Received: from mout.gmx.net (mout.gmx.net [212.227.17.22]) by h1439878.stratoserver.net (8.14.4/8.14.4/Debian-2ubuntu2.1) with ESMTP id t6DHC9qm003163 for ; Mon, 13 Jul 2015 19:12:10 +0200 Received: from relay.uni-heidelberg.de ([129.206.100.212]) by mx-ha.gmx.net (mxgmx103) with ESMTPS (Nemesis) id 0Ll0Uz-1YgBrl0FYD-00apqK for ; Mon, 13 Jul 2015 19:12:04 +0200 Received: from listserv.uni-heidelberg.de (listserv.uni-heidelberg.de [129.206.100.94]) by relay.uni-heidelberg.de (8.14.1/8.14.1) with ESMTP id t6DHAYoL028102 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Mon, 13 Jul 2015 19:10:34 +0200 Received: from listserv.uni-heidelberg.de (listserv.uni-heidelberg.de [127.0.0.1]) by listserv.uni-heidelberg.de (8.13.8/8.13.8) with ESMTP id t6DFJ5KB014507; Mon, 13 Jul 2015 19:10:34 +0200 Received: by LISTSERV.UNI-HEIDELBERG.DE (LISTSERV-TCP/IP release 16.0) with spool id 12399525 for LATEX-L@LISTSERV.UNI-HEIDELBERG.DE; Mon, 13 Jul 2015 19:10:33 +0200 Received: from relay.uni-heidelberg.de (relay.uni-heidelberg.de [129.206.100.212]) by listserv.uni-heidelberg.de (8.13.8/8.13.8) with ESMTP id t6DHAXDo026352 for ; Mon, 13 Jul 2015 19:10:33 +0200 Received: from nov-007-i575.relay.mailchannels.net (nov-007-i575.relay.mailchannels.net [46.232.183.129]) by relay.uni-heidelberg.de (8.14.1/8.14.1) with ESMTP id t6DHAPSo028040 for ; Mon, 13 Jul 2015 19:10:30 +0200 X-Sender-Id: netnames|x-authuser|joseph.wright@morningstar2.co.uk Received: from smtp3.easily.co.uk (ip-10-213-14-133.us-west-2.compute.internal [10.213.14.133]) by relay.mailchannels.net (Postfix) with ESMTPA id D92711D445B for ; Mon, 13 Jul 2015 17:10:22 +0000 (UTC) X-Sender-Id: netnames|x-authuser|joseph.wright@morningstar2.co.uk Received: from smtp3.easily.co.uk (smtp3.easily.co.uk [10.21.145.197]) (using TLSv1 with cipher DHE-RSA-AES256-SHA) by 0.0.0.0:2500 (trex/5.5.1); Mon, 13 Jul 2015 17:10:23 +0000 X-MC-Relay: Neutral X-MailChannels-SenderId: netnames|x-authuser|joseph.wright@morningstar2.co.uk X-MailChannels-Auth-Id: netnames X-MC-Loop-Signature: 1436807423391:549132535 X-MC-Ingress-Time: 1436807423390 Received: from [139.222.114.154] (port=53386 helo=[139.222.114.154]) by smtp3.easily.co.uk with esmtpa (Exim 4.43) id 1ZEhFE-000398-J8 for LATEX-L@LISTSERV.UNI-HEIDELBERG.DE; Mon, 13 Jul 2015 18:10:20 +0100 References: <559AE0A8.3080907@morningstar2.co.uk> <55A3E1DD.8050707@posteo.net> <55A3E5E6.2060405@morningstar2.co.uk> User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:38.0) Gecko/20100101 Thunderbird/38.0.1 MIME-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit X-AuthUser: joseph.wright@morningstar2.co.uk Message-ID: <55A3F0FC.2060407@morningstar2.co.uk> Date: Mon, 13 Jul 2015 18:10:20 +0100 Reply-To: Mailing list for the LaTeX3 project Sender: Mailing list for the LaTeX3 project From: Joseph Wright Subject: Re: LuaTeX support in the LaTeX kernel To: LATEX-L@LISTSERV.UNI-HEIDELBERG.DE In-Reply-To: <55A3E5E6.2060405@morningstar2.co.uk> Precedence: list List-Help: , List-Unsubscribe: List-Subscribe: List-Owner: List-Archive: Envelope-To: X-GMX-Antispam: 0 (Mail was not recognized as spam); Detail=V3; X-GMX-Antivirus: 0 (no virus found) X-UI-Filterresults: notjunk:1;V01:K0:XVO0ZBw6k7c=:zPkZbZe5dRjXdUABlQ4F+dxPk7 7H3M0yWaLQs2BZ5skM7vRKiUmObOOIlEzZQmi8MNYHKduXOY+bl4cP9qTjFRhznnGuTRXhY1Z ZhLZGJHww//2Kv34IduUtjxfEcsO9sfg4qCVY5iqQdvYdspbSZpsdNp+x6dLij8DmY+m4TonS M0wiTfRu6RFgxy/ubcpKkz1/XL1y8nfE5NfnhTBRH5CSzKD99gqdVmLpSuwvLpO80n46+6V6l UvaRDtKWS/iBZTsfTo0sWpeCHHBAMOUW6I8TrRn+RHEN1bQ085o5no/SxiWdBh+PtSstX2WH3 QX61IJ/lGWIuTIFZI50HbcXLHy4W81msB5nBh/WL9WRQZZQrXxRo2773+tgYUNGnmUYPDIYay aEUMIRgreiudn27zhJTOubNnKxwgwlK3W4xouL9m7rHvdVoTZXK2KEQZH6hAXjLmfYasWECri SW5JS8urX+dZldweY9XpBJ6Jjn6d3oHpCykYy3WhlJqCmL2/KWAmwfHLAMQjQDQU52Ntzpk4O z75j78g5EfUkNQckKa5JXcHfPhWSn/IxIXbqn73tdd+hF7V+mb4VPAbBwtW88ClfocVwACl+w 73RHJs5/snAoaWJxFs3/EbZ0fKuDtXYK8plw9eWI4/GoFzJcTvfcJ/XiaFIIiI5twm40UZh6/ PCaSxWpa9ObXlSyyVmf18uckmCCvl/TmdsxpM0fnX1peuP+hpMkhhktpKmuqvvN8gieziG7ex CzwJCqjkvmvH4pcRRO+jIc4HOJanCbZxXDvJcPVHU+kWw9ixW4oPDcpHWDI0V2Xzq4iyGvMUf PF7KezbfXfQQy4a5/9UXwHr+HvS3YDruE+zqIPZYNNFSmwakmFFwadFoqVMmbD395lvxoot4A Ymm41wgbCIOtOrKSH0vcWZFjiyoC5q/t+5U8wXaf+j5SUZmz05JHE27EBao6+UP0nBGRtg966 ZS80kC1pmpVOv4oXepCcbk4s6i8hzjJ+vZKsZpVEKZrnshq3zJxcdfn+/iAxDAPy3dmZ/dpL/ dkAafxSkNngKF3LHCdhhMkPWPYcd+f1ao/co7O5YULy+KRTHFnkkXqzC53juKOzrtK987EtYn E87U0BY3CMeXJ8wvEaKapKp19ET+rCF/XWe9GWE3tJB9UHwHLtLu1B7YlRzs5o7Ek1MBbBVuM Av+Vq2285imd4YiJv/X7/LxN03Y2gfHkxOO9pz1cgcDUsT9TT44k4fnHYyFxnPC6IDaxNCnZy /2GPlA0bN7mQxRoi1reTQuDqzeGfmy1lI8BfR9a1ORtPo/zzc/vHjK8ni8hNLpdQA0InPMZom 9z3W/osCc2ds+GIUPytD5XwkwUNQolPmu5IgVvW9T0kzbsi4sZFzaMCFsXVINSBvlhoyDh/ub N3ewjKhHXX7GMAQIWVAF8avUd6UqJxb9F3jCzOiP4H6xoSY/2EGg+8xb6ui36I/IogTM6Y5oi LUCJS11+jc/TVK9bILrdOlZh3tvfW5DTg8PBr7yqppXHSj8Vdxzok5LBtjrP4i4sSEicYrFw= = X-UI-Loop:V01:lZL2g3h68vg=:GAL+HRkvpXgcUSaUxWx6O+79X31wkB7rz+eA7H5vtRE= X-UI-Out-Filterresults: notjunk:1;V01:K0:i2ke6TeT3AQ=:VNCDKVjWLk5a9fwL+7b13/ YR4WHcXt0uPw1ryo2lDVNb1emOEo4GiUQ7XOAKpqAAdMc7L+GALSGN6I07y8N3ofBA8YqQ3Oe b26Dgcijjf0K4umS6dZbnGiRSxZIK/eKwL/LWxIV4WdYpBcXg9Ae/anUgB6YMugUS2LbKX6U3 aR6uoTomj/uOZB4Z3hivrrI+fP+5xO3BFhbuL2vK6HxsKad70heswm8CMZiFK86QEw+he6UkQ r1MPGJvVERw1gbvVFu1Ergq1VpdOl2Fv1F2HjUiDhZCjT4Zg8kRvzddkZa5DlPfhfeykpEwXR kMEZ5RndRom8eS6iRbG/6tat2bljKDsjEiKclDHTO1diNKAywVVWemYiz0Tf8cpEHSGOIo/ZB hZ2AWyyHeTzdhwDFGcv41zYhoOVHDYZmwN4LFeLuQaGFxqfkjvocvF7WNqsm1So0E9obuQdqI abJasXnH4fpi5p6vvwm0JbyW3MgEtqPpgXy0cW6JJUTIu2po3K5x X-Scanned-By: MIMEDefang 2.71 on 85.214.41.38 Status: R X-Status: X-Keywords: X-UID: 7762 On 13/07/2015 17:23, Joseph Wright wrote: > On 13/07/2015 17:05, Stephan Hennig wrote: >> Am 06.07.2015 um 22:10 schrieb Joseph Wright: >> >>> We are looking for feedback in two areas. Firstly, does the >>> functionality provided cover the needs of package authors using LuaTeX? >>> Secondly, we would like input on how a transition to new support code >>> can be managed. We are reluctant to suggest a complex compatibility >>> layer for either our code or the existing packages, and suspect at this >>> stage that a clean 'step change' may be needed for packages working with >>> LuaTeX. However, this is a complex area and needs careful consideration. >> >> Let me question the idea of implementing resource allocation stuff in >> the LaTeX kernel. What is the appropriate way to claim resources, e.g., >> attributes, in a format agnostic pure Lua low-level module foo.lua >> (which is then used in format specific wrappers/packages)? That foo.lua >> module can't certainly make use of function latex.new_attribute, since >> that is only available to LaTeX users. Or do I miss something? >> >> Best regards, >> Stephan Hennig > > We are not unmindful of the problem of code that works with multiple > formats! Most obviously, luaoftload is required for using the > higher-level fontspec interface in LaTeX but also works with plain > LuaTeX to extend the font loader. Whilst we want to work to improve the > LaTeX support in this area we certainly don't want to break working with > plain. > > The luatebase package is loaded with plain TeX and is the common way to > allocate resources. There are a variety of possible approaches where we > either make it 'aware' of kernel changes or provide a new format-neutral > package which incorporates the proposed kernel code when used with > plain. There are also other possibilities and we are considering them. > > Notably, the base registers are all allocated by the routines in the > formats (plain naming inherited by LaTeX and ConTeXt). The e-TeX format > continued that tradition with \newmarks, which we've recently added to > the LaTeX kernel, whilst \newXeTeXintercharclass has been in both the > plain and LaTeX format files for XeTeX for some time. > > The key point we are looking at is that allocators really have to be > loaded very early, ideally as part of a format. We think that LaTeX > should address this problem but are not going to rush anything. First we > are learning what support is needed (see ongoing checkins to > http://github.com/josephwright/ltluatex) and then building on that to > examine how such support changes can work with existing code. > > Not a proposed solution, but at least worth remembering, in Lua one > normally allocates local names for functions. Thus for testing at least > one can do > > new_attribute = latex.new_attribute or luatexbase.new_attribute > > or similar. We are doing that ourselves to explore what we have or have > not broken or must consider: see for example > https://github.com/josephwright/ltluatex/blob/master/doctest.tex where > David has done a minimal set of 'fix ups'. > -- > Joseph Wright To be clear, at this stage nothing is decided, up to and including whether any of this is ever added to the kernel (though I hope that something will come of it). The code is quite deliberately not in the team SVN but is separate so we can make sure that any changes are workable before altering the core codebase. -- Joseph Wright