByteArrayRef(org.apache.hadoop.hive.serde2.lazy.ByteArrayRef) JavaFloatObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaFloatObjectInspector) Value(org.apache.accumulo.core.data.Value) LazyInteger(org.apache.hadoop.hive.serde2.lazy.LazyInteger) Mutation(org.apache.accumulo.core.data.Mutation) LazyDouble(org.apache.hadoop.hive.serde2.lazy.LazyDouble) Key(org.apache.accumulo.core.data.Key) Connector(org.apache.accumulo.core.client.Connector) LazyHiveCharObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyHiveCharObjectInspector) JavaIntObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaIntObjectInspector) LazyTimestamp(org.apache.hadoop.hive.serde2.lazy.LazyTimestamp) HiveChar(org.apache.hadoop.hive.common.type.HiveChar) LazyHiveChar(org.apache.hadoop.hive.serde2.lazy.LazyHiveChar) LazyBooleanObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyBooleanObjectInspector) LazyFloat(org.apache.hadoop.hive.serde2.lazy.LazyFloat) Timestamp(java.sql.Timestamp) LazyTimestamp(org.apache.hadoop.hive.serde2.lazy.LazyTimestamp) JavaDoubleObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaDoubleObjectInspector) LazyShort(org.apache.hadoop.hive.serde2.lazy.LazyShort) LazyString(org.apache.hadoop.hive.serde2.lazy.LazyString) JavaHiveDecimalObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaHiveDecimalObjectInspector) MockInstance(org.apache.accumulo.core.client.mock.MockInstance) JavaDateObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaDateObjectInspector) JavaByteObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaByteObjectInspector) LazyHiveDecimal(org.apache.hadoop.hive.serde2.lazy.LazyHiveDecimal) HiveDecimal(org.apache.hadoop.hive.common.type.HiveDecimal) JavaBooleanObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaBooleanObjectInspector) InputSplit(org.apache.hadoop.mapred.InputSplit) Path(org.apache.hadoop.fs.Path) DateWritable(org.apache.hadoop.hive.serde2.io.DateWritable) Text(org.apache.hadoop.io.Text) ByteArrayOutputStream(java.io.ByteArrayOutputStream) HiveVarchar(org.apache.hadoop.hive.common.type.HiveVarchar) LazyHiveVarchar(org.apache.hadoop.hive.serde2.lazy.LazyHiveVarchar) LazyFloatObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyFloatObjectInspector) LazyTimestampObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyTimestampObjectInspector) LazyDate(org.apache.hadoop.hive.serde2.lazy.LazyDate) LazyDate(org.apache.hadoop.hive.serde2.lazy.LazyDate) Date(java.sql.Date) LazyHiveDecimalObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyHiveDecimalObjectInspector) LazyByteObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.primitive.LazyByteObjectInspector) BatchWriter(org.apache.accumulo.core.client.BatchWriter) JavaShortObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.primitive.JavaShortObjectInspector) Test(org.junit.Test)

Aggregations

ByteArrayRef (org.apache.hadoop.hive.serde2.lazy.ByteArrayRef)19 LazySimpleStructObjectInspector (org.apache.hadoop.hive.serde2.lazy.objectinspector.LazySimpleStructObjectInspector)14 Mutation (org.apache.accumulo.core.data.Mutation)10 Test (org.junit.Test)10 Configuration (org.apache.hadoop.conf.Configuration)9 LazySerDeParameters (org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters)9 Text (org.apache.hadoop.io.Text)9 Properties (java.util.Properties)8 LazyStruct (org.apache.hadoop.hive.serde2.lazy.LazyStruct)8 ObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector)8 TypeInfo (org.apache.hadoop.hive.serde2.typeinfo.TypeInfo)8 LazyMapObjectInspector (org.apache.hadoop.hive.serde2.lazy.objectinspector.LazyMapObjectInspector)7 Entry (java.util.Map.Entry)6 Connector (org.apache.accumulo.core.client.Connector)6 MockInstance (org.apache.accumulo.core.client.mock.MockInstance)6 PasswordToken (org.apache.accumulo.core.client.security.tokens.PasswordToken)6 Key (org.apache.accumulo.core.data.Key)6 Value (org.apache.accumulo.core.data.Value)6 Authorizations (org.apache.accumulo.core.security.Authorizations)6 JobConf (org.apache.hadoop.mapred.JobConf)6