> it won't parse u-escaped invalid codepoints: ["\ud800"] How is this not expected behaviour? The string is not well-formed. Same thing with decent XML parse

> it won't parse u-escaped invalid codepoints: ["\ud800"] How is this not expected behaviour? The string is not well-formed. Same thing with decent XML parsers. They croak when you give them invalid codepoints.