Search in sources :

Example 6 with SchemaTracker

use of org.apache.drill.exec.physical.impl.protocol.SchemaTracker in project drill by apache.

the class TestScanOrchestratorEarlySchema method testModeSmoothing.

@Test
public void testModeSmoothing() {
    ScanOrchestratorBuilder builder = new MockScanBuilder();
    builder.enableSchemaSmoothing(true);
    builder.projection(RowSetTestUtils.projectList("a"));
    ScanSchemaOrchestrator scanner = new ScanSchemaOrchestrator(fixture.allocator(), builder);
    // Most general schema: nullable, with precision.
    TupleMetadata schema1 = new SchemaBuilder().addNullable("a", MinorType.VARCHAR, 10).buildSchema();
    SchemaTracker tracker = new SchemaTracker();
    int schemaVersion;
    {
        // Table 1: most permissive type
        ReaderSchemaOrchestrator reader = scanner.startReader();
        ResultSetLoader loader = reader.makeTableLoader(schema1);
        // Create a batch
        reader.startBatch();
        loader.writer().addRow("fred").addRow("wilma");
        reader.endBatch();
        tracker.trackSchema(scanner.output());
        schemaVersion = tracker.schemaVersion();
        // Verify
        SingleRowSet expected = fixture.rowSetBuilder(schema1).addRow("fred").addRow("wilma").build();
        RowSetUtilities.verify(expected, fixture.wrap(scanner.output()));
        scanner.closeReader();
    }
    {
        // Table 2: required, use nullable
        // Required version.
        TupleMetadata schema2 = new SchemaBuilder().add("a", MinorType.VARCHAR, 10).buildSchema();
        ReaderSchemaOrchestrator reader = scanner.startReader();
        ResultSetLoader loader = reader.makeTableLoader(schema2);
        // Create a batch
        reader.startBatch();
        loader.writer().addRow("barney").addRow("betty");
        reader.endBatch();
        // Verify, using persistent schema
        tracker.trackSchema(scanner.output());
        assertEquals(schemaVersion, tracker.schemaVersion());
        SingleRowSet expected = fixture.rowSetBuilder(schema1).addRow("barney").addRow("betty").build();
        RowSetUtilities.verify(expected, fixture.wrap(scanner.output()));
        scanner.closeReader();
    }
    {
        // Table 3: narrower precision, use wider
        // Required version with narrower precision.
        TupleMetadata schema3 = new SchemaBuilder().add("a", MinorType.VARCHAR, 5).buildSchema();
        ReaderSchemaOrchestrator reader = scanner.startReader();
        ResultSetLoader loader = reader.makeTableLoader(schema3);
        // Create a batch
        reader.startBatch();
        loader.writer().addRow("bam-bam").addRow("pebbles");
        reader.endBatch();
        // Verify, using persistent schema
        tracker.trackSchema(scanner.output());
        assertEquals(schemaVersion, tracker.schemaVersion());
        SingleRowSet expected = fixture.rowSetBuilder(schema1).addRow("bam-bam").addRow("pebbles").build();
        RowSetUtilities.verify(expected, fixture.wrap(scanner.output()));
        scanner.closeReader();
    }
    scanner.close();
}
Also used : SingleRowSet(org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet) ResultSetLoader(org.apache.drill.exec.physical.resultSet.ResultSetLoader) ScanOrchestratorBuilder(org.apache.drill.exec.physical.impl.scan.project.ScanSchemaOrchestrator.ScanOrchestratorBuilder) TupleMetadata(org.apache.drill.exec.record.metadata.TupleMetadata) SchemaBuilder(org.apache.drill.exec.record.metadata.SchemaBuilder) BatchSchemaBuilder(org.apache.drill.exec.record.BatchSchemaBuilder) SchemaTracker(org.apache.drill.exec.physical.impl.protocol.SchemaTracker) MockScanBuilder(org.apache.drill.exec.physical.impl.scan.ScanTestUtils.MockScanBuilder) ScanSchemaOrchestrator(org.apache.drill.exec.physical.impl.scan.project.ScanSchemaOrchestrator) ReaderSchemaOrchestrator(org.apache.drill.exec.physical.impl.scan.project.ReaderSchemaOrchestrator) SubOperatorTest(org.apache.drill.test.SubOperatorTest) Test(org.junit.Test)

Aggregations

SchemaTracker (org.apache.drill.exec.physical.impl.protocol.SchemaTracker)6 MockScanBuilder (org.apache.drill.exec.physical.impl.scan.ScanTestUtils.MockScanBuilder)6 ScanOrchestratorBuilder (org.apache.drill.exec.physical.impl.scan.project.ScanSchemaOrchestrator.ScanOrchestratorBuilder)6 SchemaBuilder (org.apache.drill.exec.record.metadata.SchemaBuilder)6 TupleMetadata (org.apache.drill.exec.record.metadata.TupleMetadata)6 SubOperatorTest (org.apache.drill.test.SubOperatorTest)6 Test (org.junit.Test)6 ReaderSchemaOrchestrator (org.apache.drill.exec.physical.impl.scan.project.ReaderSchemaOrchestrator)5 ScanSchemaOrchestrator (org.apache.drill.exec.physical.impl.scan.project.ScanSchemaOrchestrator)5 ResultSetLoader (org.apache.drill.exec.physical.resultSet.ResultSetLoader)5 SingleRowSet (org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet)5 BatchSchemaBuilder (org.apache.drill.exec.record.BatchSchemaBuilder)4 File (java.io.File)1 SchemaPath (org.apache.drill.common.expression.SchemaPath)1 ImplicitColumnManager (org.apache.drill.exec.physical.impl.scan.file.ImplicitColumnManager)1 BatchSchema (org.apache.drill.exec.record.BatchSchema)1 VectorContainer (org.apache.drill.exec.record.VectorContainer)1 RowSetComparison (org.apache.drill.test.rowSet.RowSetComparison)1 Path (org.apache.hadoop.fs.Path)1