spark-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chris Teoh <chris.t...@gmail.com>
Subject Re: Convert RDD[Iterrable[MyCaseClass]] to RDD[MyCaseClass]
Date Sat, 01 Dec 2018 11:17:00 GMT
Hi James,

Try flatMap (_.toList). See below example:-

scala> case class MyClass(i:Int)
defined class MyClass

scala> val r = 1 to 100
r: scala.collection.immutable.Range.Inclusive = Range(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, 95, 96, 97, 98, 99, 100)

scala> val r2 = 101 to 200
r2: scala.collection.immutable.Range.Inclusive = Range(101, 102, 103, 104,
105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119,
120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134,
135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149,
150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164,
165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179,
180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194,
195, 196, 197, 198, 199, 200)

scala> val c1 = r.map(MyClass(_)).toIterable
c1: Iterable[MyClass] = Vector(MyClass(1), MyClass(2), MyClass(3),
MyClass(4), MyClass(5), MyClass(6), MyClass(7), MyClass(8), MyClass(9),
MyClass(10), MyClass(11), MyClass(12), MyClass(13), MyClass(14),
MyClass(15), MyClass(16), MyClass(17), MyClass(18), MyClass(19),
MyClass(20), MyClass(21), MyClass(22), MyClass(23), MyClass(24),
MyClass(25), MyClass(26), MyClass(27), MyClass(28), MyClass(29),
MyClass(30), MyClass(31), MyClass(32), MyClass(33), MyClass(34),
MyClass(35), MyClass(36), MyClass(37), MyClass(38), MyClass(39),
MyClass(40), MyClass(41), MyClass(42), MyClass(43), MyClass(44),
MyClass(45), MyClass(46), MyClass(47), MyClass(48), MyClass(49),
MyClass(50), MyClass(51), MyClass(52), MyClass(53), MyClass(54),
MyClass(55), MyClass(56), MyClass(57), MyClass(58), MyClass(59), MyClass(...

scala> val c2 = r2.map(MyClass(_)).toIterable
c2: Iterable[MyClass] = Vector(MyClass(101), MyClass(102), MyClass(103),
MyClass(104), MyClass(105), MyClass(106), MyClass(107), MyClass(108),
MyClass(109), MyClass(110), MyClass(111), MyClass(112), MyClass(113),
MyClass(114), MyClass(115), MyClass(116), MyClass(117), MyClass(118),
MyClass(119), MyClass(120), MyClass(121), MyClass(122), MyClass(123),
MyClass(124), MyClass(125), MyClass(126), MyClass(127), MyClass(128),
MyClass(129), MyClass(130), MyClass(131), MyClass(132), MyClass(133),
MyClass(134), MyClass(135), MyClass(136), MyClass(137), MyClass(138),
MyClass(139), MyClass(140), MyClass(141), MyClass(142), MyClass(143),
MyClass(144), MyClass(145), MyClass(146), MyClass(147), MyClass(148),
MyClass(149), MyClass(150), MyClass(151), MyClass(152), MyClass(153),
MyClass(154), MyClass(15...
scala> val rddIt = sc.parallelize(Seq(c1,c2))
rddIt: org.apache.spark.rdd.RDD[Iterable[MyClass]] =
ParallelCollectionRDD[2] at parallelize at <console>:28

scala> rddIt.flatMap(_.toList)
res4: org.apache.spark.rdd.RDD[MyClass] = MapPartitionsRDD[3] at flatMap at
<console>:26

res4 is what you're looking for.


On Sat, 1 Dec 2018 at 21:09, Chris Teoh <chris.teoh@gmail.com> wrote:

> Do you have the full code example?
>
> I think this would be similar to the mapPartitions code flow, something
> like flatMap( _ =>  _.toList )
>
> I haven't yet tested this out but this is how I'd first try.
>
> On Sat, 1 Dec 2018 at 01:02, James Starks <suserft@protonmail.com.invalid>
> wrote:
>
>> When processing data, I create an instance of RDD[Iterable[MyCaseClass]]
>> and I want to convert it to RDD[MyCaseClass] so that it can be further
>> converted to dataset or dataframe with toDS() function. But I encounter a
>> problem that SparkContext can not be instantiated within SparkSession.map
>> function because it already exists, even with allowMultipleContexts set to
>> true.
>>
>>     val sc = new SparkConf()
>>     sc.set("spark.driver.allowMultipleContexts", "true")
>>     new SparkContext(sc).parallelize(seq)
>>
>> How can I fix this?
>>
>> Thanks.
>>
>
>
> --
> Chris
>


-- 
Chris

Mime
View raw message