[Freeciv-Dev] Re: (PR#13664) On the romanization of Japanese names

["Jason Short" <jdorje@xxxxxxxxxxxxxxxxxxxxx>]

<URL: http://bugs.freeciv.org/Ticket/Display.html?id=13664 >

Daniel Markstedt wrote:

> As Unicode indeed is not completely universal (sic) yet I will of course 
> avoid more exotic accented latin in rulesets in the future.

Charset support depends entirely on the GUI client.

gui-gtk-1.2 uses the local charset.  However this client is removed in 2.1.

gui-gtk-2.0 uses all unicode (UTF-8).  Some characters may still not 
show properly if the user doesn't have sufficient fonts installed 
(particularly for Kanji).  We could get around this problem by including 
a fontset in our distribution.

gui-win32 uses the standard windows charset (which varies by location). 
  It could be rewritten to use "unicode" (UCS-2 I think) but I doubt 
this will happen.  However gui-win32 uses transliteration for 
unsupported characters (for an English user this includes a lot of the 
east-European characters), and since gui-win32 uses GNU's libiconv 
transliteration support is good.

In 2.1, gui-xaw uses UTF-8 also (I think?).  In 2.0 it also used the 
local charset.

-jason