[erlang-questions] [erlang-bugs] xmerl and unicode data

[Anthony Ramine]

It isn't weird. It is an iolist, a nested list of bytes and binaries,
not a list of characters. As the encoding of an XML document can only
be known by actually reading it, why should the input have some predefined
charset embedded in its type?

Regards,

-- 
Anthony Ramine

Le 22 oct. 2012 à 16:27, Patrik Nyblom a écrit :

> Me too - it's really weird. Rather the input format is weird, lists of UTF-8 characters are usually considered "broken". I would wrap this in a function taking a proper Unicode list as a parameter...