if the input is negative, won't this add an extra instruction?

Fixed.

Some O(1) solutions can pass random tests on 3-5 run, they are not precise! But right solutions with less than O(nlogn) don't pass 12s timeout, although in my editor they executes less than half sec. Thats nice kata, but its broken

Will this be fixed anytime soon?

PHP "expected" and "actual" the wrong way around.

Asking for translations is not a suggestion.

PHP translate please. I'm very satisfied this kata series

replacement may contain references of the form \n or $n, with the latter form being the preferred one. Every such reference will be replaced by the text captured by the n'th parenthesized pattern. n can be from 0 to 99, and \0 or $0 refers to the text matched by the whole pattern. Opening parentheses are counted from left to right (starting from 1) to obtain the number of the capturing subpattern. Note that backslashes in string literals may require to be escaped.

This kata need to implement own div? No bcmath for bcdiv. Looks like extra complexity unrelated to the task. Please indicate at least the accuracy in the description to which you want to round

Haha, I didn’t even think that there is a finite and such a small number of solutions. Meme

Before php 7.4 % works bad with negative numbers. PHP 8.0 server throws 422