Data360 Analyze

Hi,
I'm currently tyring to use the graph that list all users from: https://support.infogix.com/hc/en-us/articles/360042748714-How-to-List-Users-in-Analyze

However, i get an error when i try to parse the Json for the users.

The error says: 

Field "lastName" does not exist on output (0) "out1". Error in "ProcessRecords" at line "10".

I've tried to add it as a variable in the configureFields part of the Transformer node but then i simply get another error due to a Unknown Type (using str).

The Json looks as followed :

{
  "users" : [ {
    "user" : {
    "id" : "xxxxx-yyyy-zzzz-aaaa-bbbb",
    "tenant" : "defaultTenant",
    "name" : "John",
    "imported" : true,
    "active" : true,
    "state" : "ACTIVE",
    "principalImportDetails" : {
    "uniqueImportReference" : "xxxx,yyyy,zzzz",
    "importSource" : "xxxx,yyyy,zzz",
    "removedFromSource" : false
    },
    "attributes" : null,
    "firstName" : null,
    "lastName" : null,
    "emailAddress" : null
   }
  }
 ]
}

The code that is used in the Transformer node looks like this (Default from the downloaded graph):
ConfigureFields

out1 += in1
out1.id=str
out1.name=str
out1.principalImportDetails=str
out1.active=str
out1.state=str
out1.imported=str
out1.tenant=str

out1 -= in1.BaseURL
out1 -= in1.LTK
out1 -= in1.URL
out1 -= in1.response
out1 -= in1.StatusCode
out1 -= in1.StatusMessage

import json

ProcessRecords

response=json.loads(fields.response)
if in1.StatusCode<>200:
  node.logger.error("Failed:"+in1.StatusMessage)
  raise node.fail()
out1+=in1

users=response['users']
for j in users:
  user=j['user']
  out1 += user
  node.write(0,out1)

When i manually extract all the fields it works but i assume this should work automatically (assuming the graph on the site was correct).
The bottom 4 fields about the FirstName, Lastname, Attributes and emailAddress never work.
When i use a JSON data node. It extracts all the fields properly except the 4 null value fields at the bottom.

Am i missing something here? How do i get those last 4 fields and why do i get the error in the first place when i run the graph as it was without chaning the transformer node?

Thanks,
Jurgen

The addtibutes, firstName,lastName and emailAddress elements were added to a user's profile in a relatively recent release. It looks like that data flow has not been modified to handle the presence of the additional fields.

You can modify the script in the 'Parse Json' Transform node's ConfigureFields property so that they are handled correctly by inserting the following lines:

out1.attributes=unicode
out1.firstName=unicode
out1.lastName=unicode
out1.emailAddress=unicode

The above changes have been made in the attached data flow.

Attached files

D3SA_List_Users_367 - 3 Feb 2021.lna

The JSON Data node does not output a field when all of the values of a particular attribute are null.

<x-zendesk-user data-user-name="Adrian Williams">366746295667</x-zendesk-user> Thank's a lot for this info! Good to know that the JSON Data node ignores attributes if they are all null.

Regarding the Transform node, i actualy tried to do this but got an error. I get the same if i run the model you've attached. It says: 

java.lang.IllegalArgumentException: java.lang.IllegalArgumentException: Unknown type: dict Error executing "ProcessRecords" at line "10"

I assume its having problems saving a null into unicode but cant tell for sure (very new to both d360 and python)