>>467
See http://perldoc.perl.org/perlunicode.html#Porting-code-from-perl-5.6.X.
The charset situation in Wakaba is a confusing mess. IIRC, the wrapper functions are either used to retain compatibility with Perl 5.6, or to retrieve Unicode data from a non-Unicode database, or both. Note how the utf8
pragma isn't used anywhere in the script, while the encoding
pragma is used sparsely.
It's been a few years since I've done anything with Wakaba, but I recall having to use get_decoded_hashref()
or its arrayref
counterpart every time I wanted to retrieve text that might contain Unicode.