@Test public void pivot() { Dataset<Row> df = spark.table("courseSales"); List<Row> actual = df.groupBy("year") .pivot("course", Arrays.asList("dotNET", "Java")) .agg(sum("earnings")).orderBy("year").collectAsList(); Assert.assertEquals(2012, actual.get(0).getInt(0)); Assert.assertEquals(15000.0, actual.get(0).getDouble(1), 0.01); Assert.assertEquals(20000.0, actual.get(0).getDouble(2), 0.01); Assert.assertEquals(2013, actual.get(1).getInt(0)); Assert.assertEquals(48000.0, actual.get(1).getDouble(1), 0.01); Assert.assertEquals(30000.0, actual.get(1).getDouble(2), 0.01); }
@Test public void pivot() { Dataset<Row> df = spark.table("courseSales"); List<Row> actual = df.groupBy("year") .pivot("course", Arrays.asList("dotNET", "Java")) .agg(sum("earnings")).orderBy("year").collectAsList(); Assert.assertEquals(2012, actual.get(0).getInt(0)); Assert.assertEquals(15000.0, actual.get(0).getDouble(1), 0.01); Assert.assertEquals(20000.0, actual.get(0).getDouble(2), 0.01); Assert.assertEquals(2013, actual.get(1).getInt(0)); Assert.assertEquals(48000.0, actual.get(1).getDouble(1), 0.01); Assert.assertEquals(30000.0, actual.get(1).getDouble(2), 0.01); }
@Test public void pivot() { Dataset<Row> df = spark.table("courseSales"); List<Row> actual = df.groupBy("year") .pivot("course", Arrays.asList("dotNET", "Java")) .agg(sum("earnings")).orderBy("year").collectAsList(); Assert.assertEquals(2012, actual.get(0).getInt(0)); Assert.assertEquals(15000.0, actual.get(0).getDouble(1), 0.01); Assert.assertEquals(20000.0, actual.get(0).getDouble(2), 0.01); Assert.assertEquals(2013, actual.get(1).getInt(0)); Assert.assertEquals(48000.0, actual.get(1).getDouble(1), 0.01); Assert.assertEquals(30000.0, actual.get(1).getDouble(2), 0.01); }
.agg(sum("value")) .head() .getLong(0);