Splunk Search
Highlighted

How to count Max Sub-sequence of identical numbers?

Explorer

Hello ,

I need to calculate the maximum length of identical numbers
for example : 0,0,0,0,0,1,0,1,1,0,0 and search for the sequence of 0, the result should be 7 in this case

Anyone have any ideas how this could be accomplished?

0 Karma
Highlighted

Re: How to count Max Sub-sequence of identical numbers?

Path Finder

Can you give more examples to this? I can't see how you would get to 7 based on that data - maybe I'm missing something?

0 Karma
Highlighted

Re: How to count Max Sub-sequence of identical numbers?

Explorer

because i want to count the sequence of zeroes the sequence goes like this:

the values:0,0,0,0,0,1,0,1,1,0,0,
the result : 3,4,5,6,7,-, 1,- ,-,1,2

the zeros at the start continues the zeroes at the end
each value is in a different row same column

0 Karma
Highlighted

Re: How to count Max Sub-sequence of identical numbers?

I used the data you supplied in your example to populate a little table with a single row/field called digit. If we didn't have to account for wrapping around the end of the list (the way you get to 7 in your example), it would be really straightforward. But here's a code snippet that achieves what you laid out:

| makeresults 
| eval digit="0,0,0,0,0,1,0,1,1,0,0" 
| makemv delim="," digit 
| mvexpand digit 
| fields - _time 
| streamstats count BY digit reset_on_change=true 
| eventstats first(digit) AS first_digit, last(digit) AS last_digit 
| eventstats max(count) AS max_count BY digit 
| eventstats last(count) AS final_count 
| eval total_count=if(first_digit=last_digit AND first_digit=digit, max_count+final_count, max_count) 
| fields digit, total_count

You can adjust it for other digit lists by adjusting the second line - or customize it to match your data source by editing/removing the first five lines.

View solution in original post

0 Karma
Highlighted

Re: How to count Max Sub-sequence of identical numbers?

Explorer

thank you it works !

0 Karma