Opening files with UNICODE file names

Hi all,

I would like to know how to open files with UNICODE file names. Is there any function that accepts a wchar_t as file name and returns a file descriptor of that file?

Thanks,
Alan

if your linux system is configured to support unicode, then the standard file open will handle the filename.

most linux systems can easily have unicode set as default charset, so everything on the system is encoded in unicode.
[ lord knows all my linux boxes are unicode only, it helps with deciphering those ms codepage encoded crap files ]

Unicode is just a name for a group of standards, despite MS's use of the word. What is known as Unicode in Windows is really just the UCS-2 (I think, or perhaps UTF-16) encoding of the Unicode character set.

Under *nix, the typical Unicode encoding used is UTF-8, a variable size encoding using a single byte as the base unit (a char, unlike UCS-2, which uses a wchar_t). The wchar_t in *nix is typically actually a 32-bit value and uses the UTF-32 encoding by default.

On the other hand, I believe wxWidgets uses UCS-2 or UTF-16 in its APIs, as might some other wrapper libraries. GTK+ uses GLib's UTF-8 string types. Qt is annoyingly vague about the encoding it uses (it keeps referring to it as "Unicode encoding" in the 3.1 docs), but as far as I can see it's UCS-2 or UTF-16 - a QChar is 16 bits wide.

So, what do you want?