Instructions are clear

2 years later... description updated! :-)

Did you read any of the intro comments within the sample test section?

"www.xakep.ru", no prefix "http://" nor suffix ".com"

Just a guess, but "markers" can actually be more than 1 character.
Sadly this does not explain the error completely cos' you do escape the markers.

I agree. I think that at least should be 6 or 5th Kyu. The tests cases put an pretty good emphasis on performance.

Edit:
My bad, no performance issues here, just that the feedback from the test cases is all messed up in those invalid chunk size edge cases. It always says process took too long. I would suggest to pass those test cases upfront, and just leave the random ones hidden.

Is this sarcasm?

The test case is correct.

It's true that all the letter are there, but that's not the only requirement. :-)

The regex picks any character followed by an uppercase letter and sets two groups

\1 = [any character]

\2 = [the uppercase letter that follows]

Then it replaces this pattern by the format "\1_\2"

So, for the string 'CodeWarsPascalCase' it will pick 'eW', 'rP' and 'lC' and replace with the new given pattern ('e_W', 'r_P' and 'l_C')

Pretty clever, indeed.

5 +4 is 9 right ;)

The input may have 0's in unexpected places - try making the ranges 0-9 and see if that is the issue...

Can you show your code?

Which language???