Splunk Search

Need Count of values in multivalued field, grouped by another field in json data

rajeshku348
New Member

hi everyone,
I need count of "id" field against the sequence field

parentRecord sequence count(id)

ABC162856 2 1
ABC162856 38 4

I am using spath to extract the multivalued field but it clubs all "id" fields together, so i am unable to group by "sequence" field. I have tried extracting via REX as well but not been able to make it work. I have searched Q&A but couldn't find a similar query.

...
| spath input=jsonpayload output=ids path=data.relationships.elements.data{}.records.data{}.id
| spath input=jsonpayload output=parentID path=data.parentRecord
| spath input=jsonpayload output=sequence path=data.relationships.elements.data{}.sequence
| table parentID, sequence
| stats count(ids) by parentRecord, sequence

my query's result:

sequence count(ids)

2 5
38 5

data:

{
"data": {
"relationships": {
"elements": {
"data": [
{
"sequence": "2",
"records": {
"data": [
{
"id": "ABC62856"
}
]
},
},
{
"sequence": "38",
"records": {
"data": [
{
"id": "ABC9247022"
},
{
"id": "ABC161214"
},
{
"id": "ABC162549"
},
{
"id": "ABC162679"
},
]
},
}
]
}
},
"parentRecord": "ABC162856"
}
}

0 Karma

poete
Builder

Hello @rajeshku348,

I found a few errors in the json, and fixed it.

Here is the solution to your problem.

| makeresults 
| eval _raw="{\"data\": {\"relationships\": {\"elements\": {\"data\": [{\"sequence\": \"2\",\"records\": {\"data\": [{\"id\": \"ABC62856\"}]}},{\"sequence\": \"38\",\"records\": {\"data\": [{\"id\": \"ABC9247022\"},{\"id\": \"ABC161214\"},{\"id\": \"ABC162549\"},{\"id\": \"ABC162679\"}]}}]}},\"parentRecord\": \"ABC162856\"}}"
| spath path=data.relationships.elements.data{}
| rename data.relationships.elements.data{} as datas
| mvexpand datas
| spath input=datas
| rename records.data{}.id as id
| table sequence,id
0 Karma

rajeshku348
New Member

I was able to do it as below but I guess there must be some better way of doing this using spath directly instead of having to use rex

| rex field=_raw max_match=200 "records\": {\"data\": [(?.*?)]},"
| spath input=jsonpayload output=parentRecord path=data.id
| eval z=mvzip(sequence1,ids)
| mvexpand z
| makemv z delim=","
| eval sequence = mvindex(z,0)
| eval recordCount = mvcount(z) -1
| table _time, parentRecord , sequence, recordCount

0 Karma
Get Updates on the Splunk Community!

Index This | What’s a riddle wrapped in an enigma?

September 2025 Edition Hayyy Splunk Education Enthusiasts and the Eternally Curious!  We’re back with this ...

BORE at .conf25

Boss Of Regular Expression (BORE) was an interactive session run again this year at .conf25 by the brilliant ...

OpenTelemetry for Legacy Apps? Yes, You Can!

This article is a follow-up to my previous article posted on the OpenTelemetry Blog, "Your Critical Legacy App ...