Search in sources :

Example 21 with Mutation

use of org.apache.accumulo.core.data.Mutation in project hive by apache.

the class TestAccumuloRowSerializer method testVisibilityLabel.

@Test
public void testVisibilityLabel() throws IOException, SerDeException {
    List<String> columns = Arrays.asList("row", "cq1", "cq2", "cq3");
    List<TypeInfo> types = Arrays.<TypeInfo>asList(TypeInfoFactory.stringTypeInfo, TypeInfoFactory.intTypeInfo, TypeInfoFactory.intTypeInfo, TypeInfoFactory.stringTypeInfo);
    List<String> typeNames = new ArrayList<String>(types.size());
    for (TypeInfo type : types) {
        typeNames.add(type.getTypeName());
    }
    Properties tableProperties = new Properties();
    tableProperties.setProperty(AccumuloSerDeParameters.COLUMN_MAPPINGS, ":rowid,cf:cq1#b,cf:cq2#b,cf:cq3");
    tableProperties.setProperty(serdeConstants.FIELD_DELIM, " ");
    tableProperties.setProperty(serdeConstants.LIST_COLUMNS, Joiner.on(',').join(columns));
    tableProperties.setProperty(serdeConstants.LIST_COLUMN_TYPES, Joiner.on(',').join(typeNames));
    AccumuloSerDeParameters accumuloSerDeParams = new AccumuloSerDeParameters(new Configuration(), tableProperties, AccumuloSerDe.class.getSimpleName());
    LazySerDeParameters serDeParams = accumuloSerDeParams.getSerDeParameters();
    LazySimpleStructObjectInspector oi = (LazySimpleStructObjectInspector) LazyFactory.createLazyStructInspector(columns, types, serDeParams.getSeparators(), serDeParams.getNullSequence(), serDeParams.isLastColumnTakesRest(), serDeParams.isEscaped(), serDeParams.getEscapeChar());
    AccumuloRowSerializer serializer = new AccumuloRowSerializer(0, serDeParams, accumuloSerDeParams.getColumnMappings(), new ColumnVisibility("foo"), accumuloSerDeParams.getRowIdFactory());
    // Create the LazyStruct from the LazyStruct...Inspector
    LazyStruct obj = (LazyStruct) LazyFactory.createLazyObject(oi);
    ByteArrayRef byteRef = new ByteArrayRef();
    byteRef.setData(new byte[] { 'r', 'o', 'w', '1', ' ', '1', '0', ' ', '2', '0', ' ', 'v', 'a', 'l', 'u', 'e' });
    obj.init(byteRef, 0, byteRef.getData().length);
    Mutation m = (Mutation) serializer.serialize(obj, oi);
    Assert.assertArrayEquals("row1".getBytes(), m.getRow());
    List<ColumnUpdate> updates = m.getUpdates();
    Assert.assertEquals(3, updates.size());
    ByteArrayOutputStream baos = new ByteArrayOutputStream();
    DataOutputStream out = new DataOutputStream(baos);
    ColumnUpdate update = updates.get(0);
    Assert.assertEquals("cf", new String(update.getColumnFamily()));
    Assert.assertEquals("cq1", new String(update.getColumnQualifier()));
    Assert.assertEquals("foo", new String(update.getColumnVisibility()));
    out.writeInt(10);
    Assert.assertArrayEquals(baos.toByteArray(), update.getValue());
    update = updates.get(1);
    Assert.assertEquals("cf", new String(update.getColumnFamily()));
    Assert.assertEquals("cq2", new String(update.getColumnQualifier()));
    Assert.assertEquals("foo", new String(update.getColumnVisibility()));
    baos.reset();
    out.writeInt(20);
    Assert.assertArrayEquals(baos.toByteArray(), update.getValue());
    update = updates.get(2);
    Assert.assertEquals("cf", new String(update.getColumnFamily()));
    Assert.assertEquals("cq3", new String(update.getColumnQualifier()));
    Assert.assertEquals("foo", new String(update.getColumnVisibility()));
    Assert.assertEquals("value", new String(update.getValue()));
}
Also used : LazySimpleStructObjectInspector(org.apache.hadoop.hive.serde2.lazy.objectinspector.LazySimpleStructObjectInspector) ColumnUpdate(org.apache.accumulo.core.data.ColumnUpdate) Configuration(org.apache.hadoop.conf.Configuration) LazySerDeParameters(org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters) DataOutputStream(java.io.DataOutputStream) ArrayList(java.util.ArrayList) ByteArrayOutputStream(java.io.ByteArrayOutputStream) Properties(java.util.Properties) TypeInfo(org.apache.hadoop.hive.serde2.typeinfo.TypeInfo) ByteArrayRef(org.apache.hadoop.hive.serde2.lazy.ByteArrayRef) ColumnVisibility(org.apache.accumulo.core.security.ColumnVisibility) Mutation(org.apache.accumulo.core.data.Mutation) LazyStruct(org.apache.hadoop.hive.serde2.lazy.LazyStruct) Test(org.junit.Test)

Example 22 with Mutation

use of org.apache.accumulo.core.data.Mutation 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 23 with Mutation

use of org.apache.accumulo.core.data.Mutation 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)

Example 24 with Mutation

use of org.apache.accumulo.core.data.Mutation in project Gaffer by gchq.

the class RowIdAggregatorTest method testAggregatingMultiplePropertySetsAcrossRowIDRange.

private void testAggregatingMultiplePropertySetsAcrossRowIDRange(final AccumuloStore store, final AccumuloElementConverter elementConverter) throws StoreException, AccumuloElementConversionException, RangeFactoryException {
    String visibilityString = "public";
    try {
        // Create table
        // (this method creates the table, removes the versioning iterator, and adds the SetOfStatisticsCombiner iterator).
        TableUtils.createTable(store);
        final Properties properties1 = new Properties();
        properties1.put(AccumuloPropertyNames.COUNT, 1);
        final Properties properties2 = new Properties();
        properties2.put(AccumuloPropertyNames.COUNT, 1);
        final Properties properties3 = new Properties();
        properties3.put(AccumuloPropertyNames.COUNT, 2);
        // Create edge
        final Edge edge = new Edge(TestGroups.EDGE);
        edge.setSource("2");
        edge.setDestination("1");
        edge.setDirected(true);
        edge.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 1);
        edge.putProperty(AccumuloPropertyNames.PROP_1, 0);
        edge.putProperty(AccumuloPropertyNames.PROP_2, 0);
        edge.putProperty(AccumuloPropertyNames.PROP_3, 0);
        edge.putProperty(AccumuloPropertyNames.PROP_4, 0);
        final Edge edge2 = new Edge(TestGroups.EDGE);
        edge2.setSource("B");
        edge2.setDestination("Z");
        edge2.setDirected(true);
        edge2.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 1);
        edge2.putProperty(AccumuloPropertyNames.PROP_1, 1);
        edge2.putProperty(AccumuloPropertyNames.PROP_2, 1);
        edge2.putProperty(AccumuloPropertyNames.PROP_3, 1);
        edge2.putProperty(AccumuloPropertyNames.PROP_4, 1);
        final Edge edge3 = new Edge(TestGroups.EDGE);
        edge3.setSource("3");
        edge3.setDestination("8");
        edge3.setDirected(true);
        edge3.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 1);
        edge3.putProperty(AccumuloPropertyNames.PROP_1, 0);
        edge3.putProperty(AccumuloPropertyNames.PROP_2, 0);
        edge3.putProperty(AccumuloPropertyNames.PROP_3, 0);
        edge3.putProperty(AccumuloPropertyNames.PROP_4, 0);
        final Edge edge6 = new Edge("BasicEdge2");
        edge6.setSource("1");
        edge6.setDestination("5");
        edge6.setDirected(true);
        edge6.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 2);
        edge6.putProperty(AccumuloPropertyNames.PROP_1, 0);
        edge6.putProperty(AccumuloPropertyNames.PROP_2, 0);
        edge6.putProperty(AccumuloPropertyNames.PROP_3, 0);
        edge6.putProperty(AccumuloPropertyNames.PROP_4, 0);
        final Edge edge7 = new Edge("BasicEdge2");
        edge7.setSource("2");
        edge7.setDestination("6");
        edge7.setDirected(true);
        edge7.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 1);
        edge7.putProperty(AccumuloPropertyNames.PROP_1, 0);
        edge7.putProperty(AccumuloPropertyNames.PROP_2, 0);
        edge7.putProperty(AccumuloPropertyNames.PROP_3, 0);
        edge7.putProperty(AccumuloPropertyNames.PROP_4, 0);
        final Edge edge8 = new Edge("BasicEdge2");
        edge8.setSource("4");
        edge8.setDestination("8");
        edge8.setDirected(true);
        edge8.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 2);
        edge8.putProperty(AccumuloPropertyNames.PROP_1, 0);
        edge8.putProperty(AccumuloPropertyNames.PROP_2, 0);
        edge8.putProperty(AccumuloPropertyNames.PROP_3, 0);
        edge8.putProperty(AccumuloPropertyNames.PROP_4, 0);
        final Edge edge9 = new Edge("BasicEdge2");
        edge9.setSource("5");
        edge9.setDestination("9");
        edge9.setDirected(true);
        edge9.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 2);
        edge9.putProperty(AccumuloPropertyNames.PROP_1, 0);
        edge9.putProperty(AccumuloPropertyNames.PROP_2, 0);
        edge9.putProperty(AccumuloPropertyNames.PROP_3, 0);
        edge9.putProperty(AccumuloPropertyNames.PROP_4, 0);
        // Accumulo key
        final Key key = elementConverter.getKeysFromEdge(edge).getFirst();
        final Key key2 = elementConverter.getKeysFromEdge(edge2).getFirst();
        final Key key3 = elementConverter.getKeysFromEdge(edge3).getFirst();
        final Key key4 = elementConverter.getKeysFromEdge(edge6).getFirst();
        final Key key5 = elementConverter.getKeysFromEdge(edge7).getFirst();
        final Key key6 = elementConverter.getKeysFromEdge(edge8).getFirst();
        final Key key7 = elementConverter.getKeysFromEdge(edge9).getFirst();
        // Accumulo values
        final Value value1 = elementConverter.getValueFromProperties(TestGroups.EDGE, properties1);
        final Value value2 = elementConverter.getValueFromProperties(TestGroups.EDGE, properties2);
        final Value value3 = elementConverter.getValueFromProperties(TestGroups.EDGE, properties3);
        final Value value4 = elementConverter.getValueFromProperties(TestGroups.EDGE_2, properties1);
        final Value value5 = elementConverter.getValueFromProperties(TestGroups.EDGE_2, properties2);
        //Create mutation
        final Mutation m1 = new Mutation(key.getRow());
        m1.put(key.getColumnFamily(), key.getColumnQualifier(), new ColumnVisibility(key.getColumnVisibility()), key.getTimestamp(), value1);
        final Mutation m2 = new Mutation(key.getRow());
        m2.put(key.getColumnFamily(), key.getColumnQualifier(), new ColumnVisibility(key.getColumnVisibility()), key.getTimestamp(), value2);
        final Mutation m3 = new Mutation(key.getRow());
        m3.put(key.getColumnFamily(), key.getColumnQualifier(), new ColumnVisibility(key.getColumnVisibility()), key.getTimestamp(), value3);
        final Mutation m4 = new Mutation(key2.getRow());
        m4.put(key2.getColumnFamily(), key2.getColumnQualifier(), new ColumnVisibility(key2.getColumnVisibility()), key2.getTimestamp(), value1);
        final Mutation m5 = new Mutation(key.getRow());
        m5.put(key3.getColumnFamily(), key3.getColumnQualifier(), new ColumnVisibility(key3.getColumnVisibility()), key3.getTimestamp(), value1);
        final Mutation m6 = new Mutation(key4.getRow());
        m6.put(key4.getColumnFamily(), key4.getColumnQualifier(), new ColumnVisibility(key4.getColumnVisibility()), key4.getTimestamp(), value4);
        final Mutation m7 = new Mutation(key5.getRow());
        m7.put(key5.getColumnFamily(), key5.getColumnQualifier(), new ColumnVisibility(key5.getColumnVisibility()), key5.getTimestamp(), value5);
        final Mutation m8 = new Mutation(key6.getRow());
        m8.put(key6.getColumnFamily(), key6.getColumnQualifier(), new ColumnVisibility(key6.getColumnVisibility()), key6.getTimestamp(), value5);
        final Mutation m9 = new Mutation(key7.getRow());
        m9.put(key7.getColumnFamily(), key7.getColumnQualifier(), new ColumnVisibility(key7.getColumnVisibility()), key7.getTimestamp(), value5);
        // Write mutation
        final BatchWriterConfig writerConfig = new BatchWriterConfig();
        writerConfig.setMaxMemory(1000000L);
        writerConfig.setMaxLatency(1000L, TimeUnit.MILLISECONDS);
        writerConfig.setMaxWriteThreads(1);
        final BatchWriter writer = store.getConnection().createBatchWriter(store.getProperties().getTable(), writerConfig);
        writer.addMutation(m1);
        writer.addMutation(m2);
        writer.addMutation(m3);
        writer.addMutation(m4);
        writer.addMutation(m5);
        writer.addMutation(m6);
        writer.addMutation(m7);
        writer.addMutation(m8);
        writer.addMutation(m9);
        writer.close();
        // Read data back and check we get one merged element
        final Authorizations authorizations = new Authorizations(visibilityString);
        final BatchScanner scanner = store.getConnection().createBatchScanner(store.getProperties().getTable(), authorizations, 1000);
        try {
            scanner.addScanIterator(store.getKeyPackage().getIteratorFactory().getRowIDAggregatorIteratorSetting(store, "BasicEdge2"));
        } catch (IteratorSettingException e) {
            fail(e.getMessage());
        }
        final RangeFactory rangeF = store.getKeyPackage().getRangeFactory();
        final Range r = rangeF.getRangeFromPair(new Pair<ElementSeed>((new EntitySeed("1")), new EntitySeed("4")), new SummariseGroupOverRanges());
        final Range r2 = rangeF.getRangeFromPair(new Pair<ElementSeed>((new EntitySeed("5")), new EntitySeed("5")), new SummariseGroupOverRanges());
        scanner.setRanges(Arrays.asList(r, r2));
        final Iterator<Entry<Key, Value>> it = scanner.iterator();
        Entry<Key, Value> entry = it.next();
        Element readEdge = elementConverter.getFullElement(entry.getKey(), entry.getValue());
        Edge expectedEdge = new Edge("BasicEdge2");
        expectedEdge.setSource("4");
        expectedEdge.setDestination("8");
        expectedEdge.setDirected(true);
        expectedEdge.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 5);
        expectedEdge.putProperty(AccumuloPropertyNames.COUNT, 3);
        assertEquals(expectedEdge, readEdge);
        assertEquals(5, readEdge.getProperty(AccumuloPropertyNames.COLUMN_QUALIFIER));
        assertEquals(3, readEdge.getProperty(AccumuloPropertyNames.COUNT));
        // Check we get the Result of the second provided range
        assertTrue(it.hasNext());
        entry = it.next();
        readEdge = elementConverter.getFullElement(entry.getKey(), entry.getValue());
        expectedEdge = new Edge("BasicEdge2");
        expectedEdge.setSource("5");
        expectedEdge.setDestination("9");
        expectedEdge.setDirected(true);
        expectedEdge.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 2);
        expectedEdge.putProperty(AccumuloPropertyNames.COUNT, 1);
        assertEquals(expectedEdge, readEdge);
        //Check no additional rows are found. (For a table of this size we shouldn't see this)
        if (it.hasNext()) {
            fail("Additional row found.");
        }
    } catch (AccumuloException | TableExistsException | TableNotFoundException e) {
        fail(this.getClass().getSimpleName() + " failed with exception: " + e);
    }
}
Also used : Element(uk.gov.gchq.gaffer.data.element.Element) BatchScanner(org.apache.accumulo.core.client.BatchScanner) Properties(uk.gov.gchq.gaffer.data.element.Properties) AccumuloProperties(uk.gov.gchq.gaffer.accumulostore.AccumuloProperties) RangeFactory(uk.gov.gchq.gaffer.accumulostore.key.RangeFactory) TableNotFoundException(org.apache.accumulo.core.client.TableNotFoundException) Entry(java.util.Map.Entry) SummariseGroupOverRanges(uk.gov.gchq.gaffer.accumulostore.operation.impl.SummariseGroupOverRanges) BatchWriterConfig(org.apache.accumulo.core.client.BatchWriterConfig) ColumnVisibility(org.apache.accumulo.core.security.ColumnVisibility) AccumuloException(org.apache.accumulo.core.client.AccumuloException) Authorizations(org.apache.accumulo.core.security.Authorizations) IteratorSettingException(uk.gov.gchq.gaffer.accumulostore.key.exception.IteratorSettingException) Range(org.apache.accumulo.core.data.Range) Value(org.apache.accumulo.core.data.Value) EntitySeed(uk.gov.gchq.gaffer.operation.data.EntitySeed) TableExistsException(org.apache.accumulo.core.client.TableExistsException) ElementSeed(uk.gov.gchq.gaffer.operation.data.ElementSeed) Mutation(org.apache.accumulo.core.data.Mutation) BatchWriter(org.apache.accumulo.core.client.BatchWriter) Edge(uk.gov.gchq.gaffer.data.element.Edge) Key(org.apache.accumulo.core.data.Key)

Example 25 with Mutation

use of org.apache.accumulo.core.data.Mutation in project Gaffer by gchq.

the class CoreKeyGroupByAggregatorIteratorTest method testAggregatingEmptyColumnQualifier.

public void testAggregatingEmptyColumnQualifier(final AccumuloStore store, final AccumuloElementConverter elementConverter) throws StoreException, AccumuloElementConversionException {
    final String visibilityString = "public";
    try {
        // Create edge
        final Edge edge = new Edge(TestGroups.EDGE);
        edge.setSource("1");
        edge.setDestination("2");
        edge.setDirected(true);
        edge.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 8);
        edge.putProperty(AccumuloPropertyNames.PROP_1, 0);
        edge.putProperty(AccumuloPropertyNames.PROP_2, 0);
        edge.putProperty(AccumuloPropertyNames.PROP_3, 0);
        edge.putProperty(AccumuloPropertyNames.PROP_4, 0);
        edge.putProperty(AccumuloPropertyNames.COUNT, 1);
        //THIS EDGE WILL BE REDUCED MEANING ITS CQ (columnQualifier) will only occur once because its key is equal.
        final Edge edge2 = new Edge(TestGroups.EDGE);
        edge2.setSource("1");
        edge2.setDestination("2");
        edge2.setDirected(true);
        edge2.putProperty(AccumuloPropertyNames.PROP_1, 0);
        edge2.putProperty(AccumuloPropertyNames.PROP_2, 0);
        edge2.putProperty(AccumuloPropertyNames.PROP_3, 0);
        edge2.putProperty(AccumuloPropertyNames.PROP_4, 0);
        edge2.putProperty(AccumuloPropertyNames.COUNT, 2);
        final Edge edge3 = new Edge(TestGroups.EDGE);
        edge3.setSource("1");
        edge3.setDestination("2");
        edge3.setDirected(true);
        edge3.putProperty(AccumuloPropertyNames.PROP_1, 0);
        edge3.putProperty(AccumuloPropertyNames.PROP_2, 0);
        edge3.putProperty(AccumuloPropertyNames.PROP_3, 0);
        edge3.putProperty(AccumuloPropertyNames.PROP_4, 0);
        edge3.putProperty(AccumuloPropertyNames.COUNT, 10);
        // Accumulo key
        final Key key = elementConverter.getKeysFromEdge(edge).getFirst();
        final Key key2 = elementConverter.getKeysFromEdge(edge2).getFirst();
        final Key key3 = elementConverter.getKeysFromEdge(edge3).getFirst();
        // Accumulo values
        final Value value1 = elementConverter.getValueFromProperties(TestGroups.EDGE, edge.getProperties());
        final Value value2 = elementConverter.getValueFromProperties(TestGroups.EDGE, edge2.getProperties());
        final Value value3 = elementConverter.getValueFromProperties(TestGroups.EDGE, edge3.getProperties());
        // Create mutation
        final Mutation m1 = new Mutation(key.getRow());
        m1.put(key.getColumnFamily(), key.getColumnQualifier(), new ColumnVisibility(key.getColumnVisibility()), key.getTimestamp(), value1);
        final Mutation m2 = new Mutation(key.getRow());
        m2.put(key.getColumnFamily(), key.getColumnQualifier(), new ColumnVisibility(key.getColumnVisibility()), key.getTimestamp(), value2);
        final Mutation m3 = new Mutation(key.getRow());
        m3.put(key.getColumnFamily(), key.getColumnQualifier(), new ColumnVisibility(key.getColumnVisibility()), key.getTimestamp(), value3);
        final Mutation m4 = new Mutation(key2.getRow());
        m4.put(key2.getColumnFamily(), key2.getColumnQualifier(), new ColumnVisibility(key2.getColumnVisibility()), key2.getTimestamp(), value1);
        final Mutation m5 = new Mutation(key.getRow());
        m5.put(key3.getColumnFamily(), key3.getColumnQualifier(), new ColumnVisibility(key3.getColumnVisibility()), key3.getTimestamp(), value1);
        // Write mutation
        final BatchWriterConfig writerConfig = new BatchWriterConfig();
        writerConfig.setMaxMemory(1000000L);
        writerConfig.setMaxLatency(1000L, TimeUnit.MILLISECONDS);
        writerConfig.setMaxWriteThreads(1);
        final BatchWriter writer = store.getConnection().createBatchWriter(store.getProperties().getTable(), writerConfig);
        writer.addMutation(m1);
        writer.addMutation(m2);
        writer.addMutation(m3);
        writer.addMutation(m4);
        writer.addMutation(m5);
        writer.close();
        Edge expectedEdge = new Edge(TestGroups.EDGE);
        expectedEdge.setSource("1");
        expectedEdge.setDestination("2");
        expectedEdge.setDirected(true);
        expectedEdge.putProperty(AccumuloPropertyNames.COLUMN_QUALIFIER, 8);
        expectedEdge.putProperty(AccumuloPropertyNames.COUNT, 15);
        expectedEdge.putProperty(AccumuloPropertyNames.PROP_1, 0);
        expectedEdge.putProperty(AccumuloPropertyNames.PROP_2, 0);
        expectedEdge.putProperty(AccumuloPropertyNames.PROP_3, 0);
        expectedEdge.putProperty(AccumuloPropertyNames.PROP_4, 0);
        // Read data back and check we get one merged element
        final Authorizations authorizations = new Authorizations(visibilityString);
        final Scanner scanner = store.getConnection().createScanner(store.getProperties().getTable(), authorizations);
        final IteratorSetting iteratorSetting = new IteratorSettingBuilder(AccumuloStoreConstants.COLUMN_QUALIFIER_AGGREGATOR_ITERATOR_PRIORITY, "KeyCombiner", CoreKeyGroupByAggregatorIterator.class).all().view(new View.Builder().edge(TestGroups.EDGE, new ViewElementDefinition.Builder().groupBy().build()).build()).schema(store.getSchema()).keyConverter(store.getKeyPackage().getKeyConverter()).build();
        scanner.addScanIterator(iteratorSetting);
        final Iterator<Entry<Key, Value>> it = scanner.iterator();
        final Entry<Key, Value> entry = it.next();
        final Element readEdge = elementConverter.getFullElement(entry.getKey(), entry.getValue());
        assertEquals(expectedEdge, readEdge);
        assertEquals(8, readEdge.getProperty(AccumuloPropertyNames.COLUMN_QUALIFIER));
        assertEquals(15, readEdge.getProperty(AccumuloPropertyNames.COUNT));
        // Check no more entries
        if (it.hasNext()) {
            fail("Additional row found.");
        }
    } catch (AccumuloException | TableNotFoundException e) {
        fail(this.getClass().getSimpleName() + " failed with exception: " + e);
    }
}
Also used : Scanner(org.apache.accumulo.core.client.Scanner) AccumuloException(org.apache.accumulo.core.client.AccumuloException) Authorizations(org.apache.accumulo.core.security.Authorizations) IteratorSettingBuilder(uk.gov.gchq.gaffer.accumulostore.utils.IteratorSettingBuilder) Element(uk.gov.gchq.gaffer.data.element.Element) IteratorSettingBuilder(uk.gov.gchq.gaffer.accumulostore.utils.IteratorSettingBuilder) TableNotFoundException(org.apache.accumulo.core.client.TableNotFoundException) Entry(java.util.Map.Entry) IteratorSetting(org.apache.accumulo.core.client.IteratorSetting) Value(org.apache.accumulo.core.data.Value) BatchWriterConfig(org.apache.accumulo.core.client.BatchWriterConfig) Mutation(org.apache.accumulo.core.data.Mutation) ColumnVisibility(org.apache.accumulo.core.security.ColumnVisibility) BatchWriter(org.apache.accumulo.core.client.BatchWriter) Edge(uk.gov.gchq.gaffer.data.element.Edge) Key(org.apache.accumulo.core.data.Key)

Aggregations

Mutation (org.apache.accumulo.core.data.Mutation)31 Value (org.apache.accumulo.core.data.Value)21 Key (org.apache.accumulo.core.data.Key)17 ColumnVisibility (org.apache.accumulo.core.security.ColumnVisibility)16 Authorizations (org.apache.accumulo.core.security.Authorizations)15 BatchWriter (org.apache.accumulo.core.client.BatchWriter)13 Entry (java.util.Map.Entry)12 BatchWriterConfig (org.apache.accumulo.core.client.BatchWriterConfig)12 Test (org.junit.Test)11 ByteArrayRef (org.apache.hadoop.hive.serde2.lazy.ByteArrayRef)10 Text (org.apache.hadoop.io.Text)10 Properties (java.util.Properties)8 TableNotFoundException (org.apache.accumulo.core.client.TableNotFoundException)8 Configuration (org.apache.hadoop.conf.Configuration)8 LazySerDeParameters (org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters)8 LazyStruct (org.apache.hadoop.hive.serde2.lazy.LazyStruct)8 LazySimpleStructObjectInspector (org.apache.hadoop.hive.serde2.lazy.objectinspector.LazySimpleStructObjectInspector)8 TypeInfo (org.apache.hadoop.hive.serde2.typeinfo.TypeInfo)8 Element (uk.gov.gchq.gaffer.data.element.Element)8 AccumuloException (org.apache.accumulo.core.client.AccumuloException)7