Search in sources :

Example 16 with Constraints

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

the class HiveMetadataHandlerTest method decodeContinuationToken.

@Test
public void decodeContinuationToken() throws Exception {
    TableName tableName = new TableName("testSchema", "testTable");
    Constraints constraints = Mockito.mock(Constraints.class);
    Schema partitionSchema = this.hiveMetadataHandler.getPartitionSchema("testCatalogName");
    Set<String> partitionCols = partitionSchema.getFields().stream().map(Field::getName).collect(Collectors.toSet());
    BlockAllocator blockAllocator = new BlockAllocatorImpl();
    GetTableLayoutRequest getTableLayoutRequest = new GetTableLayoutRequest(this.federatedIdentity, "testQueryId", "testCatalogName", tableName, constraints, partitionSchema, partitionCols);
    GetTableLayoutResponse getTableLayoutResponse = this.hiveMetadataHandler.doGetTableLayout(blockAllocator, getTableLayoutRequest);
    GetSplitsRequest getSplitsRequest = new GetSplitsRequest(this.federatedIdentity, "testQueryId", "testCatalogName", tableName, getTableLayoutResponse.getPartitions(), new ArrayList<>(partitionCols), constraints, "1");
    Integer splitRequestToken = 0;
    if (getSplitsRequest.hasContinuationToken()) {
        splitRequestToken = Integer.valueOf(getSplitsRequest.getContinuationToken());
    }
    Assert.assertNotNull(splitRequestToken.toString());
}
Also used : AtomicInteger(java.util.concurrent.atomic.AtomicInteger) 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) Test(org.junit.Test)

Example 17 with Constraints

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

the class HiveMuxRecordHandlerTest method buildSplitSql.

@Test
public void buildSplitSql() throws SQLException {
    ReadRecordsRequest readRecordsRequest = Mockito.mock(ReadRecordsRequest.class);
    Mockito.when(readRecordsRequest.getCatalogName()).thenReturn("recordHive");
    Connection jdbcConnection = Mockito.mock(Connection.class);
    TableName tableName = new TableName("testSchema", "tableName");
    Schema schema = Mockito.mock(Schema.class);
    Constraints constraints = Mockito.mock(Constraints.class);
    Split split = Mockito.mock(Split.class);
    this.jdbcRecordHandler.buildSplitSql(jdbcConnection, "recordHive", tableName, schema, constraints, split);
    Mockito.verify(this.hiveRecordHandler, Mockito.times(1)).buildSplitSql(Mockito.eq(jdbcConnection), Mockito.eq("recordHive"), Mockito.eq(tableName), Mockito.eq(schema), Mockito.eq(constraints), Mockito.eq(split));
}
Also used : TableName(com.amazonaws.athena.connector.lambda.domain.TableName) ReadRecordsRequest(com.amazonaws.athena.connector.lambda.records.ReadRecordsRequest) Constraints(com.amazonaws.athena.connector.lambda.domain.predicate.Constraints) Schema(org.apache.arrow.vector.types.pojo.Schema) Connection(java.sql.Connection) Split(com.amazonaws.athena.connector.lambda.domain.Split) Test(org.junit.Test)

Example 18 with Constraints

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

the class JdbcRecordHandlerTest method readWithConstraint.

@Test
public void readWithConstraint() throws SQLException {
    ConstraintEvaluator constraintEvaluator = Mockito.mock(ConstraintEvaluator.class);
    Mockito.when(constraintEvaluator.apply(Mockito.anyString(), Mockito.any())).thenReturn(true);
    TableName inputTableName = new TableName("testSchema", "testTable");
    SchemaBuilder expectedSchemaBuilder = SchemaBuilder.newBuilder();
    expectedSchemaBuilder.addField(FieldBuilder.newBuilder("testCol1", org.apache.arrow.vector.types.Types.MinorType.INT.getType()).build());
    expectedSchemaBuilder.addField(FieldBuilder.newBuilder("testCol2", org.apache.arrow.vector.types.Types.MinorType.VARCHAR.getType()).build());
    expectedSchemaBuilder.addField(FieldBuilder.newBuilder("testPartitionCol", org.apache.arrow.vector.types.Types.MinorType.VARCHAR.getType()).build());
    Schema fieldSchema = expectedSchemaBuilder.build();
    BlockAllocator allocator = new BlockAllocatorImpl();
    S3SpillLocation s3SpillLocation = S3SpillLocation.newBuilder().withIsDirectory(true).build();
    Split.Builder splitBuilder = Split.newBuilder(s3SpillLocation, null).add("testPartitionCol", String.valueOf("testPartitionValue"));
    Constraints constraints = Mockito.mock(Constraints.class, Mockito.RETURNS_DEEP_STUBS);
    String[] schema = { "testCol1", "testCol2" };
    int[] columnTypes = { Types.INTEGER, Types.VARCHAR };
    Object[][] values = { { 1, "testVal1" }, { 2, "testVal2" } };
    AtomicInteger rowNumber = new AtomicInteger(-1);
    ResultSet resultSet = mockResultSet(schema, columnTypes, values, rowNumber);
    Mockito.when(this.preparedStatement.executeQuery()).thenReturn(resultSet);
    SpillConfig spillConfig = Mockito.mock(SpillConfig.class);
    Mockito.when(spillConfig.getSpillLocation()).thenReturn(s3SpillLocation);
    BlockSpiller s3Spiller = new S3BlockSpiller(this.amazonS3, spillConfig, allocator, fieldSchema, constraintEvaluator);
    ReadRecordsRequest readRecordsRequest = new ReadRecordsRequest(this.federatedIdentity, "testCatalog", "testQueryId", inputTableName, fieldSchema, splitBuilder.build(), constraints, 1024, 1024);
    Mockito.when(amazonS3.putObject(Mockito.anyString(), Mockito.anyString(), Mockito.any(), Mockito.any())).thenAnswer((Answer<PutObjectResult>) invocation -> {
        ByteArrayInputStream byteArrayInputStream = (ByteArrayInputStream) invocation.getArguments()[2];
        int n = byteArrayInputStream.available();
        byte[] bytes = new byte[n];
        byteArrayInputStream.read(bytes, 0, n);
        String data = new String(bytes, StandardCharsets.UTF_8);
        Assert.assertTrue(data.contains("testVal1") || data.contains("testVal2") || data.contains("testPartitionValue"));
        return new PutObjectResult();
    });
    this.jdbcRecordHandler.readWithConstraint(s3Spiller, readRecordsRequest, queryStatusChecker);
}
Also used : Schema(org.apache.arrow.vector.types.pojo.Schema) Connection(java.sql.Connection) S3BlockSpiller(com.amazonaws.athena.connector.lambda.data.S3BlockSpiller) com.amazonaws.athena.connector.lambda.data.writers.extractors(com.amazonaws.athena.connector.lambda.data.writers.extractors) BlockAllocator(com.amazonaws.athena.connector.lambda.data.BlockAllocator) ConstraintEvaluator(com.amazonaws.athena.connector.lambda.domain.predicate.ConstraintEvaluator) Answer(org.mockito.stubbing.Answer) BlockSpiller(com.amazonaws.athena.connector.lambda.data.BlockSpiller) SQLException(java.sql.SQLException) JdbcCredentialProvider(com.amazonaws.athena.connectors.jdbc.connection.JdbcCredentialProvider) ByteArrayInputStream(java.io.ByteArrayInputStream) AtomicInteger(java.util.concurrent.atomic.AtomicInteger) SchemaBuilder(com.amazonaws.athena.connector.lambda.data.SchemaBuilder) ResultSet(java.sql.ResultSet) Map(java.util.Map) TestBase(com.amazonaws.athena.connectors.jdbc.TestBase) AmazonS3(com.amazonaws.services.s3.AmazonS3) QueryStatusChecker(com.amazonaws.athena.connector.lambda.QueryStatusChecker) BlockAllocatorImpl(com.amazonaws.athena.connector.lambda.data.BlockAllocatorImpl) FederatedIdentity(com.amazonaws.athena.connector.lambda.security.FederatedIdentity) PutObjectResult(com.amazonaws.services.s3.model.PutObjectResult) GetSecretValueResult(com.amazonaws.services.secretsmanager.model.GetSecretValueResult) Before(org.junit.Before) AmazonAthena(com.amazonaws.services.athena.AmazonAthena) Split(com.amazonaws.athena.connector.lambda.domain.Split) ReadRecordsRequest(com.amazonaws.athena.connector.lambda.records.ReadRecordsRequest) AWSSecretsManager(com.amazonaws.services.secretsmanager.AWSSecretsManager) Test(org.junit.Test) DatabaseConnectionConfig(com.amazonaws.athena.connectors.jdbc.connection.DatabaseConnectionConfig) PreparedStatement(java.sql.PreparedStatement) TableName(com.amazonaws.athena.connector.lambda.domain.TableName) StandardCharsets(java.nio.charset.StandardCharsets) FieldBuilder(com.amazonaws.athena.connector.lambda.data.FieldBuilder) Constraints(com.amazonaws.athena.connector.lambda.domain.predicate.Constraints) Mockito(org.mockito.Mockito) S3SpillLocation(com.amazonaws.athena.connector.lambda.domain.spill.S3SpillLocation) SpillConfig(com.amazonaws.athena.connector.lambda.data.SpillConfig) Assert(org.junit.Assert) JdbcConnectionFactory(com.amazonaws.athena.connectors.jdbc.connection.JdbcConnectionFactory) Collections(java.util.Collections) GetSecretValueRequest(com.amazonaws.services.secretsmanager.model.GetSecretValueRequest) Types(java.sql.Types) PutObjectResult(com.amazonaws.services.s3.model.PutObjectResult) Schema(org.apache.arrow.vector.types.pojo.Schema) ConstraintEvaluator(com.amazonaws.athena.connector.lambda.domain.predicate.ConstraintEvaluator) TableName(com.amazonaws.athena.connector.lambda.domain.TableName) Constraints(com.amazonaws.athena.connector.lambda.domain.predicate.Constraints) SpillConfig(com.amazonaws.athena.connector.lambda.data.SpillConfig) ReadRecordsRequest(com.amazonaws.athena.connector.lambda.records.ReadRecordsRequest) BlockAllocatorImpl(com.amazonaws.athena.connector.lambda.data.BlockAllocatorImpl) AtomicInteger(java.util.concurrent.atomic.AtomicInteger) ByteArrayInputStream(java.io.ByteArrayInputStream) S3SpillLocation(com.amazonaws.athena.connector.lambda.domain.spill.S3SpillLocation) BlockAllocator(com.amazonaws.athena.connector.lambda.data.BlockAllocator) SchemaBuilder(com.amazonaws.athena.connector.lambda.data.SchemaBuilder) ResultSet(java.sql.ResultSet) S3BlockSpiller(com.amazonaws.athena.connector.lambda.data.S3BlockSpiller) Split(com.amazonaws.athena.connector.lambda.domain.Split) S3BlockSpiller(com.amazonaws.athena.connector.lambda.data.S3BlockSpiller) BlockSpiller(com.amazonaws.athena.connector.lambda.data.BlockSpiller) Test(org.junit.Test)

Example 19 with Constraints

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

the class PostGreSqlMetadataHandlerTest method doGetSplitsContinuation.

@Test
public void doGetSplitsContinuation() throws Exception {
    BlockAllocator blockAllocator = new BlockAllocatorImpl();
    Constraints constraints = Mockito.mock(Constraints.class);
    TableName tableName = new TableName("testSchema", "testTable");
    Schema partitionSchema = this.postGreSqlMetadataHandler.getPartitionSchema("testCatalogName");
    Set<String> partitionCols = partitionSchema.getFields().stream().map(Field::getName).collect(Collectors.toSet());
    GetTableLayoutRequest getTableLayoutRequest = new GetTableLayoutRequest(this.federatedIdentity, "testQueryId", "testCatalogName", tableName, constraints, partitionSchema, partitionCols);
    PreparedStatement preparedStatement = Mockito.mock(PreparedStatement.class);
    Mockito.when(this.connection.prepareStatement(PostGreSqlMetadataHandler.GET_PARTITIONS_QUERY)).thenReturn(preparedStatement);
    String[] columns = { "child_schema", "child" };
    int[] types = { Types.VARCHAR, Types.VARCHAR };
    Object[][] values = { { "s0", "p0" }, { "s1", "p1" } };
    ResultSet resultSet = mockResultSet(columns, types, values, new AtomicInteger(-1));
    final String expectedQuery = String.format(PostGreSqlMetadataHandler.GET_PARTITIONS_QUERY, tableName.getTableName(), tableName.getSchemaName());
    Mockito.when(preparedStatement.executeQuery()).thenReturn(resultSet);
    Mockito.when(this.connection.getMetaData().getSearchStringEscape()).thenReturn(null);
    GetTableLayoutResponse getTableLayoutResponse = this.postGreSqlMetadataHandler.doGetTableLayout(blockAllocator, getTableLayoutRequest);
    BlockAllocator splitBlockAllocator = new BlockAllocatorImpl();
    GetSplitsRequest getSplitsRequest = new GetSplitsRequest(this.federatedIdentity, "testQueryId", "testCatalogName", tableName, getTableLayoutResponse.getPartitions(), new ArrayList<>(partitionCols), constraints, "1");
    GetSplitsResponse getSplitsResponse = this.postGreSqlMetadataHandler.doGetSplits(splitBlockAllocator, getSplitsRequest);
    Set<Map<String, String>> expectedSplits = new HashSet<>();
    expectedSplits.add(ImmutableMap.of("partition_schema_name", "s1", "partition_name", "p1"));
    Assert.assertEquals(expectedSplits.size(), getSplitsResponse.getSplits().size());
    Set<Map<String, String>> actualSplits = getSplitsResponse.getSplits().stream().map(Split::getProperties).collect(Collectors.toSet());
    Assert.assertEquals(expectedSplits, actualSplits);
}
Also used : GetSplitsRequest(com.amazonaws.athena.connector.lambda.metadata.GetSplitsRequest) Schema(org.apache.arrow.vector.types.pojo.Schema) PreparedStatement(java.sql.PreparedStatement) TableName(com.amazonaws.athena.connector.lambda.domain.TableName) Constraints(com.amazonaws.athena.connector.lambda.domain.predicate.Constraints) GetTableLayoutResponse(com.amazonaws.athena.connector.lambda.metadata.GetTableLayoutResponse) BlockAllocatorImpl(com.amazonaws.athena.connector.lambda.data.BlockAllocatorImpl) AtomicInteger(java.util.concurrent.atomic.AtomicInteger) GetSplitsResponse(com.amazonaws.athena.connector.lambda.metadata.GetSplitsResponse) BlockAllocator(com.amazonaws.athena.connector.lambda.data.BlockAllocator) GetTableLayoutRequest(com.amazonaws.athena.connector.lambda.metadata.GetTableLayoutRequest) ResultSet(java.sql.ResultSet) Map(java.util.Map) ImmutableMap(com.google.common.collect.ImmutableMap) HashSet(java.util.HashSet) Test(org.junit.Test)

Example 20 with Constraints

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

the class PostGreSqlMetadataHandlerTest method doGetTableLayoutWithSQLException.

@Test(expected = RuntimeException.class)
public void doGetTableLayoutWithSQLException() throws Exception {
    Constraints constraints = Mockito.mock(Constraints.class);
    TableName tableName = new TableName("testSchema", "testTable");
    Schema partitionSchema = this.postGreSqlMetadataHandler.getPartitionSchema("testCatalogName");
    Set<String> partitionCols = partitionSchema.getFields().stream().map(Field::getName).collect(Collectors.toSet());
    GetTableLayoutRequest getTableLayoutRequest = new GetTableLayoutRequest(this.federatedIdentity, "testQueryId", "testCatalogName", tableName, constraints, partitionSchema, partitionCols);
    Connection connection = Mockito.mock(Connection.class, Mockito.RETURNS_DEEP_STUBS);
    JdbcConnectionFactory jdbcConnectionFactory = Mockito.mock(JdbcConnectionFactory.class);
    Mockito.when(jdbcConnectionFactory.getConnection(Mockito.any(JdbcCredentialProvider.class))).thenReturn(connection);
    Mockito.when(connection.getMetaData().getSearchStringEscape()).thenThrow(new SQLException());
    PostGreSqlMetadataHandler postGreSqlMetadataHandler = new PostGreSqlMetadataHandler(databaseConnectionConfig, this.secretsManager, this.athena, jdbcConnectionFactory);
    postGreSqlMetadataHandler.doGetTableLayout(Mockito.mock(BlockAllocator.class), getTableLayoutRequest);
}
Also used : JdbcConnectionFactory(com.amazonaws.athena.connectors.jdbc.connection.JdbcConnectionFactory) TableName(com.amazonaws.athena.connector.lambda.domain.TableName) Constraints(com.amazonaws.athena.connector.lambda.domain.predicate.Constraints) SQLException(java.sql.SQLException) Schema(org.apache.arrow.vector.types.pojo.Schema) GetTableLayoutRequest(com.amazonaws.athena.connector.lambda.metadata.GetTableLayoutRequest) BlockAllocator(com.amazonaws.athena.connector.lambda.data.BlockAllocator) Connection(java.sql.Connection) JdbcCredentialProvider(com.amazonaws.athena.connectors.jdbc.connection.JdbcCredentialProvider) Test(org.junit.Test)

Aggregations

Constraints (com.amazonaws.athena.connector.lambda.domain.predicate.Constraints)182 Test (org.junit.Test)172 TableName (com.amazonaws.athena.connector.lambda.domain.TableName)148 Schema (org.apache.arrow.vector.types.pojo.Schema)136 ValueSet (com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet)64 HashMap (java.util.HashMap)63 Split (com.amazonaws.athena.connector.lambda.domain.Split)59 ReadRecordsRequest (com.amazonaws.athena.connector.lambda.records.ReadRecordsRequest)55 AtomicInteger (java.util.concurrent.atomic.AtomicInteger)54 BlockAllocator (com.amazonaws.athena.connector.lambda.data.BlockAllocator)50 GetTableLayoutRequest (com.amazonaws.athena.connector.lambda.metadata.GetTableLayoutRequest)47 BlockAllocatorImpl (com.amazonaws.athena.connector.lambda.data.BlockAllocatorImpl)44 ArrayList (java.util.ArrayList)44 PreparedStatement (java.sql.PreparedStatement)42 GetTableLayoutResponse (com.amazonaws.athena.connector.lambda.metadata.GetTableLayoutResponse)41 Block (com.amazonaws.athena.connector.lambda.data.Block)36 SchemaBuilder (com.amazonaws.athena.connector.lambda.data.SchemaBuilder)35 RecordResponse (com.amazonaws.athena.connector.lambda.records.RecordResponse)33 ResultSet (java.sql.ResultSet)32 GetSplitsResponse (com.amazonaws.athena.connector.lambda.metadata.GetSplitsResponse)30