Re: Unicode and C++

[Per Hedbor <per idonex se>]

> Excuse me, but filenames, when in unicode, are written in utf-8.
> UTF-8 was specifically designed to the purpose of allowing unicode in
> filenames, another name of utf-8 isn't "file system safe UCS transformation
> format" ?
> And if the file names are not in unicode, then using UCS2 or UCS4 will
> need a conversion anyway.

Consider:

we have the file 'åke'. 
This file is selected by a GTK file-selector.
This results in the string 'Ã¥ke'.
Now, if we try to do some operation on this string, we have problems.

The reverse is also true, when you read that the username is 
'Åke Rök' from the passwd database, you have to convert it to utf8
before you can use it.

So, all that I am saying is that all GTK 1.4 programs that read data
from external sources (files etc) have to be modified anyway.

-- 
Per Hedbor                                   http://per.hedbor.org/