From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:36246) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bzL5u-0002MD-2z for qemu-devel@nongnu.org; Wed, 26 Oct 2016 06:06:02 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bzL5q-0001cw-Vo for qemu-devel@nongnu.org; Wed, 26 Oct 2016 06:06:02 -0400 Received: from mx1.redhat.com ([209.132.183.28]:46750) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1bzL5q-0001ca-Ps for qemu-devel@nongnu.org; Wed, 26 Oct 2016 06:05:58 -0400 Message-ID: <1477476355.18984.32.camel@redhat.com> From: Gerd Hoffmann Date: Wed, 26 Oct 2016 12:05:55 +0200 In-Reply-To: <20161025222616.GE2801@var.home> References: <20161015195308.20473-1-samuel.thibault@ens-lyon.org> <20161025222616.GE2801@var.home> Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Mime-Version: 1.0 Subject: Re: [Qemu-devel] [PATCH 0/5] curses: wide character support List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Samuel Thibault Cc: qemu-devel@nongnu.org, pbonzini@redhat.com, berrange@redhat.com, peter.maydell@linaro.org, mjt@tls.msk.ru On Mi, 2016-10-26 at 00:26 +0200, Samuel Thibault wrote: > Hello, >=20 > Ping? Back online now after being sick for a while. Sorry for the delay. > Samuel Thibault, on Sat 15 Oct 2016 21:53:03 +0200, wrote: > > This patch series adds wide character support to the curses frontend of= qemu, > > thus allowing to fix a lot of input and output issues with e.g. accente= d letters > > and semi-graphic glyphs. Since qemu can't know the encoding of the VGA = font, the > > user has to specify it (just like he has to specify the keyboard layout= with > > -k). I used option -f to make it simple for now, but I welcome any othe= r idea :) >=20 > I forgot to mention that I updated the patches according to the reviews > on the list (assume cursesw support, and use -display option). Good, looks much better now without all the #ifdefs. /me wonders whenever we should do the same with iconv. It's part of the POSIX.1-2001 specs, and we have 2016 now. Do we *really* need configure checks and #ifdefs for it? Also we could use iconv for the cp437-to-unicode mapping instead of having a hard-coded table for it. Given how close the freeze deadline is now I'm tempted to cherry-pick patches 0-3 and prepare a pull request ASAP. cheers, Gerd