@postmodern I recognize that surrogate codepoint prefix (0xD8) from some Unicode work I was doing a couple of years ago. Javascript uses UTF-16 internally, so JavaScript strings use surrogate pairs for any codepoints over 0xffff. The simplest approach might be to just use JSON.parse, e.g: JSON.parse '"\\ud83d\\ude01"' # => "😁"

@nick_evans now I just need to figure out the other UTF-8 / JSON related bug, how to embed binary Strings into generated JSON. Apparently JSON is UTF-8 by default, so it will convert any binary string to UTF-8, attempt to escape special chars, and then hit an invalid byte sequence.

@postmodern JSON is a silly encoding! 😂 Mandates UTF-8, but also mandates surrogate pairs! I don't think binary data is natively possible in JSON. You just need to use Base64 (or some other UTF-8 safe encoding) for anything like that.