I think you might be right. Either way, the description I ended up with for this kata was way too confusing. I'm also not sure why I settled on allowing words to be preceded by non-word characters in the description instead of just ^ \n \r I believe I had in mind that words could "contain" hyphens but not end with them, but that's not fully clear from the kata description. Tempted to completely remake this without allowing spec creep

Have updated the description to make this clearer. Thanks :)

Have updated RNG to ensure numbers are produced!

I wasn't able to recreate the reaction time issue in repl.it, and adding the first two of your examples as static test cases on here passes as expected (with "11.34 seconds" and "25.18 seconds" respectively). Could you tell me if you still get this issue? I'll avoid resolving it for now

Silly me forgot to look into the string issue before publishing (as I'd put a temporary fix in my solution). Will update!

Thanks for bringing up the reaction time issue. Unpublishing now to fix

You're definitely right @JohanWiltink ! I'm going to see if I can improve the description this evening, before I resort to updating the whole solution/specification again! I reckon I should be able to improve on what's currently there.

@Firefly2002 thanks for the heads-up :). To be honest I'd be interested to see how someone might attempt this efficiently without RegEx

Apologies, I meant for single letters to count. I have now clarified this as an additional bullet point for the "word" definition! Hope that's clearer now. Thanks again!

Have updated the description (first bullet point under "Word boundaries") - had included a description along those lines but had to move it to the "scrapped rules" section. Hope that works!

I believe this is now resolved (see updated description and solution). Let me know if you disagree or spot any other issues. You're highly encouraged to attempt the new version :). Thanks again!

I believe this is now resolved (see updated description and testing). Let me know if you disagree or spot any other issues. Thanks again!

You're highly encouraged to attempt the new version :).

I believe this is now resolved (see updated description). Let me know if you disagree or spot any other issues. Thanks again!

You're highly encouraged to attempt the new version :).

I believe that - as far as regex patterns are concerned - the very start of a string still matches. You're right, though. I should make this clear in the description.

As for the end of a word, I will define that as ending with whitespace, a full stop, an exclamation mark, or a closed parenthesis.

Once this has been added in, would you consider the issue resolved?

Thanks again for the feedback! Really useful.

Thanks. Will do (to the best of my ability/knowledge)!

Thanks for the feedback! I'm not sure I follow you, though. Unless I'm misunderstanding something...

In my aim for this kata, "zzZzz zzZzz" should be untouched, as "zzZzz" is treated as a whole "invalid" word.

I'm planning to update the description with a note to say that a "word" must preceded by whitespace or an open parenthesis. Do you feel that this would clear the issue up? Note that - with this - the solution will be changed slightly, so you may want to hold off on any more attempts for a little while.