this is a lightweight and fast unicode detector i made for use in one of my future projects. it runs a lot faster than qucchia's detector and takes up a lot less JSON space (on the order of 260KB total). it works for all* characters in UTF-16; or every character with codepoints in ranges from 0 to 65535 (\u0000 to \uFFFF). the main improvements this has from other unicode detectors i've seen on scratch: - the project knows when a character isnt case sensitive at all and thus eliminate case checking for no reason. - it binary searches instead of using item# which helps a lot with speed. - the detector takes less JSON space than others. - can handle the 'weird' characters like "NJNjnj", and the characters with code points less than 32 if you have a use for that. idk what else to say here *some characters have been a bit funny in other browsers like chromium. i have fixed them i think however, but if there are bugs then tell me. there is a 'test' option in the project which will test the validity of the detector by attempting to encode and decode every character. any failures will be printed out (if there are any besides \u0008 then let me know). note that the character \u0008 is weird for some reason and gets pruned from scratch.* ok so \u0008 is partially fixable due to the ability to obtain that character through the translate extension of all things. will require internet to obtain however.
