From mboxrd@z Thu Jan 1 00:00:00 1970 X-Msuck: nntp://news.gmane.io/gmane.comp.tex.context/5129 Path: main.gmane.org!not-for-mail From: Giuseppe Bilotta Newsgroups: gmane.comp.tex.context Subject: Re[2]: the "xdesc" module Date: Thu, 19 Jul 2001 12:10:38 +0200 Sender: owner-ntg-context@let.uu.nl Message-ID: <4712675332.20010719121038@bigfoot.com> References: <5.1.0.14.1.20010718124207.03e7b9e8@server-1> Reply-To: Giuseppe Bilotta NNTP-Posting-Host: coloc-standby.netfonds.no Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: main.gmane.org 1035395740 31485 80.91.224.250 (23 Oct 2002 17:55:40 GMT) X-Complaints-To: usenet@main.gmane.org NNTP-Posting-Date: Wed, 23 Oct 2002 17:55:40 +0000 (UTC) Cc: ntg-context@ntg.nl, ntg-context@ntg.nl Original-To: Hans Hagen Xref: main.gmane.org gmane.comp.tex.context:5129 X-Report-Spam: http://spam.gmane.org/gmane.comp.tex.context:5129 Wednesday, July 18, 2001 Hans Hagen wrote: >>Since the "objects" code could be used as a base for lots of >>things, among which epigraphs and bibliographies, in a >>"consistent" and highly customizable way (see the template >>feature), I think that core support for it can be taken in >>consideration :-) HH> sure, but then more xml based; this is something i'm working on I thought that XML was just a "wrapper" around TeX code? That is, TeX code is defined first, and then a XML wrapper to turn XML code to TeX code? >>Now, what am I exactly thinking about? The .tui file would contain >>entries of this kind: >> >>o d {objname} {keyname} {keyname} ... {keyname} >> >>[o d = object define] >> >>which would inform TeXUtil of the order of the keys of the object >>{objname}. Then we would have entries in the form >> >>o e {objname} {tag} {keyval} {keyval} ... {keyval} {pagenumber} >>{realpage} HH> since we want predicatibility: HH> {keyval,keyval,keyval} Ok. HH> and not pagenumber/realpagenumber, but the full location stamp [including HH> chapter and so, so that we can select! Ok. HH> also, this may be combined with indexes -) I thought about using indexes, initially, but I could not find a way to sort on different, possibly multiple, keys. >>produced by the \get[objref] commands, (I still have to >>think of a "nice" way to allow sanitized keyvals) and finally >>entries in the form >> >>o s {objname} {keyname} {keyname} {keyname} ... {keyname} >> >>[o s = object sort] HH> there is already sort key support Uh? I missed that! How do you do it? >>Can we discuss the implementation of this feature? HH> sure, but keep in mind that i have to work to finish, machines to install, HH> etc -) Oops :-) I guess I'm the only one here with nothing to do in my hands ... (well, unless I gather enough will to start to study for the doctorate ...) ;-) -- Giuseppe "Oblomov" Bilotta