Do these two test satisfy your needs?

It’s an ACF thing - Lucee doesn’t have it as far as I know.

Funnily enough I saw today that escaping for regular expressions is a Stage 3 proposal for ECMAScript. Here are some links if of interest.

Cool, here’s CF code I’ve used for this in the past (I TOTALLY didn’t know this BIF existed!)

We can probably just adapt that.

I found it due to our ACF tests failing when running on BoxLang. I think it might be related to POSIX classes…?

ACF will output:

********[:upper:]

BoxLang will output:

[:upper:]****[:upper:]

So an edge case!

Is this causing an actual issue. The BoxLang output (which comes directly from the Java regex libs is technically correct. Here’s ChatGTP’s explaination:

In Java regex, \Q...\E is used to quote (escape) a sequence of characters, treating them as literal text rather than as special regex metacharacters.

So, the regex \Q[]{}()\E is equivalent to:

CopyEdit

\[\]\{\}\(\)

which matches exactly the string []{}(), interpreting the brackets, braces, and parentheses as literal characters instead of their usual regex meanings.