[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: The return of the acute-cedilla BUG FROM HELL



On May 24, 2004, Z <zleite mminternet com> wrote:

> FC2 has again the bug on the us-acentos keymap. Pressing "'" + c gives
> an non-existing symbol (at least non-existing in the languages I know)
> instad of the cedilla.

Well, Ä surely exists in some languages, and you have to agree that it
would be damn surprising if Ä were to prefer Ä over Ã.  Why the heck
is the acute accent *under* the letter, one would think...

If your locale is pt (or pt_BR?), gtk apps will map 'c to Ã, but X
will still compose 'c into Ä.  That's bad, and inconsistent.  The
solution (untested) is to create a file in
/usr/lib/X11/locale/pt_BR.UTF-8/Compose, adapted from
/usr/lib/X11/locale/en_US.UTF-8/Compose, in which the combinations of
<dead_acute> and <c> or <C> are mapped to the à and à characters,
instead of Ä and Ä as they are.  Then adjust compose.dir in the parent
directory such that pt_BR.UTF-8 is mapped to this new Compose rules.

Maybe there's a way to create a specialization of the Compose rules; I
don't know.

Then, work to get this change into upstream Xorg, and it will be fixed
in whatever Fedora Core release happens to integrate the Xorg release
that has your change.


Personally, I just got used to entering <Multi_key> <,> <c> to
generate Ã.  Being a native pt_BR speaker, and writing a
non-negligible amount of e-mail in Portuguese, I don't find it to be
too much of a pain, as long as you know about it, and use a reasonable
key for <Multi_key> (AKA Compose).  I use the AltGR key for this
purpose; others choose the Win-key, or the Menu key.  You can easily
select one of them in the Layout Options in Keyboard Preferences.

> I saw some bugs related to this in bugzilla. Shoud I file it anyway?

You're probably better off filing an enhancement request with upstream
Xorg.  We don't have a bug here, just an inconvenience that takes some
getting-used-to.

> It seems to be a reversion from the xorg migration.

As far as I can tell you're mistaken.  From personal experience, FC1
(and probably RHL9) worked just the same in this regard, at least as
far as X11 is concerned.  I haven't checked for changes in gtk within
pt_BR locales, though; this might have changed.  Maybe you had
different i18n settings.  For example, switching from ISO-8859-1 to
UTF-8, or from pt_BR to en_US would have changed the 'c compose rules
on at least some applications.

-- 
Alexandre Oliva             http://www.ic.unicamp.br/~oliva/
Red Hat Compiler Engineer   aoliva {redhat com, gcc.gnu.org}
Free Software Evangelist  oliva {lsd ic unicamp br, gnu.org}



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]