Help with custom response handler for REST API Modular input.

[quihong]

Hello,

I've been tasked with ingesting some audit events from a online application (Lever Hire). I'm looking at using the REST API Modular input and need help parsing the data. Originally I thought I had to create a new sourcetype to do the parsing, but now I believe the correct method is to create a custom Response Handler.

I've found some examples here: https://github.com/damiendallimore/SplunkModularInputsPythonFramework/blob/master/implementations/re...

However, I'm not familiar enough with python to write my own response handler. I'm hoping someone in the community can quickly help me with the code.

Here is what the raw data looks like. Any help is appreciated.

{
    "data": [{
        "id": "5b628f1e-2bcf-45f7-90fa-7b1264987d42",
        "user": {
            "role": "super admin",
            "id": "8810816c-03da-48db-b3c1-d47a8f5c024f",
            "name": "Joe Mama",
            "email": "[email protected]"
        },
        "type": "key:added",
        "createdAt": 1515609233213,
        "target": {
            "type": "key",
            "id": "8d0501a2-f613-4154-b2e7-fc4b416ad213",
            "label": "Splunk"
        },
        "meta": {
            "key": {
                "tokenLastFour": "w2jU",
                "id": "8d0501a2-f613-4154-b2e7-fc4b416ad213",
                "name": "Splunk",
                "partner": false,
                "service": "data-api"
            }
        }
    }, {
        "id": "85374119-2af3-48b2-838f-7821fb15ef7c",
        "user": {
            "role": "super admin",
            "id": "8810816c-03da-48db-b3c1-d47a8f5c024f",
            "name": "Joe Mama",
            "email": "[email protected]"
        },
        "type": "key:removed",
        "createdAt": 1515609175385,
        "target": {
            "type": "key",
            "id": "21b0fb88-006d-4a9a-a1e1-2164fcd8d243",
            "label": "Splunk"
        },
        "meta": {
            "key": {
                "tokenLastFour": "RhgT",
                "id": "21b0fb88-006d-4a9a-a1e1-2164fcd8d243",
                "name": "Splunk",
                "partner": false,
                "service": "data-api"
            }
        }
    }, {
        "id": "b368c76a-f7a5-4cc8-8201-ce4051847976",
        "user": {
            "role": "super admin",
            "id": "8810816c-03da-48db-b3c1-d47a8f5c024f",
            "name": "Joe Mama",
            "email": "[email protected]"
        },
        "type": "user.authentication:succeeded",
        "createdAt": 1515609122117,
        "target": {
            "type": "user",
            "id": "8810816c-03da-48db-b3c1-d47a8f5c024f",
            "label": "Joe Mama"
        },
        "meta": {
            "user": {
                "role": "super admin",
                "id": "8810816c-03da-48db-b3c1-d47a8f5c024f",
                "name": "Joe Mama",
                "email": "[email protected]"
            },
            "authentication": {
                "method": "direct"
            }
        }
    }, {
        "id": "5b88b646-f141-4be7-a970-e39c56ce13ad",
        "user": {
            "role": "super admin",
            "id": "lever-support",
            "name": "Lever Support",
            "email": "[email protected]"
        },
        "type": "key:added",
        "createdAt": 1515520786845,
        "target": {
            "type": "key",
            "id": "82cedc33-87ff-4d68-bc44-7dcc7559da4c",
            "label": "click-boarding"
        },
        "meta": {
            "key": {
                "tokenLastFour": "RAof",
                "id": "82cedc33-87ff-4d68-bc44-7dcc7559da4c",
                "name": "click-boarding",
                "partner": false,
                "service": "data-api"
            }
        }
    }, {
        "id": "c4ef90e5-449d-4a2b-a724-8cde900f1a1f",
        "user": {
            "role": "super admin",
            "id": "cd6751d7-998a-451b-ab22-fb2e0fa96da5",
            "name": "superman",
            "email": "[email protected]"
        },
        "type": "user.authentication:succeeded",
        "createdAt": 1515456274871,
        "target": {
            "type": "user",
            "id": "cd6751d7-998a-451b-ab22-fb2e0fa96da5",
            "label": "superman"
        },
        "meta": {
            "user": {
                "role": "super admin",
                "id": "cd6751d7-998a-451b-ab22-fb2e0fa96da5",
                "name": "superman",
                "email": "[email protected]"
            },
            "authentication": {
                "method": "direct"
            }
        }

    }],
    "hasNext": false
}

[Damien_Dallimor · Accepted solution]

Handler example below.

Then use standard Splunk timestamp extraction in props.conf for your sourcetype to use the createdAt time as the prefix.

[yoursourcetype]
TIME_PREFIX = createdAt": 

Handler

class ExampleHandler:

    def __init__(self,**args):
        pass

    def __call__(self, response_object,raw_response_output,response_type,req_args,endpoint):
        if response_type == "json":        
            output = json.loads(raw_response_output)

            for item in output["data"]:
                print_xml_stream(json.dumps(item))   
        else:
            print_xml_stream(raw_response_output)

[quihong]

Thank you very much!

Had to escape the quotes, other than that perfect.
[yoursourcetype]
TIME_PREFIX =\" createdAt\":


[Damien_Dallimor]

Please describe what it is you want the custom response handler to do with the raw JSON ?

[quihong]

Sorry I was not clear...

I would like the custom response handler to break out the raw json into individual events with proper timestamp (createdAt field). Each event starts with the {"id":