[SPARK-11625][SQL] add java test for typed aggregate

[cloud-fan]

No description provided.

[cloud-fan]

java doesn't support contravariant in TypedColumn, which makes this typed aggregate API hard to use at java side.

cc @marmbrus @rxin

[marmbrus]

Yeah, this is unfortunate. I think that the only thing we can do is use the DataFrame version of agg and then cast afterwards.

    Dataset<Tuple4<String, Integer, Long, Long>> agged2 = grouped.agg(
            new IntSumOf().toColumn(e.INT(), e.INT()),
            expr("sum(_2)"),
            count("*")).as(e.tuple(e.STRING(), e.INT(), e.LONG(), e.LONG()));

[SparkQA]

Test build #45511 has finished for PR 9591 at commit 479d8c5.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n * case class Record(data: ByteBuffer, time: Long, promise: Promise[WriteAheadLogRecordHandle])\n

[SparkQA]

Test build #45523 has finished for PR 9591 at commit 4e0c686.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[marmbrus]

We should do something about this. I think that if we rename the object to Encoders then Scala will build static forwarders and we can just call Encoders.STRING()

[SparkQA]

Test build #45595 has finished for PR 9591 at commit ae55976.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[cloud-fan]

retest this please.

[SparkQA]

Test build #45623 has finished for PR 9591 at commit ae55976.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[SparkQA]

Test build #45632 has finished for PR 9591 at commit 3806107.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[cloud-fan]

retest this please

[SparkQA]

Test build #45635 has finished for PR 9591 at commit 3806107.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[SparkQA]

Test build #2045 has finished for PR 9591 at commit 3806107.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[SparkQA]

Test build #2046 has finished for PR 9591 at commit 3806107.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[SparkQA]

Test build #45709 has finished for PR 9591 at commit b6de790.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[cloud-fan]

retest this please.

[SparkQA]

Test build #45723 has finished for PR 9591 at commit b6de790.

• This patch fails Spark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[cloud-fan]

retest this please.

[SparkQA]

Test build #45740 has finished for PR 9591 at commit b6de790.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[marmbrus]

Maybe we should just move this to Queryable with docs.

[marmbrus]

Oh sorry, maybe we need to make GroupQueryable?

[marmbrus]

@rxin I think we need to expose this for the java users to work, while still getting nice inference for scala

[cloud-fan]

We need to provide a tEncoder here, which is not the case of GroupedData.

[marmbrus]

Okay, we still need to add scaladoc then.

[SparkQA]

Test build #45821 has finished for PR 9591 at commit 7c9eb43.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * class TypedColumn[-T, U](\n * abstract class Aggregator[-A, B, C] extends Serializable\n * class JavaTrackStateDStream[KeyType, ValueType, StateType, EmittedType](\n

[SparkQA]

Test build #45907 has finished for PR 9591 at commit ae752fd.

• This patch fails PySpark unit tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[cloud-fan]

retest this please.

[SparkQA]

Test build #45925 has finished for PR 9591 at commit ae752fd.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds the following public classes (experimental):\n * abstract class Aggregator[-A, B, C] extends Serializable\n

[marmbrus]

Thanks! Merging to master and 1.6.