UTF-8 encoder/decoder title is misleading

[zvodd]

This is problematic.
There are people posting on stack overflow and being generally confused by what UTF-8 is.
In part, thanks to the complete lack of clarification on the page https://mothereff.in/utf-8

[mathiasbynens]

Could you post a link to an example Stack Overflow post where this confusion occurs?

Would it help to link to https://encoding.spec.whatwg.org/#utf-8?

[zvodd]

Yes, sorry, I had intended to elaborate earlier.
I think it would be helpful to have small disclaimer, explaining that the app/page encodes JavaScript strings (which are UTF-8) into escaped hexadecimal byte strings, and vise versa.

"Disclaimer:
JavaScript, which encoder is made with, internally represents all strings in UTF-8 format. This page converts UTF-8 strings to escaped hexadecimal values (byte literals), e.g. "\x41\x42\x43" and vise versa. It also decodes escaped Unicode literals e.g. "\u0041\u0042\u0043" into plain strings.
Consider viewing this link for more information about UTF-8 and Unicode."
...or something like that.

I think otherwise, new players will get the notion that hexadecimal escape sequences are the definition of UTF-8.

https://stackoverflow.com/search?q=https%3A%2F%2Fmothereff.in%2Futf-8
https://stackoverflow.com/questions/60839049/is-there-any-function-to-get-utf-8-encoding-output-text-in-python

[mathiasbynens]

JavaScript, which encoder is made with, internally represents all strings in UTF-8 format.

This is false. See https://mathiasbynens.be/notes/javascript-encoding.

This page converts UTF-8 strings to escaped hexadecimal values (byte literals), e.g. "\x41\x42\x43" and vise versa.

This is also incorrect. The page takes Unicode strings as input and converts them to their UTF-8-encoded byte sequence, in the format \x41\x42\x43.

It also decodes escaped Unicode literals e.g. "\u0041\u0042\u0043" into plain strings.

The tool does no such thing. Are you thinking of https://mothereff.in/js-escapes?

[zvodd]

My mistake, on the internal representation.

On the second point, fair and accurate, as they are Unicode strings. 👍
My point about an explanation being included in the page still stands.

On the third, I was just reporting the observed behavior of the script. So, not sure about that?
Thought it was an intended feature.
https://i.imgur.com/4VALCkr.png

[mathiasbynens]

@zvodd What's the URL you're seeing that on, and in which browser are you seeing this? https://mothereff.in/utf-8#ABC produces the expected output of \x41\x42\x43 for me.

[zvodd]

The user input is from the bottom so user input is \u0041\u0042\u0043; and upon close inspection the decoder seems to have a weird case of accepting string parts beginning with "\u" followed by 2 "0" and a 2 digit hex number, and interprets them as if it were reading the escaped version of that 2 digit hex number.
So that part of my description was also inaccurate.

Let me try to be clear, I am not hear to try to tell you I know how you encoder works better than you. I am just saying that I think the description of UT8 encoding on the page should let users know that the "\x00" format of output is a hexadecimal format escape string and not exclusively related to UTF-8.

My initial attempt at the explanation was an example with a number of inaccuracies, for which I apologize.