.match() function suddenly not working

Brock Jolet_wkfaaa

I built an Automation rule a while back that allows certain people to manually trigger a request to transition an issue to a hidden status if it's already linked to another issue.

 

The basic setup is as follows:

• User triggers rule
• The user is asked to provide a text input which gets stored as {{userInputs.ticketNum}}
• I then parsed the input for a ticket number and saved it as a variable called {{ticketNum}} using this smart value: {{userInputs.ticketNum.upperCase.match("([A-Z]+-\d+)")}}
  • That forced all caps and stripped anything not matching the correct ticket format so that users could copy and paste a URL for example.
• I then used an If/Else block to verify that {{ticketNum}} matched the regex 

  [A-Z]+-\d+

• If correct, I would proceed to transition the issue as requested.

 

This worked for some time, but wasn't used often. It looks like it was last successfully executed on 2025-08-14. At some point between then and today, it stopped working.

 

Doing some testing, I learned that any use of .upperCase now returns a blank value and similarly .match() also returns a blank value. Thankfully, I was able to find .toUpperCase() as an alternative that works, but there is no alternative to .match().

 

I tried passing the raw input to my If/Else block and can confirm that the regex is correct. However, I need a method of stripping the extraneous parts of the string so that I can pass the key into a Lookup.

Bill Sheboy

Hi @Brock Jolet 

Without seeing your entire rule and audit log details...

I hypothesize the problem is any extra characters you do not want to match upon are not handled with the regular expression passed to the match() function.  Perhaps try adding some tokens outside of the selection, writing the results to the audit log for testing, such as:

{{userInputs.ticketNum.trim().toUpperCase().match("^.*([A-Z]+-\d+).*")}}

 

And as a reminder, there is no automation documentation for what actually is (or is not) supported for regular expressions in automation rules, and so experimentation is required.  What the docs say, with emphasis added by me, is:

match()

Performs a regular expression search and returns the first (and only one) matching regular expression group.

 

The underlying implementation is based on Java's Pattern class and uses Matcher.find() to find matches. If multiple matches are found, they return as a collection that can be iterated...

 

Kind regards,
Bill