Search in sources :

Example 6 with ValueSet

use of com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet in project aws-athena-query-federation by awslabs.

the class ReadRecordsRequestSerDeTest method beforeTest.

@Before
public void beforeTest() throws IOException {
    String yearCol = "year";
    String monthCol = "month";
    String dayCol = "day";
    Schema schema = SchemaBuilder.newBuilder().addField(yearCol, new ArrowType.Int(32, true)).addField(monthCol, new ArrowType.Int(32, true)).addField(dayCol, new ArrowType.Int(32, true)).addField("col2", new ArrowType.Utf8()).addField("col3", Types.MinorType.FLOAT8.getType()).addField("col4", Types.MinorType.FLOAT8.getType()).addField("col5", Types.MinorType.FLOAT8.getType()).build();
    Map<String, ValueSet> constraintsMap = new HashMap<>();
    constraintsMap.put("col3", SortedRangeSet.copyOf(Types.MinorType.FLOAT8.getType(), ImmutableList.of(Range.greaterThan(allocator, Types.MinorType.FLOAT8.getType(), -10000D)), false));
    constraintsMap.put("col4", EquatableValueSet.newBuilder(allocator, Types.MinorType.FLOAT8.getType(), false, true).add(1.1D).build());
    constraintsMap.put("col5", new AllOrNoneValueSet(Types.MinorType.FLOAT8.getType(), false, true));
    Constraints constraints = new Constraints(constraintsMap);
    Block partitions = allocator.createBlock(schema);
    int num_partitions = 10;
    for (int i = 0; i < num_partitions; i++) {
        BlockUtils.setValue(partitions.getFieldVector(yearCol), i, 2016 + i);
        BlockUtils.setValue(partitions.getFieldVector(monthCol), i, (i % 12) + 1);
        BlockUtils.setValue(partitions.getFieldVector(dayCol), i, (i % 28) + 1);
    }
    partitions.setRowCount(num_partitions);
    SpillLocation spillLocation = S3SpillLocation.newBuilder().withBucket("athena-virtuoso-test").withPrefix("lambda-spill").withQueryId("test-query-id").withSplitId("test-split-id").withIsDirectory(true).build();
    EncryptionKey encryptionKey = new EncryptionKey("test-key".getBytes(), "test-nonce".getBytes());
    Split split = Split.newBuilder(spillLocation, encryptionKey).add("year", "2017").add("month", "11").add("day", "1").build();
    expected = new ReadRecordsRequest(federatedIdentity, "test-query-id", "test-catalog", new TableName("test-schema", "test-table"), schema, split, constraints, 100_000_000_000L, 100_000_000_000L);
    String expectedSerDeFile = utils.getResourceOrFail("serde/v2", "ReadRecordsRequest.json");
    expectedSerDeText = utils.readAllAsString(expectedSerDeFile).trim();
}
Also used : SpillLocation(com.amazonaws.athena.connector.lambda.domain.spill.SpillLocation) S3SpillLocation(com.amazonaws.athena.connector.lambda.domain.spill.S3SpillLocation) HashMap(java.util.HashMap) AllOrNoneValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.AllOrNoneValueSet) Schema(org.apache.arrow.vector.types.pojo.Schema) ArrowType(org.apache.arrow.vector.types.pojo.ArrowType) EncryptionKey(com.amazonaws.athena.connector.lambda.security.EncryptionKey) TableName(com.amazonaws.athena.connector.lambda.domain.TableName) Constraints(com.amazonaws.athena.connector.lambda.domain.predicate.Constraints) ReadRecordsRequest(com.amazonaws.athena.connector.lambda.records.ReadRecordsRequest) Block(com.amazonaws.athena.connector.lambda.data.Block) Split(com.amazonaws.athena.connector.lambda.domain.Split) EquatableValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.EquatableValueSet) ValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet) AllOrNoneValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.AllOrNoneValueSet) Before(org.junit.Before)

Example 7 with ValueSet

use of com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet in project aws-athena-query-federation by awslabs.

the class HiveRecordHandlerTest method buildSplitSql.

@Test
public void buildSplitSql() throws SQLException {
    TableName tableName = new TableName("testSchema", "testTable");
    SchemaBuilder schemaBuilder = SchemaBuilder.newBuilder();
    schemaBuilder.addField(FieldBuilder.newBuilder("testCol1", Types.MinorType.INT.getType()).build());
    schemaBuilder.addField(FieldBuilder.newBuilder("testCol2", Types.MinorType.DATEDAY.getType()).build());
    schemaBuilder.addField(FieldBuilder.newBuilder("testCol3", Types.MinorType.DATEMILLI.getType()).build());
    schemaBuilder.addField(FieldBuilder.newBuilder("testCol4", Types.MinorType.VARBINARY.getType()).build());
    schemaBuilder.addField(FieldBuilder.newBuilder("partition", Types.MinorType.VARCHAR.getType()).build());
    Schema schema = schemaBuilder.build();
    Split split = Mockito.mock(Split.class);
    Mockito.when(split.getProperties()).thenReturn(Collections.singletonMap("partition", "p0"));
    Mockito.when(split.getProperty(Mockito.eq("partition"))).thenReturn("p0");
    Range range1a = Mockito.mock(Range.class, Mockito.RETURNS_DEEP_STUBS);
    Mockito.when(range1a.isSingleValue()).thenReturn(true);
    Mockito.when(range1a.getLow().getValue()).thenReturn(1);
    Range range1b = Mockito.mock(Range.class, Mockito.RETURNS_DEEP_STUBS);
    Mockito.when(range1b.isSingleValue()).thenReturn(true);
    Mockito.when(range1b.getLow().getValue()).thenReturn(2);
    ValueSet valueSet1 = Mockito.mock(SortedRangeSet.class, Mockito.RETURNS_DEEP_STUBS);
    Mockito.when(valueSet1.getRanges().getOrderedRanges()).thenReturn(ImmutableList.of(range1a, range1b));
    final long dateDays = TimeUnit.DAYS.toDays(Date.valueOf("2020-01-05").getTime());
    ValueSet valueSet2 = getSingleValueSet(dateDays);
    Constraints constraints = Mockito.mock(Constraints.class);
    Mockito.when(constraints.getSummary()).thenReturn(new ImmutableMap.Builder<String, ValueSet>().put("testCol2", valueSet2).build());
    PreparedStatement expectedPreparedStatement = Mockito.mock(PreparedStatement.class);
    Mockito.when(this.connection.prepareStatement(Mockito.anyString())).thenReturn(expectedPreparedStatement);
    PreparedStatement preparedStatement = this.hiveRecordHandler.buildSplitSql(this.connection, "testCatalogName", tableName, schema, constraints, split);
    Assert.assertEquals(expectedPreparedStatement, preparedStatement);
}
Also used : TableName(com.amazonaws.athena.connector.lambda.domain.TableName) Constraints(com.amazonaws.athena.connector.lambda.domain.predicate.Constraints) Schema(org.apache.arrow.vector.types.pojo.Schema) SchemaBuilder(com.amazonaws.athena.connector.lambda.data.SchemaBuilder) PreparedStatement(java.sql.PreparedStatement) Split(com.amazonaws.athena.connector.lambda.domain.Split) Range(com.amazonaws.athena.connector.lambda.domain.predicate.Range) ValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet) ImmutableMap(com.google.common.collect.ImmutableMap) Test(org.junit.Test)

Example 8 with ValueSet

use of com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet in project aws-athena-query-federation by awslabs.

the class HiveRecordHandlerTest method getSingleValueSet.

private ValueSet getSingleValueSet(Object value) {
    Range range = Mockito.mock(Range.class, Mockito.RETURNS_DEEP_STUBS);
    Mockito.when(range.isSingleValue()).thenReturn(true);
    Mockito.when(range.getLow().getValue()).thenReturn(value);
    ValueSet valueSet = Mockito.mock(SortedRangeSet.class, Mockito.RETURNS_DEEP_STUBS);
    Mockito.when(valueSet.getRanges().getOrderedRanges()).thenReturn(Collections.singletonList(range));
    return valueSet;
}
Also used : Range(com.amazonaws.athena.connector.lambda.domain.predicate.Range) ValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet)

Example 9 with ValueSet

use of com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet in project aws-athena-query-federation by awslabs.

the class PropertyGraphHandler method getQueryPartForContraintsMap.

/**
 * Used to generate Gremlin Query part for Constraint Map
 *
 * @param traversal Gremlin Traversal, traversal is updated based on constraints
 *                  map
 * @param hasMap    Constraint Hash Map
 *
 * @return A Gremlin Query Part equivalent to Contraint.
 */
public GraphTraversal getQueryPartForContraintsMap(GraphTraversal traversal, final ReadRecordsRequest recordsRequest) {
    final Map<String, ValueSet> hashMap = recordsRequest.getConstraints().getSummary();
    if (hashMap.size() == 0) {
        return traversal;
    }
    logger.info("readWithContraint: Constaints Map " + hashMap.toString());
    final Set<String> setOfkeys = (Set<String>) (hashMap.keySet());
    for (final String key : setOfkeys) {
        if (hashMap.get(key) instanceof SortedRangeSet) {
            final List<Range> ranges = ((SortedRangeSet) hashMap.get(key)).getOrderedRanges();
            for (final Range range : ranges) {
                if (!range.getLow().isNullValue() && !range.getHigh().isNullValue()) {
                    if (range.getLow().getValue().toString().equals(range.getHigh().getValue().toString())) {
                        traversal = GremlinQueryPreProcessor.generateGremlinQueryPart(traversal, key, range.getLow().getValue().toString(), range.getType(), range.getLow().getBound(), GremlinQueryPreProcessor.Operator.EQUALTO);
                        break;
                    }
                }
                if (!range.getLow().isNullValue()) {
                    traversal = GremlinQueryPreProcessor.generateGremlinQueryPart(traversal, key, range.getLow().getValue().toString(), range.getType(), range.getLow().getBound(), GremlinQueryPreProcessor.Operator.GREATERTHAN);
                }
                if (!range.getHigh().isNullValue()) {
                    traversal = GremlinQueryPreProcessor.generateGremlinQueryPart(traversal, key, range.getHigh().getValue().toString(), range.getType(), range.getHigh().getBound(), GremlinQueryPreProcessor.Operator.LESSTHAN);
                }
            }
        }
        if (hashMap.get(key) instanceof EquatableValueSet) {
            final EquatableValueSet valueSet = ((EquatableValueSet) hashMap.get(key));
            if (valueSet.isWhiteList()) {
                traversal = GremlinQueryPreProcessor.generateGremlinQueryPart(traversal, key, valueSet.getValue(0).toString(), valueSet.getType(), null, GremlinQueryPreProcessor.Operator.EQUALTO);
            } else {
                traversal = GremlinQueryPreProcessor.generateGremlinQueryPart(traversal, key, valueSet.getValue(0).toString(), valueSet.getType(), null, GremlinQueryPreProcessor.Operator.NOTEQUALTO);
            }
        }
    }
    return traversal;
}
Also used : EquatableValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.EquatableValueSet) ValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet) SortedRangeSet(com.amazonaws.athena.connector.lambda.domain.predicate.SortedRangeSet) Set(java.util.Set) SortedRangeSet(com.amazonaws.athena.connector.lambda.domain.predicate.SortedRangeSet) EquatableValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.EquatableValueSet) Range(com.amazonaws.athena.connector.lambda.domain.predicate.Range) EquatableValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.EquatableValueSet) ValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet)

Example 10 with ValueSet

use of com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet in project aws-athena-query-federation by awslabs.

the class MySqlRecordHandlerTest method getRangeSet.

private ValueSet getRangeSet(Marker.Bound lowerBound, Object lowerValue, Marker.Bound upperBound, Object upperValue) {
    Range range = Mockito.mock(Range.class, Mockito.RETURNS_DEEP_STUBS);
    Mockito.when(range.isSingleValue()).thenReturn(false);
    Mockito.when(range.getLow().getBound()).thenReturn(lowerBound);
    Mockito.when(range.getLow().getValue()).thenReturn(lowerValue);
    Mockito.when(range.getHigh().getBound()).thenReturn(upperBound);
    Mockito.when(range.getHigh().getValue()).thenReturn(upperValue);
    ValueSet valueSet = Mockito.mock(SortedRangeSet.class, Mockito.RETURNS_DEEP_STUBS);
    Mockito.when(valueSet.getRanges().getOrderedRanges()).thenReturn(Collections.singletonList(range));
    return valueSet;
}
Also used : Range(com.amazonaws.athena.connector.lambda.domain.predicate.Range) ValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet)

Aggregations

ValueSet (com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet)104 Test (org.junit.Test)66 Constraints (com.amazonaws.athena.connector.lambda.domain.predicate.Constraints)63 HashMap (java.util.HashMap)48 TableName (com.amazonaws.athena.connector.lambda.domain.TableName)47 Schema (org.apache.arrow.vector.types.pojo.Schema)37 Split (com.amazonaws.athena.connector.lambda.domain.Split)31 Range (com.amazonaws.athena.connector.lambda.domain.predicate.Range)27 ReadRecordsRequest (com.amazonaws.athena.connector.lambda.records.ReadRecordsRequest)27 EquatableValueSet (com.amazonaws.athena.connector.lambda.domain.predicate.EquatableValueSet)26 ArrayList (java.util.ArrayList)25 Matchers.anyString (org.mockito.Matchers.anyString)25 RecordResponse (com.amazonaws.athena.connector.lambda.records.RecordResponse)24 Block (com.amazonaws.athena.connector.lambda.data.Block)23 S3SpillLocation (com.amazonaws.athena.connector.lambda.domain.spill.S3SpillLocation)21 RemoteReadRecordsResponse (com.amazonaws.athena.connector.lambda.records.RemoteReadRecordsResponse)18 SchemaBuilder (com.amazonaws.athena.connector.lambda.data.SchemaBuilder)17 ReadRecordsResponse (com.amazonaws.athena.connector.lambda.records.ReadRecordsResponse)17 InvocationOnMock (org.mockito.invocation.InvocationOnMock)17 BlockAllocatorImpl (com.amazonaws.athena.connector.lambda.data.BlockAllocatorImpl)13