Convert flattened DataFrame to nested JSON

This article explains how to convert a flattened DataFrame to a nested structure, by nesting a case class within another case class.

You can use this technique to build a JSON file, that can then be sent to an external API.

Define nested schema

We’ll start with a flattened DataFrame.

Example flattened DataFrame

Using this example DataFrame, we define a custom nested schema using case classes.

case class empId(id:String)
case class depId(dep_id:String)
case class details(id:empId,name:String,position:String,depId:depId)
case class code(manager_id:String)
case class reporting(reporting:Array[code])
case class hireDate(hire_date:String)
case class emp_record(emp_details:details,incrementDate:String,commission:String,country:String,hireDate:hireDate,reports_to:reporting)

You can see that the case classes nest different data types within one another.

Convert flattened DataFrame to a nested structure

Use to pass every row object to the corresponding case class.

import spark.implicits._
val nestedDF=>{
val empID_1= empId(r.getString(0))
val depId_1 = depId(r.getString(7))
val details_1=details(empID_1,r.getString(1),r.getString(2),depId_1)
val code_1=code(r.getString(3))
val reporting_1 = reporting(Array(code_1))
val hireDate_1 = hireDate(r.getString(4))


This creates a nested DataFrame.

Example nested DataFrame

Write out nested DataFrame as a JSON file

Use the repartition().write.option function to write the nested DataFrame to a JSON file.


Example notebook

Run the example notebook to see each of these steps performed.

DataFrame to nested JSON example

Open notebook in new tab