Search in sources :

Example 51 with SingleRowSet

use of org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet in project drill by apache.

the class TestSorter method testTwoRows.

// Paranoia: sort with two rows.
@Test
public void testTwoRows() throws Exception {
    TupleMetadata schema = SortTestUtilities.nonNullSchema();
    SingleRowSet rowSet = new RowSetBuilder(fixture.allocator(), schema).addRow(1, "1").addRow(0, "0").withSv2().build();
    SingleRowSet expected = new RowSetBuilder(fixture.allocator(), schema).addRow(0, "0").addRow(1, "1").build();
    runSorterTest(rowSet, expected);
}
Also used : RowSetBuilder(org.apache.drill.exec.physical.rowSet.RowSetBuilder) SingleRowSet(org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet) TupleMetadata(org.apache.drill.exec.record.metadata.TupleMetadata) DrillTest(org.apache.drill.test.DrillTest) OperatorTest(org.apache.drill.categories.OperatorTest) Test(org.junit.Test)

Example 52 with SingleRowSet

use of org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet in project drill by apache.

the class TestResultSetLoaderDictArray method testScalarValue.

@Test
public void testScalarValue() {
    TupleMetadata schema = new SchemaBuilder().add("a", MinorType.INT).addDictArray("d", MinorType.VARCHAR).value(MinorType.INT).resumeSchema().buildSchema();
    ResultSetLoaderImpl.ResultSetOptions options = new ResultSetOptionBuilder().readerSchema(schema).build();
    ResultSetLoader rsLoader = new ResultSetLoaderImpl(fixture.allocator(), options);
    RowSetLoader rootWriter = rsLoader.writer();
    // Write a couple of rows
    rsLoader.startBatch();
    rootWriter.addRow(10, objArray(map("a", 1, "b", 2, "d", 4), map("a", 2, "c", 3, "d", 1, "e", 4))).addRow(20, objArray()).addRow(30, objArray(map("a", 2, "c", 4, "d", 5, "e", 6, "f", 11), map("a", 1, "d", 6, "c", 3), map("b", 2, "a", 3)));
    // Verify the batch
    RowSet actual = fixture.wrap(rsLoader.harvest());
    SingleRowSet expected = fixture.rowSetBuilder(schema).addRow(10, objArray(map("a", 1, "b", 2, "d", 4), map("a", 2, "c", 3, "d", 1, "e", 4))).addRow(20, objArray()).addRow(30, objArray(map("a", 2, "c", 4, "d", 5, "e", 6, "f", 11), map("a", 1, "d", 6, "c", 3), map("b", 2, "a", 3))).build();
    RowSetUtilities.verify(expected, actual);
    rsLoader.close();
}
Also used : SingleRowSet(org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet) ResultSetLoader(org.apache.drill.exec.physical.resultSet.ResultSetLoader) TupleMetadata(org.apache.drill.exec.record.metadata.TupleMetadata) SchemaBuilder(org.apache.drill.exec.record.metadata.SchemaBuilder) SingleRowSet(org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet) RowSet(org.apache.drill.exec.physical.rowSet.RowSet) RowSetLoader(org.apache.drill.exec.physical.resultSet.RowSetLoader) SubOperatorTest(org.apache.drill.test.SubOperatorTest) Test(org.junit.Test)

Example 53 with SingleRowSet

use of org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet in project drill by apache.

the class TestResultSetLoaderDicts method testDictValue.

@Test
public void testDictValue() {
    final TupleMetadata schema = new SchemaBuilder().add("a", MinorType.INT).addDict("d", MinorType.INT).dictValue().key(MinorType.INT).nullableValue(MinorType.VARCHAR).resumeDict().resumeSchema().buildSchema();
    final ResultSetLoaderImpl.ResultSetOptions options = new ResultSetOptionBuilder().readerSchema(schema).build();
    final ResultSetLoader rsLoader = new ResultSetLoaderImpl(fixture.allocator(), options);
    final RowSetLoader rootWriter = rsLoader.writer();
    // Write some rows
    rsLoader.startBatch();
    rootWriter.addRow(10, map(1, map(1, "a", 2, "b", 4, "c"), 2, map(2, "a2", 1, "c2"))).addRow(20, map()).addRow(30, map(1, map(), 2, map(1, "a3"), 3, map(2, "b4", 4, "n4", 1, null), 4, map(3, "m5", 1, "a5", 2, "c5", 8, "m5", 21, "h5")));
    // Validate first batch
    RowSet actual = fixture.wrap(rsLoader.harvest());
    SingleRowSet expected = fixture.rowSetBuilder(schema).addRow(10, map(1, map(1, "a", 2, "b", 4, "c"), 2, map(2, "a2", 1, "c2"))).addRow(20, map()).addRow(30, map(1, map(), 2, map(1, "a3"), 3, map(2, "b4", 4, "n4", 1, null), 4, map(3, "m5", 1, "a5", 2, "c5", 8, "m5", 21, "h5"))).build();
    RowSetUtilities.verify(expected, actual);
    // Add another rows in the second batch.
    rsLoader.startBatch();
    rootWriter.addRow(40, map(1, map(1, "j6", 0, "k6"))).addRow(50, map(1, map(2, "l7"), 2, map(1, "o8", 5, "p8", 7, "u8")));
    // Validate first batch. The new dict should have been back-filled with
    // empty offsets for the missing rows.
    actual = fixture.wrap(rsLoader.harvest());
    expected = fixture.rowSetBuilder(actual.schema()).addRow(40, map(1, map(1, "j6", 0, "k6"))).addRow(50, map(1, map(2, "l7"), 2, map(1, "o8", 5, "p8", 7, "u8"))).build();
    RowSetUtilities.verify(expected, actual);
    rsLoader.close();
}
Also used : SingleRowSet(org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet) ResultSetLoader(org.apache.drill.exec.physical.resultSet.ResultSetLoader) TupleMetadata(org.apache.drill.exec.record.metadata.TupleMetadata) SchemaBuilder(org.apache.drill.exec.record.metadata.SchemaBuilder) SingleRowSet(org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet) RowSet(org.apache.drill.exec.physical.rowSet.RowSet) RowSetLoader(org.apache.drill.exec.physical.resultSet.RowSetLoader) SubOperatorTest(org.apache.drill.test.SubOperatorTest) Test(org.junit.Test)

Example 54 with SingleRowSet

use of org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet in project drill by apache.

the class TestResultSetLoaderMaps method testNestedMapsRequired.

/**
 * Create nested maps. Then, add columns to each map
 * on the fly. Use required, variable-width columns since
 * those require the most processing and are most likely to
 * fail if anything is out of place.
 */
@Test
public void testNestedMapsRequired() {
    final TupleMetadata schema = new SchemaBuilder().add("a", MinorType.INT).addMap("m1").add("b", MinorType.VARCHAR).addMap("m2").add("c", MinorType.VARCHAR).resumeMap().resumeSchema().buildSchema();
    final ResultSetLoaderImpl.ResultSetOptions options = new ResultSetOptionBuilder().readerSchema(schema).build();
    final ResultSetLoader rsLoader = new ResultSetLoaderImpl(fixture.allocator(), options);
    assertEquals(5, rsLoader.schemaVersion());
    final RowSetLoader rootWriter = rsLoader.writer();
    rsLoader.startBatch();
    rootWriter.addRow(10, mapValue("b1", mapValue("c1")));
    // Validate first batch
    RowSet actual = fixture.wrap(rsLoader.harvest());
    assertEquals(5, rsLoader.schemaVersion());
    SingleRowSet expected = fixture.rowSetBuilder(schema).addRow(10, mapValue("b1", mapValue("c1"))).build();
    RowSetUtilities.verify(expected, actual);
    // Now add columns in the second batch.
    rsLoader.startBatch();
    rootWriter.addRow(20, mapValue("b2", mapValue("c2")));
    final TupleWriter m1Writer = rootWriter.tuple("m1");
    m1Writer.addColumn(SchemaBuilder.columnSchema("d", MinorType.VARCHAR, DataMode.REQUIRED));
    final TupleWriter m2Writer = m1Writer.tuple("m2");
    m2Writer.addColumn(SchemaBuilder.columnSchema("e", MinorType.VARCHAR, DataMode.REQUIRED));
    rootWriter.addRow(30, mapValue("b3", mapValue("c3", "e3"), "d3"));
    // And another set while the write proceeds.
    m1Writer.addColumn(SchemaBuilder.columnSchema("f", MinorType.VARCHAR, DataMode.REQUIRED));
    m2Writer.addColumn(SchemaBuilder.columnSchema("g", MinorType.VARCHAR, DataMode.REQUIRED));
    rootWriter.addRow(40, mapValue("b4", mapValue("c4", "e4", "g4"), "d4", "e4"));
    // Validate second batch
    actual = fixture.wrap(rsLoader.harvest());
    assertEquals(9, rsLoader.schemaVersion());
    final TupleMetadata expectedSchema = new SchemaBuilder().add("a", MinorType.INT).addMap("m1").add("b", MinorType.VARCHAR).addMap("m2").add("c", MinorType.VARCHAR).add("e", MinorType.VARCHAR).add("g", MinorType.VARCHAR).resumeMap().add("d", MinorType.VARCHAR).add("f", MinorType.VARCHAR).resumeSchema().buildSchema();
    expected = fixture.rowSetBuilder(expectedSchema).addRow(20, mapValue("b2", mapValue("c2", "", ""), "", "")).addRow(30, mapValue("b3", mapValue("c3", "e3", ""), "d3", "")).addRow(40, mapValue("b4", mapValue("c4", "e4", "g4"), "d4", "e4")).build();
    RowSetUtilities.verify(expected, actual);
    rsLoader.close();
}
Also used : SingleRowSet(org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet) ResultSetLoader(org.apache.drill.exec.physical.resultSet.ResultSetLoader) TupleWriter(org.apache.drill.exec.vector.accessor.TupleWriter) TupleMetadata(org.apache.drill.exec.record.metadata.TupleMetadata) SchemaBuilder(org.apache.drill.exec.record.metadata.SchemaBuilder) SingleRowSet(org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet) RowSet(org.apache.drill.exec.physical.rowSet.RowSet) RowSetLoader(org.apache.drill.exec.physical.resultSet.RowSetLoader) SubOperatorTest(org.apache.drill.test.SubOperatorTest) Test(org.junit.Test)

Example 55 with SingleRowSet

use of org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet in project drill by apache.

the class TestResultSetLoaderMaps method testBasics.

@Test
public void testBasics() {
    final TupleMetadata schema = new SchemaBuilder().add("a", MinorType.INT).addMap("m").add("c", MinorType.INT).add("d", MinorType.VARCHAR).resumeSchema().add("e", MinorType.VARCHAR).buildSchema();
    final ResultSetLoaderImpl.ResultSetOptions options = new ResultSetOptionBuilder().readerSchema(schema).build();
    final ResultSetLoader rsLoader = new ResultSetLoaderImpl(fixture.allocator(), options);
    assertFalse(rsLoader.isProjectionEmpty());
    final RowSetLoader rootWriter = rsLoader.writer();
    // Verify structure and schema
    assertEquals(5, rsLoader.schemaVersion());
    final TupleMetadata actualSchema = rootWriter.tupleSchema();
    assertEquals(3, actualSchema.size());
    assertTrue(actualSchema.metadata(1).isMap());
    assertEquals(2, actualSchema.metadata("m").tupleSchema().size());
    assertEquals(2, actualSchema.column("m").getChildren().size());
    rsLoader.startBatch();
    // Write a row the way that clients will do.
    final ScalarWriter aWriter = rootWriter.scalar("a");
    final TupleWriter mWriter = rootWriter.tuple("m");
    final ScalarWriter cWriter = mWriter.scalar("c");
    final ScalarWriter dWriter = mWriter.scalar("d");
    final ScalarWriter eWriter = rootWriter.scalar("e");
    rootWriter.start();
    aWriter.setInt(10);
    cWriter.setInt(110);
    dWriter.setString("fred");
    eWriter.setString("pebbles");
    rootWriter.save();
    // Try adding a duplicate column.
    try {
        mWriter.addColumn(SchemaBuilder.columnSchema("c", MinorType.INT, DataMode.OPTIONAL));
        fail();
    } catch (final UserException e) {
    // Expected
    }
    // Write another using the test-time conveniences
    rootWriter.addRow(20, mapValue(210, "barney"), "bam-bam");
    // Harvest the batch
    final RowSet actual = fixture.wrap(rsLoader.harvest());
    assertEquals(5, rsLoader.schemaVersion());
    assertEquals(2, actual.rowCount());
    final MapVector mapVector = (MapVector) actual.container().getValueVector(1).getValueVector();
    assertEquals(2, mapVector.getAccessor().getValueCount());
    // Validate data
    final SingleRowSet expected = fixture.rowSetBuilder(schema).addRow(10, mapValue(110, "fred"), "pebbles").addRow(20, mapValue(210, "barney"), "bam-bam").build();
    RowSetUtilities.verify(expected, actual);
    rsLoader.close();
}
Also used : SingleRowSet(org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet) SingleRowSet(org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet) RowSet(org.apache.drill.exec.physical.rowSet.RowSet) ResultSetLoader(org.apache.drill.exec.physical.resultSet.ResultSetLoader) TupleWriter(org.apache.drill.exec.vector.accessor.TupleWriter) TupleMetadata(org.apache.drill.exec.record.metadata.TupleMetadata) SchemaBuilder(org.apache.drill.exec.record.metadata.SchemaBuilder) UserException(org.apache.drill.common.exceptions.UserException) RowSetLoader(org.apache.drill.exec.physical.resultSet.RowSetLoader) ScalarWriter(org.apache.drill.exec.vector.accessor.ScalarWriter) MapVector(org.apache.drill.exec.vector.complex.MapVector) SubOperatorTest(org.apache.drill.test.SubOperatorTest) Test(org.junit.Test)

Aggregations

SingleRowSet (org.apache.drill.exec.physical.rowSet.RowSet.SingleRowSet)257 Test (org.junit.Test)241 SchemaBuilder (org.apache.drill.exec.record.metadata.SchemaBuilder)237 TupleMetadata (org.apache.drill.exec.record.metadata.TupleMetadata)234 SubOperatorTest (org.apache.drill.test.SubOperatorTest)207 RowSet (org.apache.drill.exec.physical.rowSet.RowSet)69 ScalarReader (org.apache.drill.exec.vector.accessor.ScalarReader)62 ResultSetLoader (org.apache.drill.exec.physical.resultSet.ResultSetLoader)61 RowSetLoader (org.apache.drill.exec.physical.resultSet.RowSetLoader)54 ValueVector (org.apache.drill.exec.vector.ValueVector)32 EvfTest (org.apache.drill.categories.EvfTest)30 ScalarWriter (org.apache.drill.exec.vector.accessor.ScalarWriter)29 RowSetBuilder (org.apache.drill.exec.physical.rowSet.RowSetBuilder)27 TupleWriter (org.apache.drill.exec.vector.accessor.TupleWriter)27 RowSetComparison (org.apache.drill.test.rowSet.RowSetComparison)27 ExtendableRowSet (org.apache.drill.exec.physical.rowSet.RowSet.ExtendableRowSet)25 RepeatedValueVector (org.apache.drill.exec.vector.complex.RepeatedValueVector)24 BatchSchemaBuilder (org.apache.drill.exec.record.BatchSchemaBuilder)23 VectorContainer (org.apache.drill.exec.record.VectorContainer)22 ArrayReader (org.apache.drill.exec.vector.accessor.ArrayReader)22