UNIX for Advanced & Expert Users

m using unix c with curses
im trying out the conversion of an existing code in c to i18n for the japanese language.
the problem is that, the input which is given as the japanese characters is not taken correctly.
and what it outputs is question marks(i.e. ???), equivalent to the number o f japanese characters.
im doing this in ja_JP.utf8 locale

it check for a pattern matching in the switch case, but it does not enter the particular case..

help me at the earliest

Certainly. My psychic powers are telling me that your code is failing because...

Wait, I don't have psychic powers. Dang.

Could we see your code please?

Try to verify the codepoints you get getting are really Unicode codepoints, as you can cross check the character-codepoint mappings with the Unicode character database online, and that your console really supports multibyte UTF-8 characters. Bear in mind that you may need to do some conversion between the legacy Japanese encoding and UTF-8 with iconv in certain environments. Also check whether you have the Unicode fonts which contain the Japanese characters on your OS, and whether those fonts are indeed used to render the console text.

hello cbkihong....

can u please tell me the procedure to do so..... i mean how to do conversion between utf-8 and japanese encoding...
the code is working fine on ja_JP.eucJP locale but not on ja_JP.utf8 locale

As my C memory is fading, maybe I will give you some directions rather than actual code. I am not familiar with the Japanese encodings actually, so I will state the general way of handling it.

As I said, the first thing before you start is to check whether you really are getting UTF-8 character code sequence from your input. Do you have gucharmap (Gnome Unicode character map) installed? It will help you check the UTF-8 representation quickly by entering a character from search.

http://gucharmap.sourceforge.net/

Also check your terminal that it can support the UTF-8 encoding with proper font. Check that you can read the content of a UTF-8 encoded Japanese text file, for instance, by "cat" should normally be sufficient to test it.

Then you can enter a Japanese character as input in your program, then read the char* from the input and use sprintf() to dump the character code. Just check that the sequence matches the UTF-8 character sequence then you know your input is in Unicode. Note that Japanese characters are represented in UTF-8 3 bytes each.

Encoding conversion between legacy JP encoding and UTF-8 is only needed if your input and output encodings differ. If you are sure both are UTF-8, then you don't need encoding conversion. Encoding conversion can be conducted with the iconv library.

You can find more information in the following documents:
http://www.tldp.org/HOWTO/Unicode-HOWTO.html
http://www.debian.org/doc/manuals/intro-i18n/

You may also find this thread with an example on iconv useful:
https://www.unix.com/showthread.php?t...ighlight=iconv

For background information on Unicode, you should find certain chapters (especially the introductory chapters) from the Unicode specification useful:
http://www.unicode.org/versions/Unicode4.1.0/