Search in sources :

Example 71 with SerDeException

use of org.apache.hadoop.hive.serde2.SerDeException in project haivvreo by jghoman.

the class AvroObjectInspectorGenerator method createObjectInspectorWorker.

private ObjectInspector createObjectInspectorWorker(TypeInfo ti) throws SerDeException {
    // at deserialization and the object inspector will never see the actual union.
    if (!supportedCategories(ti))
        throw new HaivvreoException("Don't yet support this type: " + ti);
    ObjectInspector result;
    switch(ti.getCategory()) {
        case PRIMITIVE:
            PrimitiveTypeInfo pti = (PrimitiveTypeInfo) ti;
            result = PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(pti.getPrimitiveCategory());
            break;
        case STRUCT:
            StructTypeInfo sti = (StructTypeInfo) ti;
            ArrayList<ObjectInspector> ois = new ArrayList<ObjectInspector>(sti.getAllStructFieldTypeInfos().size());
            for (TypeInfo typeInfo : sti.getAllStructFieldTypeInfos()) {
                ois.add(createObjectInspectorWorker(typeInfo));
            }
            result = ObjectInspectorFactory.getStandardStructObjectInspector(sti.getAllStructFieldNames(), ois);
            break;
        case MAP:
            MapTypeInfo mti = (MapTypeInfo) ti;
            result = ObjectInspectorFactory.getStandardMapObjectInspector(PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(PrimitiveObjectInspector.PrimitiveCategory.STRING), createObjectInspectorWorker(mti.getMapValueTypeInfo()));
            break;
        case LIST:
            ListTypeInfo ati = (ListTypeInfo) ti;
            result = ObjectInspectorFactory.getStandardListObjectInspector(createObjectInspectorWorker(ati.getListElementTypeInfo()));
            break;
        case UNION:
            UnionTypeInfo uti = (UnionTypeInfo) ti;
            List<TypeInfo> allUnionObjectTypeInfos = uti.getAllUnionObjectTypeInfos();
            List<ObjectInspector> unionObjectInspectors = new ArrayList<ObjectInspector>(allUnionObjectTypeInfos.size());
            for (TypeInfo typeInfo : allUnionObjectTypeInfos) {
                unionObjectInspectors.add(createObjectInspectorWorker(typeInfo));
            }
            result = ObjectInspectorFactory.getStandardUnionObjectInspector(unionObjectInspectors);
            break;
        default:
            throw new HaivvreoException("No Hive categories matched: " + ti);
    }
    return result;
}
Also used : PrimitiveObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.PrimitiveObjectInspector) ObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector) ArrayList(java.util.ArrayList)

Example 72 with SerDeException

use of org.apache.hadoop.hive.serde2.SerDeException in project haivvreo by jghoman.

the class TestAvroSerializer method serializeAndDeserialize.

/**
   * Verify that we can serialize an avro value by taking one, running it through
   * the deser process and then serialize it again.
   */
private GenericRecord serializeAndDeserialize(String recordValue, String fieldName, Object fieldValue) throws SerDeException, IOException {
    Schema s = buildSchema(recordValue);
    GenericData.Record r = new GenericData.Record(s);
    r.put(fieldName, fieldValue);
    AvroSerializer as = new AvroSerializer();
    AvroDeserializer ad = new AvroDeserializer();
    AvroObjectInspectorGenerator aoig = new AvroObjectInspectorGenerator(s);
    ObjectInspector oi = aoig.getObjectInspector();
    List<String> columnNames = aoig.getColumnNames();
    List<TypeInfo> columnTypes = aoig.getColumnTypes();
    AvroGenericRecordWritable agrw = Utils.serializeAndDeserializeRecord(r);
    Object obj = ad.deserialize(columnNames, columnTypes, agrw, s);
    Writable result = as.serialize(obj, oi, columnNames, columnTypes, s);
    assertTrue(result instanceof AvroGenericRecordWritable);
    GenericRecord r2 = ((AvroGenericRecordWritable) result).getRecord();
    assertEquals(s, r2.getSchema());
    return r2;
}
Also used : ObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector) Schema(org.apache.avro.Schema) Writable(org.apache.hadoop.io.Writable) GenericData(org.apache.avro.generic.GenericData) TypeInfo(org.apache.hadoop.hive.serde2.typeinfo.TypeInfo) GenericRecord(org.apache.avro.generic.GenericRecord) GenericRecord(org.apache.avro.generic.GenericRecord)

Example 73 with SerDeException

use of org.apache.hadoop.hive.serde2.SerDeException in project haivvreo by jghoman.

the class TestAvroObjectInspectorGenerator method primitiveTypesWorkCorrectly.

@Test
public void primitiveTypesWorkCorrectly() throws SerDeException {
    final String bunchOfPrimitives = "{\n" + "  \"namespace\": \"testing\",\n" + "  \"name\": \"PrimitiveTypes\",\n" + "  \"type\": \"record\",\n" + "  \"fields\": [\n" + "    {\n" + "      \"name\":\"aString\",\n" + "      \"type\":\"string\"\n" + "    },\n" + "    {\n" + "      \"name\":\"anInt\",\n" + "      \"type\":\"int\"\n" + "    },\n" + "    {\n" + "      \"name\":\"aBoolean\",\n" + "      \"type\":\"boolean\"\n" + "    },\n" + "    {\n" + "      \"name\":\"aLong\",\n" + "      \"type\":\"long\"\n" + "    },\n" + "    {\n" + "      \"name\":\"aFloat\",\n" + "      \"type\":\"float\"\n" + "    },\n" + "    {\n" + "      \"name\":\"aDouble\",\n" + "      \"type\":\"double\"\n" + "    },\n" + "    {\n" + "      \"name\":\"aNull\",\n" + "      \"type\":\"null\"\n" + "    }\n" + "  ]\n" + "}";
    AvroObjectInspectorGenerator aoig = new AvroObjectInspectorGenerator(Schema.parse(bunchOfPrimitives));
    String[] expectedColumnNames = { "aString", "anInt", "aBoolean", "aLong", "aFloat", "aDouble", "aNull" };
    verifyColumnNames(expectedColumnNames, aoig.getColumnNames());
    TypeInfo[] expectedColumnTypes = { STRING, INT, BOOLEAN, LONG, FLOAT, DOUBLE, VOID };
    verifyColumnTypes(expectedColumnTypes, aoig.getColumnTypes());
    // Rip apart the object inspector, making sure we got what we expect.
    final ObjectInspector oi = aoig.getObjectInspector();
    assertTrue(oi instanceof StandardStructObjectInspector);
    final StandardStructObjectInspector ssoi = (StandardStructObjectInspector) oi;
    List<? extends StructField> structFields = ssoi.getAllStructFieldRefs();
    assertEquals(expectedColumnNames.length, structFields.size());
    for (int i = 0; i < expectedColumnNames.length; i++) {
        assertEquals("Column names don't match", expectedColumnNames[i].toLowerCase(), structFields.get(i).getFieldName());
        assertEquals("Column types don't match", expectedColumnTypes[i].getTypeName(), structFields.get(i).getFieldObjectInspector().getTypeName());
    }
}
Also used : PrimitiveObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.PrimitiveObjectInspector) ObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector) StandardStructObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.StandardStructObjectInspector) AvroObjectInspectorGenerator(com.linkedin.haivvreo.AvroObjectInspectorGenerator) StandardStructObjectInspector(org.apache.hadoop.hive.serde2.objectinspector.StandardStructObjectInspector) Test(org.junit.Test)

Example 74 with SerDeException

use of org.apache.hadoop.hive.serde2.SerDeException in project haivvreo by jghoman.

the class TestAvroObjectInspectorGenerator method failOnNonRecords.

// that we can only process records
@Test
public void failOnNonRecords() throws Exception {
    String nonRecordSchema = "{ \"type\": \"enum\",\n" + "  \"name\": \"Suit\",\n" + "  \"symbols\" : [\"SPADES\", \"HEARTS\", \"DIAMONDS\", \"CLUBS\"]\n" + "}";
    Schema s = Schema.parse(nonRecordSchema);
    try {
        new AvroObjectInspectorGenerator(s);
        fail("Should not be able to handle non-record Avro types");
    } catch (SerDeException sde) {
        assertTrue(sde.getMessage().startsWith("Schema for table must be of type RECORD"));
    }
}
Also used : Schema(org.apache.avro.Schema) AvroObjectInspectorGenerator(com.linkedin.haivvreo.AvroObjectInspectorGenerator) SerDeException(org.apache.hadoop.hive.serde2.SerDeException) Test(org.junit.Test)

Example 75 with SerDeException

use of org.apache.hadoop.hive.serde2.SerDeException in project hive by apache.

the class TestAccumuloRowSerializer method testBinarySerialization.

@Test
public void testBinarySerialization() 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(), 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()));
    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()));
    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("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)

Aggregations

SerDeException (org.apache.hadoop.hive.serde2.SerDeException)124 StructObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.StructObjectInspector)108 ObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.ObjectInspector)100 ArrayList (java.util.ArrayList)98 Properties (java.util.Properties)59 Test (org.junit.Test)59 Configuration (org.apache.hadoop.conf.Configuration)52 TypeInfo (org.apache.hadoop.hive.serde2.typeinfo.TypeInfo)52 Text (org.apache.hadoop.io.Text)50 IOException (java.io.IOException)37 ListObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.ListObjectInspector)33 Schema (org.apache.avro.Schema)31 StructField (org.apache.hadoop.hive.serde2.objectinspector.StructField)31 MapObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.MapObjectInspector)28 PrimitiveObjectInspector (org.apache.hadoop.hive.serde2.objectinspector.PrimitiveObjectInspector)28 PrimitiveTypeInfo (org.apache.hadoop.hive.serde2.typeinfo.PrimitiveTypeInfo)24 Put (org.apache.hadoop.hbase.client.Put)22 LazySerDeParameters (org.apache.hadoop.hive.serde2.lazy.LazySerDeParameters)22 IntWritable (org.apache.hadoop.io.IntWritable)22 HiveException (org.apache.hadoop.hive.ql.metadata.HiveException)21