Re: [whatwg/url] Add Unicode ToASCII fallback for ASCII domains (PR #914)

hsivonen left a comment (whatwg/url#914)

I can see why this is being done, but I don't like this, since this introduces a different whole-domain mode, which means that implementations that with the previous spec text could produce output as they did a single pass over the domain now need a different API shape.

It looks to me like _domain to ASCII_ and _domain to Unicode_ still reject different domains. I think both operations should accept or reject consistently. (Of course, in actual browsers, the _domain to Unicode_ operation does not exist and the operation is really _domain to display_, but we should keep the implementation-defined stuff to whether _domain to display_ chooses to output Punycode for a label that _domain to Unicode_ outputs Unicode for. I think we shouldn't have implementation-dependent behavior other than that.)

It looks to me like the new spec text bypasses forbidden domain code point enforcement when the input is all-ASCII. That should be fixed.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/url/pull/914#issuecomment-4791010240
You are receiving this because you are subscribed to this thread.

Message ID: <whatwg/url/pull/914/c4791010240@github.com>