classification builder - regex for multiple values that adjust

[scottt6073095]

Hi All,

Having some difficulties figuring out how to create regex that adjusts to the number of groups.

I can have any number of parts ranging from 1 to 4

strings:

events:company-conferences/company-user-conference/2018

events:company-conferences/company-user-conference

events:company-conferences

I tried this regex however it only returns the correct values if it has 4 capture groups, no more no less.

^([^\:]*)\:([^\:]*)\/([^\:]*)\/([^\:]*)

I tried the below regex however it does the same

^(.+)\:(.+)\/(.+)\/(.+)

is there regex logic that is flexible to the number of capture groups in a string? I would like to capture string that have groups up to 4 but may have 3 or 2.

any help would be highly appreciated.

thanks

[JimKultgen]

I do something similar in which I essentially have a different rule for either 5 or 6 fields.  Not sure if there is a more robust way but this should be easy to build from for 2/3/4 fields.  As an FYI my deliminator is an underscore.

^((.+)\_(.+)\_(.+)\_(.+)\_(.+)\_(.+)|(.+)\_(.+)\_(.+)\_(.+)\_(.+))$

[spencerh7639004 · Accepted solution]

Something like : ^(.+)\:(.+?)(\/|$)(.*?)(\/|$) (repeating (.*?)(\/|$) to the number of possible results will get you in even numbered fields the results you want.  So for 1-4:

^(.+)\:(.+?)(\/|$)(.*?)(\/|$)(.*?)(\/|$)(.*?)(\/|$)

You can see the results here: https://regexr.com/3sjh5

[Pablo_Childe]

Check out this thread see if it helps.

Regex help in Classification Rule Builder