1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
|
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.spark.sql
import scala.beans.{BeanInfo, BeanProperty}
import org.apache.spark.rdd.RDD
import org.apache.spark.sql.functions._
import org.apache.spark.sql.test.TestSQLContext
import org.apache.spark.sql.test.TestSQLContext.{sparkContext, sql}
import org.apache.spark.sql.test.TestSQLContext.implicits._
import org.apache.spark.sql.types._
@SQLUserDefinedType(udt = classOf[MyDenseVectorUDT])
private[sql] class MyDenseVector(val data: Array[Double]) extends Serializable {
override def equals(other: Any): Boolean = other match {
case v: MyDenseVector =>
java.util.Arrays.equals(this.data, v.data)
case _ => false
}
}
@BeanInfo
private[sql] case class MyLabeledPoint(
@BeanProperty label: Double,
@BeanProperty features: MyDenseVector)
private[sql] class MyDenseVectorUDT extends UserDefinedType[MyDenseVector] {
override def sqlType: DataType = ArrayType(DoubleType, containsNull = false)
override def serialize(obj: Any): Seq[Double] = {
obj match {
case features: MyDenseVector =>
features.data.toSeq
}
}
override def deserialize(datum: Any): MyDenseVector = {
datum match {
case data: Seq[_] =>
new MyDenseVector(data.asInstanceOf[Seq[Double]].toArray)
}
}
override def userClass = classOf[MyDenseVector]
}
class UserDefinedTypeSuite extends QueryTest {
val points = Seq(
MyLabeledPoint(1.0, new MyDenseVector(Array(0.1, 1.0))),
MyLabeledPoint(0.0, new MyDenseVector(Array(0.2, 2.0))))
val pointsRDD = sparkContext.parallelize(points).toDF()
test("register user type: MyDenseVector for MyLabeledPoint") {
val labels: RDD[Double] = pointsRDD.select('label).rdd.map { case Row(v: Double) => v }
val labelsArrays: Array[Double] = labels.collect()
assert(labelsArrays.size === 2)
assert(labelsArrays.contains(1.0))
assert(labelsArrays.contains(0.0))
val features: RDD[MyDenseVector] =
pointsRDD.select('features).rdd.map { case Row(v: MyDenseVector) => v }
val featuresArrays: Array[MyDenseVector] = features.collect()
assert(featuresArrays.size === 2)
assert(featuresArrays.contains(new MyDenseVector(Array(0.1, 1.0))))
assert(featuresArrays.contains(new MyDenseVector(Array(0.2, 2.0))))
}
test("UDTs and UDFs") {
TestSQLContext.udf.register("testType", (d: MyDenseVector) => d.isInstanceOf[MyDenseVector])
pointsRDD.registerTempTable("points")
checkAnswer(
sql("SELECT testType(features) from points"),
Seq(Row(true), Row(true)))
}
}
|