RE: [dita] 2 requested fixes in DTDs

["Gershon Joseph (gerjosep)" <gerjosep@cisco.com>]

Hi Robert,

I'm OK fixing both. If I understand correctly, these fixes don't require
any rework to the arch or lang spec, correct? 


--
Gershon

-----Original Message-----
From: Robert D Anderson [mailto:robander@us.ibm.com] 
Sent: Tuesday, January 12, 2010 11:31 PM
To: dita
Subject: [dita] 2 requested fixes in DTDs


Erik Hennum sent me notifications about the following two items that
should be changed in the classification domain. These came up while Eric
Sirois was doing additional Schema testing.

1. Eric noticed that the current definition of subjectCell violates the
Unique Particle Attribution laws of XML Schema. Specifically, it allows
topicsubject twice, initially as a direct child, and again because it is
a domain substitution element for topicref. The fix for this is to
remove topicsubject as a direct child. This fix does not impact any
existing documents, and the same content model is possible, it just
makes it legal.

Testing of this led to the second discovery:

2. The <topicapply> element is a specialization of topicref. It is
currently only available inside <subjectCell>, but was intended to be a
domain substitution for topicref itself. The example in the lang spec
actually shows it this way. Erik commented that the element is mostly
useless unless defined as originally intended, but if the TC is
unwilling to make the change at this point, we will have to fix it in
1.3. The change will not invalidate any existing Classification maps; it
will provide the topicapply element as an alternative to the
topicsubject element everywhere that the topicsubject element can appear
(as originally intended).

The first item is a bug fix, and must be updated. The second item is
less clear. Are there any objections to making this change?

Thanks -

Robert D Anderson
IBM Authoring Tools Development
Chief Architect, DITA Open Toolkit


---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that
generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php