<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8">
<META NAME="GENERATOR" CONTENT="GtkHTML/3.28.3">
</HEAD>
<BODY>
Hi,<BR>
<BR>
narrow no-break space chars (lets say nnbsp) are used in French before ";?!»" chars and after "«". It is also used in other language, for example for the short form of the Czech dates.<BR>
<BR>
Currently, when I switch to a TTY and type a nnbsp, it displays a kind of diamond (Ubuntu 10.04).<BR>
(if you can't easily type a nnbsp with your keyboard layout, run "/usr/bin/printf '\u202Fx\n'")<BR>
<BR>
I had a look in kbd .uni files, and I didn't found any sign of u+202f<BR>
So I propose to just use nnbsp as U+0020 (white space). The same was done for "no-break space" (U+00a0) and it works.<BR>
<BR>
Someone pointed to me that this behavior is consistent with <TT><FONT COLOR="#737373">t</FONT></TT>he <TT><FONT COLOR="#737373">'wcwidth' function of glibc:</FONT></TT><BR>
<BR>
This program:<BR>
<TT><FONT COLOR="#737373">=============================================================</FONT></TT><BR>
<TT><FONT COLOR="#737373">#include <locale.h></FONT></TT><BR>
<TT><FONT COLOR="#737373">#include <wctype.h></FONT></TT><BR>
<TT><FONT COLOR="#737373">#include <wchar.h></FONT></TT><BR>
<TT><FONT COLOR="#737373">#include <stdio.h></FONT></TT><BR>
<TT><FONT COLOR="#737373">int main () {</FONT></TT><BR>
<TT><FONT COLOR="#737373"> setlocale (LC_ALL, "fr_FR.UTF-8");</FONT></TT><BR>
<TT><FONT COLOR="#737373"> printf ("wcwidth (0x00A0) = %d\n", wcwidth (0x00A0));</FONT></TT><BR>
<TT><FONT COLOR="#737373"> printf ("wcwidth (0x202F) = %d\n", wcwidth (0x202F));</FONT></TT><BR>
<TT><FONT COLOR="#737373"> return 0;</FONT></TT><BR>
<TT><FONT COLOR="#737373">}</FONT></TT><BR>
<TT><FONT COLOR="#737373">=============================================================</FONT></TT><BR>
return<BR>
<TT><FONT COLOR="#737373">wcwidth (0x00A0) = 1</FONT></TT><BR>
<TT><FONT COLOR="#737373">wcwidth (0x202F) = 1</FONT></TT><BR>
<BR>
<BR>
I don't know if adding this support is just as simple as adding "u+202f" at the 0x20 line of each .uni file (if so I can provide a patch if you want).<BR>
<BR>
Thank you for your attention.<BR>
Nicolas<BR>
<BR>
ps: it also works after running these commands:<BR>
setfont -v -ou unimap<BR>
echo '0x20 U+202f' >> unimap<BR>
setfont -v -u unimap
</BODY>
</HTML>