UTF-8 and wchar_t

[Joerg Sonnenberger]

Hi all,
reviewing some libarchive code I stumpled about the code that converts
UTF-8 to wide strings.  As done by a lot of other software, it currently
blindly assumes that wchar_t ~= UCS-4.  My question is whether FreeBSD
intentionally makes that decision what (and therefore should define
__STDC_ISO_10646__ according to ISO C99) or what correct way for reading
UTF-8 it allows.  Contrary to NetBSD, FreeBSD still lacks iconv(3)
support in base, so the usual approach of converting to the locale
charset and using mbtowc etc. is not possible.

Joerg

PS: Please keep me in CC.