Read this: https://www.codewars.com/kata/5a34af40e1ce0eb1f5000036/discuss#6478df8fe58a25005c6ddb2d

You have the same problem. The third post below yours.

expected
'0,1,2,3,45\n10,11,12,13,14\n20,21,22,…'
to equal
'0,1,2,3,45\n10,11,12,13,14\n20,21,22,…'
Why....

Your current latest solution will always return "" after the first character in text is read. Also, the output format is also wrong for res

OP solved it, closing

Your code is wrong, once again. And tested with the same input (for God's sake, learn that for once) it will fail in VSCode too. Your code doesn't have a problem with toLowerCase at all, read the error message:

Testing ",vx:+xi?"
expected ' 22 24   24 9 ' to equal '22 24 24 9'

And this part of the description again:

If anything in the text isn't a letter, ignore it and don't return it.

Your code adds some spaces when it shouldn't.

My code as always works in VSCode but doesn't work here, unfortunately. My arg can't transform to LowerCase in JavaScript.

The mistake is once again in your code, see which input makes your code fail and debug your code, please.

Looks there is a mustake in current KATA. My code in VSCode get different inouts between here in JavaScript.

Had some issues between basic tests and random tests, seems conditions are not correct!

As always, mu code works in VSCode but have an error here:

expected 'INVALID' to equal +0

JavaScript:
RangeError: Maximum call stack size exceeded

I've tried a lot of variants, but didn't realize exactly what to do? Awful explanation of Kata...

Read the description again, your code isn't 100% right, and please stop posting "as always it works in vscode and not here" if you don't use the exact same input in both places, that's what would happen.

The kata is fine; I am the 12,880th person to succesfully solve it in JavaScript.

This means you rather instead should not like the incomplete coverage of tests in your VSCode environment.

Don't like this test, as always passed in VSCode and basic solutions but don't pass random tests.