anthr anthr - 4 months ago 89
SQL Question

Explode (transpose?) multiple columns in Spark SQL table

I am using Spark SQL (I mention that it is in Spark in case that affects the SQL syntax - I'm not familiar enough to be sure yet) and I have a table that I am trying to re-structure, but I'm getting stuck trying to transpose multiple columns at the same time.

Basically I have data that looks like:

userId someString varA varB
1 "example1" [0,2,5] [1,2,9]
2 "example2" [1,20,5] [9,null,6]


and I'd like to explode both varA and varB simultaneously (the length will always be consistent) - so that the final output looks like this:

userId someString varA varB
1 "example1" 0 1
1 "example1" 2 2
1 "example1" 5 9
2 "example2" 1 9
2 "example2" 20 null
2 "example2" 5 6


but I can only seem to get a single explode(var) statement to work in one command, and if I try to chain them (ie create a temp table after the first explode command) then I obviously get a huge number of duplicate, unnecessary rows.

Many thanks!

Answer

What you want is not possible without a custom UDF. In Scala you could do something like this:

val data = sc.parallelize(Seq(
    """{"userId": 1, "someString": "example1",
        "varA": [0, 2, 5], "varB": [1, 2, 9]}""",
    """{"userId": 2, "someString": "example2",
        "varA": [1, 20, 5], "varB": [9, null, 6]}"""
))

val df = sqlContext.read.json(data)

df.printSchema
// root
//  |-- someString: string (nullable = true)
//  |-- userId: long (nullable = true)
//  |-- varA: array (nullable = true)
//  |    |-- element: long (containsNull = true)
//  |-- varB: array (nullable = true)
//  |    |-- element: long (containsNull = true)

Now we can define zip udf:

import org.apache.spark.sql.functions.{udf, explode}

val zip = udf((xs: Seq[Long], ys: Seq[Long]) => xs.zip(ys))

df.withColumn("vars", explode(zip_udf($"varA", $"varB"))).select(
   $"userId", $"someString",
   $"vars._1".alias("varA"), $"vars._2".alias("varB")).show

// +------+----------+----+----+
// |userId|someString|varA|varB|
// +------+----------+----+----+
// |     1|  example1|   0|   1|
// |     1|  example1|   2|   2|
// |     1|  example1|   5|   9|
// |     2|  example2|   1|   9|
// |     2|  example2|  20|null|
// |     2|  example2|   5|   6|
// +------+----------+----+----+

With raw SQL:

sqlContext.udf.register("zip", (xs: Seq[Long], ys: Seq[Long]) => xs.zip(ys))
df.registerTempTable("df")

sqlContext.sql(
  """SELECT userId, someString, explode(zip(varA, varB)) AS vars FROM df""")