topic Re: character encoding in Operating System - Linux

character encoding

CA1497567 — Mon, 09 Jul 2007 17:24:29 GMT

As far as I know C++ uses ANSI in order to represent its characters.
How can I read a character in greek, obtain its unicode character and write it again as a string or as a character in console.

I'm desperate with this issue. I have tried with wchar_t but I cannot find the way to use it properly and I cannot write my characteres in a terminal.

Thanks you in advance.

Re: character encoding

A. Clay Stephenson — Mon, 09 Jul 2007 17:30:42 GMT

Your problem doesn't stem from the language but rather with the display device. You need to understand that all the host computer does is send octets (or groups or octets in the case of wide characters) to the output device (terminal, printer, plotter, ...) and how those octets appear is solely a function of the output device itself.

The typical way this is done is to use curses to select the appropriate alternate character set.

Re: character encoding

CA1497567 — Tue, 10 Jul 2007 01:30:27 GMT

Can you give an example how to use curses, or a link?

Thanks.

Re: character encoding

Dennis Handly — Tue, 10 Jul 2007 07:10:17 GMT

First of all, you need to set LANG appropriately. Then try to read and write char strings.

To read a character that's multibyte and convert to unicode, you must use mbtowc(3), or mbstowcs(3) for strings.

To convert that wchar_t string to utf8, you would have to use iconv(3):
cd1 = iconv_open("utf8", "ucs4");
ret = iconv(cd1, &in, &inl, &out, &outl);