use of org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory.stringTypeInfo in project hive by apache.
the class TestAccumuloPredicateHandler method testBinaryRangeGeneration.
@Test
public void testBinaryRangeGeneration() throws Exception {
List<String> columnNames = Arrays.asList("key", "column");
List<TypeInfo> columnTypes = Arrays.<TypeInfo>asList(TypeInfoFactory.intTypeInfo, TypeInfoFactory.stringTypeInfo);
conf.set(serdeConstants.LIST_COLUMNS, Joiner.on(',').join(columnNames));
conf.set(serdeConstants.LIST_COLUMN_TYPES, "int,string");
String columnMappingStr = ":rowID#b,cf:f1";
conf.set(AccumuloSerDeParameters.COLUMN_MAPPINGS, columnMappingStr);
columnMapper = new ColumnMapper(columnMappingStr, ColumnEncoding.STRING.getName(), columnNames, columnTypes);
int intValue = 100;
// Make binary integer value in the bytearray
ByteArrayOutputStream baos = new ByteArrayOutputStream();
JavaIntObjectInspector intOI = (JavaIntObjectInspector) PrimitiveObjectInspectorFactory.getPrimitiveJavaObjectInspector(TypeInfoFactory.getPrimitiveTypeInfo(serdeConstants.INT_TYPE_NAME));
LazyUtils.writePrimitive(baos, intValue, intOI);
// 100 < key
ExprNodeDesc column = new ExprNodeColumnDesc(TypeInfoFactory.intTypeInfo, "key", null, false);
ExprNodeDesc constant = new ExprNodeConstantDesc(TypeInfoFactory.intTypeInfo, intValue);
List<ExprNodeDesc> children = Lists.newArrayList();
children.add(constant);
children.add(column);
ExprNodeGenericFuncDesc node = new ExprNodeGenericFuncDesc(TypeInfoFactory.stringTypeInfo, new GenericUDFOPLessThan(), children);
assertNotNull(node);
String filterExpr = SerializationUtilities.serializeExpression(node);
conf.set(TableScanDesc.FILTER_EXPR_CONF_STR, filterExpr);
// Should make (100, +inf)
List<Range> ranges = handler.getRanges(conf, columnMapper);
Assert.assertEquals(1, ranges.size());
Assert.assertEquals(new Range(new Text(baos.toByteArray()), false, null, false), ranges.get(0));
}
use of org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory.stringTypeInfo in project hive by apache.
the class TestAccumuloPredicateHandler method testIgnoreIteratorPushdown.
@Test
public void testIgnoreIteratorPushdown() throws TooManyAccumuloColumnsException {
// Override what's placed in the Configuration by setup()
conf = new JobConf();
List<String> columnNames = Arrays.asList("field1", "field2", "rid");
List<TypeInfo> columnTypes = Arrays.<TypeInfo>asList(TypeInfoFactory.stringTypeInfo, TypeInfoFactory.intTypeInfo, TypeInfoFactory.stringTypeInfo);
conf.set(serdeConstants.LIST_COLUMNS, Joiner.on(',').join(columnNames));
conf.set(serdeConstants.LIST_COLUMN_TYPES, "string,int,string");
String columnMappingStr = "cf:f1,cf:f2,:rowID";
conf.set(AccumuloSerDeParameters.COLUMN_MAPPINGS, columnMappingStr);
columnMapper = new ColumnMapper(columnMappingStr, ColumnEncoding.STRING.getName(), columnNames, columnTypes);
ExprNodeDesc column = new ExprNodeColumnDesc(TypeInfoFactory.stringTypeInfo, "field1", null, false);
ExprNodeDesc constant = new ExprNodeConstantDesc(TypeInfoFactory.stringTypeInfo, "aaa");
List<ExprNodeDesc> children = Lists.newArrayList();
children.add(column);
children.add(constant);
ExprNodeDesc node = new ExprNodeGenericFuncDesc(TypeInfoFactory.stringTypeInfo, new GenericUDFOPEqualOrLessThan(), children);
assertNotNull(node);
ExprNodeDesc column2 = new ExprNodeColumnDesc(TypeInfoFactory.intTypeInfo, "field2", null, false);
ExprNodeDesc constant2 = new ExprNodeConstantDesc(TypeInfoFactory.intTypeInfo, 5);
List<ExprNodeDesc> children2 = Lists.newArrayList();
children2.add(column2);
children2.add(constant2);
ExprNodeDesc node2 = new ExprNodeGenericFuncDesc(TypeInfoFactory.stringTypeInfo, new GenericUDFOPGreaterThan(), children2);
assertNotNull(node2);
List<ExprNodeDesc> bothFilters = Lists.newArrayList();
bothFilters.add(node);
bothFilters.add(node2);
ExprNodeGenericFuncDesc both = new ExprNodeGenericFuncDesc(TypeInfoFactory.stringTypeInfo, new GenericUDFOPAnd(), bothFilters);
String filterExpr = SerializationUtilities.serializeExpression(both);
conf.set(TableScanDesc.FILTER_EXPR_CONF_STR, filterExpr);
conf.setBoolean(AccumuloSerDeParameters.ITERATOR_PUSHDOWN_KEY, false);
try {
List<IteratorSetting> iterators = handler.getIterators(conf, columnMapper);
assertEquals(iterators.size(), 0);
} catch (Exception e) {
fail(StringUtils.stringifyException(e));
}
}
use of org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory.stringTypeInfo in project hive by apache.
the class TestAccumuloPredicateHandler method testIteratorIgnoreRowIDFields.
@Test
public void testIteratorIgnoreRowIDFields() {
ExprNodeDesc column = new ExprNodeColumnDesc(TypeInfoFactory.stringTypeInfo, "rid", null, false);
ExprNodeDesc constant = new ExprNodeConstantDesc(TypeInfoFactory.stringTypeInfo, "aaa");
List<ExprNodeDesc> children = Lists.newArrayList();
children.add(column);
children.add(constant);
ExprNodeDesc node = new ExprNodeGenericFuncDesc(TypeInfoFactory.stringTypeInfo, new GenericUDFOPEqualOrLessThan(), children);
assertNotNull(node);
ExprNodeDesc column2 = new ExprNodeColumnDesc(TypeInfoFactory.stringTypeInfo, "rid", null, false);
ExprNodeDesc constant2 = new ExprNodeConstantDesc(TypeInfoFactory.stringTypeInfo, "bbb");
List<ExprNodeDesc> children2 = Lists.newArrayList();
children2.add(column2);
children2.add(constant2);
ExprNodeDesc node2 = new ExprNodeGenericFuncDesc(TypeInfoFactory.stringTypeInfo, new GenericUDFOPGreaterThan(), children2);
assertNotNull(node2);
List<ExprNodeDesc> bothFilters = Lists.newArrayList();
bothFilters.add(node);
bothFilters.add(node2);
ExprNodeGenericFuncDesc both = new ExprNodeGenericFuncDesc(TypeInfoFactory.stringTypeInfo, new GenericUDFOPAnd(), bothFilters);
String filterExpr = SerializationUtilities.serializeExpression(both);
conf.set(TableScanDesc.FILTER_EXPR_CONF_STR, filterExpr);
try {
List<IteratorSetting> iterators = handler.getIterators(conf, columnMapper);
assertEquals(iterators.size(), 0);
} catch (SerDeException e) {
StringUtils.stringifyException(e);
}
}
use of org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory.stringTypeInfo in project hive by apache.
the class TestAccumuloPredicateHandler method setup.
@Before
public void setup() throws TooManyAccumuloColumnsException {
FunctionRegistry.getFunctionNames();
conf = new JobConf();
List<String> columnNames = Arrays.asList("field1", "rid");
List<TypeInfo> columnTypes = Arrays.<TypeInfo>asList(TypeInfoFactory.stringTypeInfo, TypeInfoFactory.stringTypeInfo);
conf.set(serdeConstants.LIST_COLUMNS, Joiner.on(',').join(columnNames));
conf.set(serdeConstants.LIST_COLUMN_TYPES, "string,string");
String columnMappingStr = "cf:f1,:rowID";
conf.set(AccumuloSerDeParameters.COLUMN_MAPPINGS, columnMappingStr);
columnMapper = new ColumnMapper(columnMappingStr, ColumnEncoding.STRING.getName(), columnNames, columnTypes);
}
use of org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory.stringTypeInfo 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()));
}
Aggregations