Splunk Search

Extract data from JSON array

karlbosanquet
Path Finder

I have some data which is along the following format;

{"event":
     {
     "Timestamp":"2019-01-16 22:20:26.123"
     "Stream":"123456",
     "Id":"789",
     "Data": [
         {
        "name":"0",
            "rank":"1",
            "score":"34.56787"
         },
         {
        "name":"1",
            "rank":"3",
            "score":"4.74287"
         },
         {
        "name":"2",
            "rank":"2",
            "score":"12.54863"
         },
         {
        "name":"3",
            "rank":"4",
            "score":"1.78916"
         },

     ]
     }
}

We get around 800,000 of these per day and have around 50 data elements in each one.

I am trying to find the best way to return the top 2 rank name and score for each event, e.g.;

1_name = 0
1_score = 34.56787
2_name = 2
2_score = 12.54863

And another search to timechart all scores by name

0 Karma
1 Solution

mayurr98
Super Champion

Hi Can you try this

| makeresults 
| eval _raw = "{\"event\":
      {
      \"Timestamp\":\"2019-01-16 22:20:26.123\",
      \"Stream\":\"123456\",
      \"Id\":\"789\",
      \"Data\": [
          {
         \"name\":\"0\",
             \"rank\":\"1\",
             \"score\":\"34.56787\"
                  },
          {
         \"name\":\"1\",
             \"rank\":\"3\",
             \"score\":\"4.74287\"
                  },
          {
         \"name\":\"2\",
             \"rank\":\"2\",
             \"score\":\"12.54863\"
                  },
          {
         \"name\":\"3\",
             \"rank\":\"4\",
             \"score\":\"1.78916\"
                  },

      ]
      }
 }" 
| spath 
| rename event.Data{}.* as * event.Timestamp as Timestamp event.Id as Id event.Stream as Stream 
| eval data=mvzip(mvzip(name,rank),score) 
| mvexpand data 
| rex field=data "(?<name>[^\,]+)\,(?<rank>[^\,]+)\,(?<score>.*)" 
| table Timestamp Stream Id name rank score

Also try this for top 2 scores:

|stats values(score) as score by rank name 
|  sort 2 -score

let me know if this helps!

View solution in original post

0 Karma

mayurr98
Super Champion

Hi Can you try this

| makeresults 
| eval _raw = "{\"event\":
      {
      \"Timestamp\":\"2019-01-16 22:20:26.123\",
      \"Stream\":\"123456\",
      \"Id\":\"789\",
      \"Data\": [
          {
         \"name\":\"0\",
             \"rank\":\"1\",
             \"score\":\"34.56787\"
                  },
          {
         \"name\":\"1\",
             \"rank\":\"3\",
             \"score\":\"4.74287\"
                  },
          {
         \"name\":\"2\",
             \"rank\":\"2\",
             \"score\":\"12.54863\"
                  },
          {
         \"name\":\"3\",
             \"rank\":\"4\",
             \"score\":\"1.78916\"
                  },

      ]
      }
 }" 
| spath 
| rename event.Data{}.* as * event.Timestamp as Timestamp event.Id as Id event.Stream as Stream 
| eval data=mvzip(mvzip(name,rank),score) 
| mvexpand data 
| rex field=data "(?<name>[^\,]+)\,(?<rank>[^\,]+)\,(?<score>.*)" 
| table Timestamp Stream Id name rank score

Also try this for top 2 scores:

|stats values(score) as score by rank name 
|  sort 2 -score

let me know if this helps!

0 Karma

karlbosanquet
Path Finder

Expected output examples;

  • Search 1 (table timestamp Stream Id 1_name 1_value 2_name 2_value)

||Timestamp||Stream||Id||1_name||1_score||2_name||2_score||
|2019-01-16 22:20:26.123|123456|789|0|34.56787|2|12.54863|
|2019-01-16 22:21:28.567|123456|790|12|54.78797|46|34.82632|

  • Search 2 (timechart values(score) by name)

||Timestamp||0||1||2||3||4||5||
|2019-01-16 22:20:26.123|34.56787|4.74287|12.54863|1.78916|3.35732|
|2019-01-15 12:21:34.244|2.34689|12.45211|21.24852|14.72412|2.28342|

0 Karma
Got questions? Get answers!

Join the Splunk Community Slack to learn, troubleshoot, and make connections with fellow Splunk practitioners in real time!

Meet up IRL or virtually!

Join Splunk User Groups to connect and learn in-person by region or remotely by topic or industry.

Get Updates on the Splunk Community!

Developer Spotlight with Denis Gladkikh

From Splunk Engineer to Kubernetes App Builder Denis GladkikhWhat happens when a lifelong developer turns a ...

Governing Enterprise AI, Bringing Cisco Telemetry Home, and More from Splunk Lantern

Splunk Lantern is Splunk’s customer success center that provides practical guidance from Splunk experts on key ...

Agentic SOC Triage: Investigating Splunk ES Notables with MCP Server and a Local LLM

The Problem: Too Many Alerts, Too Little Context Security operations teams running Splunk Enterprise Security ...