W3C home > Mailing lists > Public > www-xml-schema-comments@w3.org > October to December 2000

Re: LC-215 - Easy Add-ins (i18n Comment on XML Schema Last Call Working Draft)

From: Martin J. Duerst <duerst@w3.org>
Date: Mon, 16 Oct 2000 16:51:40 +0900
Message-Id: <4.2.0.58.J.20001016162138.03ae2b00@sh.w3.mag.keio.ac.jp>
To: "Martin Gudgin" <marting@develop.com>, "Martin Gudgin" <marting@develop.com>
Cc: "Schema Comments" <www-xml-schema-comments@w3.org>
Hello Martin,

Many thanks for these examples. I have had a look at them.
As far as I understand, they show very well what they want
to show:

- It is possible to introduce additional elements using the
   substitution group construct.
- It is possible to introduce such elements into the same namespace
   ([3]/[4]) or from a new namespace ([5]/[6]).

I think that it should also be possible to create a new namespace
with everything (both the predefined elements and the new elements)
in it.

What I think these examples don't cover is the case that both the
original namespace and the additions I want to make already exist.
What are the possibilities and limitations in this case?


What I'm really quite worried about is the use of type='string' for all
these. This is a different issue (LC-216 merge mixed, textOnly, and string?)
so I won't go into details here, but it shows exactly why we have
raised this issue; if it were not that easy to just use type='string',
the chance that a schema writer would seal off <b>, <i>,... forever
without any real intention or reason would be much smaller.


Regards,   Martin.



At 00/10/08 21:06 +0100, Martin Gudgin wrote:
>Martin,
>
>Here is an example that uses substitution groups.
>
>The original schema[1] and instance[2] allow the 'p' element to contain 'b'
>and 'i' elements interleaved with test ( 'p' has a content model of mixed,
>'b' and 'i' have a content model of textOnly and are based on string ).
>
>A second schema[3] and instance[4] show how to allow new elements in a new
>namespace where 'b' and 'i' are legal.
>
>A third schema[5] and instance[6] show got to allow new elements in the same
>namespace where 'b' and 'i' are legal. This example is very similar to the
>one shown in the primer[7]
>
>I hope this illustrates how new element names ( and types ) can be
>substituted for the original element names ( and types ).
>
>Regards
>
>Martin Gudgin
>XML Schema Working Group
>
>
>[1] http://marting.develop.com/xsd/lc-215/subgroup.xsd
>[2] http://marting.develop.com/xsd/lc-215/subgroupex1.xml
>[3] http://marting.develop.com/xsd/lc-215/subgroup2.xsd
>[4] http://marting.develop.com/xsd/lc-215/subgroupex2.xml
>[5] http://marting.develop.com/xsd/lc-215/subgroup3.xsd
>[6] http://marting.develop.com/xsd/lc-215/subgroupex1ext.xml
>[7] http://www.w3.org/TR/xmlschema-0/#SubsGroups
Received on Monday, 16 October 2000 04:12:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 6 December 2009 18:12:48 GMT