Bugs in wide character API

[Marcin 'Qrczak' Kowalczyk]

Hello. I found the following bugs (31 July snapshot):

1. add_wch ignores attributes and colors of the cchar.

2. get_wch doesn't recode a printing character from the locale encoding
   to Unicode if the locale is different from UTF-8 (or ISO-8859-1 when
   no recoding is needed).

3. get_wstr and friends take wint_t * instead of wchar_t * as their
   parameter. X/Open specification seems to have a bug: it shows
   wint_t * in one place (a page with get_wstr), but wchar_t * in
   another (a page with the whole family). wchar_t * is the only type
   which makes sense.

(I made an interface to ncurses from my language Kogut. It uses wide
character API if available, narrow character API otherwise. Kogut uses
Unicode internally, so wide character API is important not only for
UTF-8 terminals.)

-- 
   __("<         Marcin Kowalczyk
   \__/       address@hidden
    ^^     http://qrnik.knm.org.pl/~qrczak/