it gives me this output:
> STRING: text
> > STRING: \r\ntest
But other implementations normally will unescape it, and it is more efficient to unecape it during the process of parsing.
As it isn't exactly a bug this tracker item was bump to feature request.