Palindromes

“Was it a car or a cat I saw?” reads the same forward and backward once you ignore spaces, punctuation and case. Checking it properly takes two steps: normalize the text first, then compare its two ends.

Normalizing means converting the letters to lowercase and removing everything that isn't a letter or a digit. The regex /[^\p{L}\p{N}]/gu handles the removal for every alphabet at once, and \p{…} only works with the u flag. Accents need one extra step: text.normalize('NFD') splits an accented letter like é or ή into its plain letter followed by a separate accent mark, and the same regex then removes the mark, because a mark isn't a letter. After that, use two pointers that start at the two ends and move toward each other comparing characters, stopping at the first mismatch, so you never need a reversed copy of the text.

Watch out: the regex /[^a-z0-9]/g deletes every letter outside a to z, accented letters included, so a Greek sentence turns into an empty string. An empty string is a palindrome, so the function answers “yes” for any Greek sentence.