void validateDataFrameWithBeans(Bean bean, DataFrame df) {
StructType schema = df.schema();
Assert.assertEquals(
new StructField("a", DoubleType$.MODULE$, false, Metadata.empty()), schema.apply("a"));
Assert.assertEquals(
new StructField("b", new ArrayType(IntegerType$.MODULE$, true), true, Metadata.empty()),
schema.apply("b"));
ArrayType valueType = new ArrayType(DataTypes.IntegerType, false);
MapType mapType = new MapType(DataTypes.StringType, valueType, true);
Assert.assertEquals(new StructField("c", mapType, true, Metadata.empty()), schema.apply("c"));
Assert.assertEquals(
new StructField("d", new ArrayType(DataTypes.StringType, true), true, Metadata.empty()),
schema.apply("d"));
Row first = df.select("a", "b", "c", "d").first();
Assert.assertEquals(bean.getA(), first.getDouble(0), 0.0);
// Now Java lists and maps are converted to Scala Seq's and Map's. Once we get a Seq below,
// verify that it has the expected length, and contains expected elements.
Seq<Integer> result = first.getAs(1);
Assert.assertEquals(bean.getB().length, result.length());
for (int i = 0; i < result.length(); i++) {
Assert.assertEquals(bean.getB()[i], result.apply(i));
}
@SuppressWarnings("unchecked")
Seq<Integer> outputBuffer = (Seq<Integer>) first.getJavaMap(2).get("hello");
Assert.assertArrayEquals(
bean.getC().get("hello"),
Ints.toArray(JavaConverters.seqAsJavaListConverter(outputBuffer).asJava()));
Seq<String> d = first.getAs(3);
Assert.assertEquals(bean.getD().size(), d.length());
for (int i = 0; i < d.length(); i++) {
Assert.assertEquals(bean.getD().get(i), d.apply(i));
}
}
@Test
public void testCrosstab() {
DataFrame df = context.table("testData2");
DataFrame crosstab = df.stat().crosstab("a", "b");
String[] columnNames = crosstab.schema().fieldNames();
Assert.assertEquals("a_b", columnNames[0]);
Assert.assertEquals("2", columnNames[1]);
Assert.assertEquals("1", columnNames[2]);
Row[] rows = crosstab.collect();
Arrays.sort(rows, crosstabRowComparator);
Integer count = 1;
for (Row row : rows) {
Assert.assertEquals(row.get(0).toString(), count.toString());
Assert.assertEquals(1L, row.getLong(1));
Assert.assertEquals(1L, row.getLong(2));
count++;
}
}