Managing Inputs with same name but different type

Users
1

Hi,
I have a client that fetches JSON from a webservice and then stores in in DGraph to be examined.

I’ve come across a case where the JSON contains to two fields with the same name but different types i.e. in the case below state is both an object and a string.


{
    Name : "Test",
    Monitoring : {
        State : "disabled"
    },
   State :  {
       Code : 16,
       Name : "running"
   }
}

When I submit this to DGraph I get an error ‘Input for predicate State of type scalar is uid’. Is there a specific way to deal with this - is there any way to ensure objects would be prefixed as they get submitted to Dgraph?

2

As “Monitoring” and “State” in your JSON are obviously Nodes. We have the “State” predicate set to UID. Therefore you can not perform a mutation for the “State” predicate with another Type. What I recommend is that you create another predicate (Alias) and then use an Alias in the query.

Nops, It is almost impossible to predict what users intend with a JSON obj. It would need an AI to handle it.

Mutation

{
  "set": [
    {
    "Name" : "Test",
    "Monitoring" : {
         "State_Monitoring" : "disabled"
  		  },
    "State" :  {
          "Name" : "running"
   }
   }
  ]
}

Query


{
  q(func: has(Monitoring)) {
    uid
    Name 
    Monitoring { uid State:State_Monitoring }
    State {uid Name}
  }
}

Result


{
  "data": {
    "q": [
      {
        "uid": "0x8",
        "Name": "Test",
        "Monitoring": [
          {
            "uid": "0x9",
            "State": "disabled"
          }
        ],
        "State": [
          {
            "uid": "0x7",
            "Name": "running"
          }
        ]
      }
    ]
  }

Cheers.

3

Hi MichelDiz, thanks I figured that was the only option. I was hoping Dgraph had something built in, but I can just pre-process the json to deal with this.

4

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.