I meant that he made the regex filter unique cases just like a set :)

(I on the other hand used both which is slower..)

When you will have complex cases, you will love it :D

Why you want he put a set ? It's work properly like that. Add a set will affect performance.

It will work cause it's not a condition for a panagram. This solution properly work's.

Genius

nice, But I wonder if there will be any duplicate letters？

nice

Nice work putting the "set" inside the regex :)

i hate Regexp, but is very friendly in cases like this

I dont even know what I am looking at
this is literally when maths in high school suddenly includes letters lmfao

Scroll down the comments to one that was made about 6 years ago. I've replied with a detailed explanation that should help you understand.

Use regex101.com for analyzing regular expressions.

I did string.match(/[a-z]/g) but it return just amount of letter this string contains(not amount of unique letters), why it wasnt working? Can anyone explain please

That's a very clever solution! Although, I think it's running in O(n^2) right?

nice! knew that regex would be the shortest way but was looking for a one liner and here I am !