From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.org!.POSTED!not-for-mail From: Eli Zaretskii Newsgroups: gmane.emacs.devel Subject: Re: Emacs text shaping using Harfbuzz Date: Thu, 13 Dec 2018 21:48:24 +0200 Message-ID: <83o99pgqd3.fsf@gnu.org> References: <83lg4tzba0.fsf@gnu.org> <241e97ba-e547-9184-b798-1692be469740@cs.ucla.edu> NNTP-Posting-Host: blaine.gmane.org X-Trace: blaine.gmane.org 1544730471 28058 195.159.176.226 (13 Dec 2018 19:47:51 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Thu, 13 Dec 2018 19:47:51 +0000 (UTC) Cc: dr.khaled.hosny@gmail.com, ebraminio@gmail.com, behdad@behdad.org, johnw@gnu.org, far.nasiri.m@gmail.com, emacs-devel@gnu.org, handa@gnu.org To: Paul Eggert Original-X-From: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Thu Dec 13 20:47:46 2018 Return-path: Envelope-to: ged-emacs-devel@m.gmane.org Original-Received: from lists.gnu.org ([208.118.235.17]) by blaine.gmane.org with esmtp (Exim 4.84_2) (envelope-from ) id 1gXWxV-0007Ac-TD for ged-emacs-devel@m.gmane.org; Thu, 13 Dec 2018 20:47:46 +0100 Original-Received: from localhost ([::1]:54549 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gXWzc-00015S-OU for ged-emacs-devel@m.gmane.org; Thu, 13 Dec 2018 14:49:56 -0500 Original-Received: from eggs.gnu.org ([2001:4830:134:3::10]:45148) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gXWyc-00014R-8f for emacs-devel@gnu.org; Thu, 13 Dec 2018 14:48:57 -0500 Original-Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gXWyb-0000C7-Ak for emacs-devel@gnu.org; Thu, 13 Dec 2018 14:48:54 -0500 Original-Received: from fencepost.gnu.org ([2001:4830:134:3::e]:40262) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gXWyI-0008Rw-9K; Thu, 13 Dec 2018 14:48:34 -0500 Original-Received: from [176.228.60.248] (port=2727 helo=home-c4e4a596f7) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1gXWyG-0002xs-JJ; Thu, 13 Dec 2018 14:48:34 -0500 In-reply-to: <241e97ba-e547-9184-b798-1692be469740@cs.ucla.edu> (message from Paul Eggert on Thu, 13 Dec 2018 10:47:18 -0800) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Received-From: 2001:4830:134:3::e X-BeenThere: emacs-devel@gnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: "Emacs development discussions." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-devel-bounces+ged-emacs-devel=m.gmane.org@gnu.org Original-Sender: "Emacs-devel" Xref: news.gmane.org gmane.emacs.devel:231804 Archived-At: > Cc: Khaled Hosny , > Ebrahim Byagowi , Kenichi Handa , > John Wiegley , Mohammad Nasirifar , > Behdad Esfahbod > From: Paul Eggert > Date: Thu, 13 Dec 2018 10:47:18 -0800 > > The Emacs harfbuzz branch built fine for me on Fedora 29 (no surprise). > I saw no problems, not that I would catch many (the Arabic I saw > displayed OK to me, but I don't read Arabic so am not the best person to > judge). Thanks for trying the branch. > One minor tweak proposed in the attached patch, for simplicity/clarity > in the mainstream-Harbuzz code. IMO, for such obviously correct fixes, you should just go ahead and push. > I see that Harbuzz is disabled by default by 'configure', and that > there's an environment variable to disable it at runtime. I propose > removing the environment variable, and enabling Harfbuzz by default at > 'configure' time (if the Harbuzz libraries are available), as I don't > see much point in nudging users away from Harfbuzz. I agree to enabling it by default if the necessary libraries are found. As for disabling at run time, I think this should be reworked to use the same framework as with other font back-ends, which can be disabled at runtime by invoking Emacs with the appropriate -xrm switch. AFAIU, currently the Harfbuzz code just piggy-backs xftfont, but it really should be a separate font backend, IMO. > I see there are some FIXMEs and suchlike comments. Will someone be > working on them? I very much hope so ;-) The branch is supposed to be WIP, not a finished result, not yet anyway.