headers
Janusz S. Bień | 12 Aug 08:24
Picon
Picon

Re: emacs-bidi for MS Windows?

On Tue, 9 Jul 2002  Eli Zaretskii <eliz <at> is.elta.co.il> wrote:

> On Tue, 9 Jul 2002, TAKAHASHI Naoto wrote:
> 
> > > My main and important questions is: should Emacs-bidi compile with
> > > Mingw and work on MS Windows?
> > 
> > No idea.  It has been tested only with the X Window System.
> 
> One problem with the Windows port of Emacs is that some of the Windows 
> output routines do logical-to-visual reordering internally.  Try 
> displaying etc/HELLO on Windows in an unpatched Emacs 21.x and look at 
> the Hebrew greeting: it will be displayed correctly in visual order, 
> right to left, even though Emacs sent the logical-order string to the 
> display!  Now move point with C-f across the Hebrew greeting, and you 
> will see it displayed in the logical order(!).

I think this should be considered as a bug in NT Emacs 21.x and
properly reported.  You are definitely the right person to do it.

> 
> So I expect some problems with bidi Emacs on Windows.  The 
> Windows-specific display code should somehow disable the automatic 
> reordering performed by the Windows runtime.
> 

Let's make the NT Emacs team aware of the problem as soon as possible.

Best regards
(Continue reading)

Permalink | Reply | 1 comment |
headers
Eli Zaretskii | 12 Aug 19:24
Picon

Re: emacs-bidi for MS Windows?

> From: jsbien <at> mimuw.edu.pl (Janusz S. =?iso-8859-2?q?Bie=F1?=)
> Date: 12 Aug 2002 08:24:50 +0200
> > 
> > One problem with the Windows port of Emacs is that some of the Windows 
> > output routines do logical-to-visual reordering internally.  Try 
> > displaying etc/HELLO on Windows in an unpatched Emacs 21.x and look at 
> > the Hebrew greeting: it will be displayed correctly in visual order, 
> > right to left, even though Emacs sent the logical-order string to the 
> > display!  Now move point with C-f across the Hebrew greeting, and you 
> > will see it displayed in the logical order(!).
> 
> I think this should be considered as a bug in NT Emacs 21.x and
> properly reported.  You are definitely the right person to do it.

I already did, when I discovered the problem about a year ago.

Of course, it never does any harm to repeat the report.  But as long
as bidi Emacs is not a reality, there's no urgent need to bug the
NTEmacs maintainers about this matter.
Permalink | Reply | 0 comments |

Gmane