W3C home > Mailing lists > Public > www-math@w3.org > November 2005

Re: Simple way to include definitions of all XML entities

From: Jirka Kosek <jirka@kosek.cz>
Date: Wed, 09 Nov 2005 11:54:45 +0100
Message-ID: <4371D575.1050804@kosek.cz>
To: David Carlisle <davidc@nag.co.uk>
Cc: www-math@w3.org, docbook@lists.oasis-open.org
David Carlisle wrote:

> Would you want the "master" file for each group of entities to include
> the others by external enity reference, much as (say) the XSLT2 map
> files are set up with the ISO 8879 set having  iso8879map.xsl which
> xsl:includes isoamsamap.xsl and friends. or would you want the "master"
> file to have copies of the entity definitions.

Actually this doesn't matter. But as you autogenerate entity files it 
might be better to create "master" file with copies of all entity 
definitions instead of just references. Imagine someone who doesn't use 
XML catalogs -- he will benefit from getting all entity definitions in a 
single HTTP response from the W3C web-server.

> Essentially anything along those lines is possible, as you may have seen
> the entities are all derived from the unicode.xml file, and how they are
> grouped and split into different files is "just a bit of xsl".

Interesting, you even have TeX mappings. Great piece of work!

> If for example docbook wanted some custom set of entities that was
> larger than html but smaller than mathml, but consistent with both (so
> far as that is possible:-) It would be easy to derive such a set by
> adding some annotations to unicode.xml and then cranking the handle...

I did some research in this area and if the following XPath 2.0 
expression isn't broken, I think that there are no incosistencies in 
mapping entity names to Unicode characters (STIX entities are excluded).

distinct-values(//character//entity/@id[. != ''][../@set != 'STIX'][ . = 
preceding::character/entity/@id[../@set != 'STIX']])

Because it is hard to predict which entites user want to use -- DocBook 
uses ISO 8879 entities, but some people want to use MathML combined with 
DocBook which leads to entities from ISO 8879 + ISO 9573 + MathML 
extensions. It is also possible to mix XHTML (table model and forms) 
with DocBook, so someone might cut'n'paste XHTML entity into DocBook file.

Because of this I think that what is really needed here is a real union 
of all entity definitions (8879 + 9573(2003) + MathML + XHTML). If I 
didn't make error in the XPath expression above, this combined entity 
set shouldn't contain any ambiguities.

So for use with DocBook (and probably with other vocabularies) it would 
be great if there will be only one entity definition file with all 
entities defined in the one place.

I can imagine that other users might benefit also from merged entity 
definitions just for MathML, ISO 8879 or XHTML.



   Jirka Kosek     e-mail: jirka@kosek.cz     http://www.kosek.cz
   Profesionální školení a poradenství v oblasti technologií XML.
      Podívejte se na náš nově spuštěný web http://DocBook.cz
        Podrobný přehled školení http://xmlguru.cz/skoleni/
Nejbližší termíny školení: XML schémata (včetně RELAX NG) 7.-9.11.
          *** DocBook 5.-7.12. *** XSL-FO 19.-20.12. ***

Received on Wednesday, 9 November 2005 10:55:03 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:27:37 UTC