Reformat JSON with Groovy. Remove child node whilst keeping the contents - json

Question

I am trying to reformat JSON to remove child nodes from some JSON objects whilst keeping whats inside the child node. (unsure how to explain, like removing the title node). In this example i am trying to have a JSONarray of groups without the UserGroup child node (whilst keeping each UserGroups content intact). Example JSON
{
"groups":[
{
"UserGroup":{
"integrationKey":"0000073807",
"uid":"0000073807"
},
"UserGroup":{
"integrationKey":"0000073810",
"uid":"0000073810"
}
}
]
}
What I'd like after processing
{
"groups":[
{
"integrationKey":"0000073807",
"uid":"0000073807"
},
{
"integrationKey":"0000073810",
"uid":"0000073810"
}
]
}

Answer 1

For the JSON with no duplicated keys, which formats the data, removing child keys may look like
import groovy.json.JsonSlurper
import groovy.json.JsonOutput
def inputJson = '''
{
"groups": [{
"UserGroup1": {
"integrationKey": "0000073807",
"uid": "0000073807"
},
"UserGroup2": {
"integrationKey": "0000073810",
"uid": "0000073810"
}
}
]
}
'''
Map parsed = new JsonSlurper().parseText(inputJson) as Map
parsed.each {
parsed[it.key] = it.value*.values().flatten()
}
println JsonOutput.toJson(parsed)
This code will print
{"groups":[{"integrationKey":"0000073807","uid":"0000073807"},{"integrationKey":"0000073810","uid":"0000073810"}]}
Handle duplicated key
But your provided input example contains 2 items with UserGroup name. It's not common for JSON and most parsers just take the last mentioned value for the key. To handle this you can use below dependency
<dependency>
<groupId>net.sf.json-lib</groupId>
<artifactId>json-lib</artifactId>
<version>2.4</version>
<classifier>jdk15</classifier>
</dependency>
And modify the above script to
import net.sf.json.JSONObject
// ...
ipuptJson = JSONObject.fromObject(inputJson).toString()
Map parsed = new JsonSlurper().parseText(inputJson) as Map
parsed.each {
parsed[it.key] = it.value*.values().flatten()
}
println JsonOutput.toJson(parsed)

Answer 2

Starting from terminology, you are not really modifiyng the json itself. In fact you can modify the in-memory representation of it exposed as a mixture of Lists, Maps and primitive object types.
The simplest way to do what you want would be:
import groovy.json.*
def json = new JsonSlurper().parseText '''
{
"groups": [{
"UserGroup1": {
"integrationKey": "0000073807",
"uid": "0000073807"
},
"UserGroup2": {
"integrationKey": "0000073810",
"uid": "0000073810"
}
}
]
}
'''
def modified = [ groups:json.groups*.values().flatten() ]
println JsonOutput.prettyPrint( JsonOutput.toJson( modified ) )
prints:
{
"groups": [
{
"integrationKey": "0000073807",
"uid": "0000073807"
},
{
"integrationKey": "0000073810",
"uid": "0000073810"
}
]
}