Hi Eric--
I myself don't know whether it should be considered a bug or a feature
;-)
In any event, my testing with Arbortext Epic 5.1 - which I think uses
Xerxes-C - discloses the following:
- mixed="true" on xs:complexType element, but
not on the xs:complexContent element that extends it to include the
class attribute, as done in your refactored xsds, results in the loss
of mixed content
- mixed="true" on the xs:complexType *and* on
the xs:complexContent elements restores mixed content
- mixed="true" on the xs:complexContent element
*alone* also restores mixed content
And this concurs with the way I read the schema
standard: xs:complexContent has a default mixed attribute value of
"false." Nowhere do I find it documented that it should inherit the
value from the xs:complexType element it modifies to override this
default - nor does that seem a typical construct in xml standards.
Usually, you'd at least need to set a value of "inherit" - or that
would need to be the default.
In any event, putting setting mixed="true" on xs:complexContent
elements where appropriate should fix the problem.
--Dana
Eric Sirois wrote:
Hi Dana,
It looks like this one is a bug in the Xerces-C parser. Xerces-J as well
as a few other parsers are returning the documents as valid.
One of two cases is happening. Either the anonymous type (complexType in
the element declaration) did not get the mixed="true" or there is
something wrong with the schemas. If it was a problem with the schema, ten
it should have returned the proper error.
That being said...I'll add the mixed="true" to the appropriate content
models as part of the design pattern. It may take a few days to make the
change and verify the changes.
Kind regards,
Eric
Eric A. Sirois
Staff Software Developer
DB2 Universal Database - Information Development
DITA Migration and Tools Development
IBM Canada Ltd. - Toronto Software Lab
Email: esirois@ca.ibm.com
Blue Pages (Internal)
"Transparency and accessibility requirements dictate that public
information and government
transactions avoid depending on technologies that imply or impose a
specific product or
platform on businesses or citizens" - EU on XML-based office document
formats.
Dana Spradley
dana.spradley@oracle.com"><dana.spradley@or
acle.com> To
dita@lists.oasis-open.org
02/13/2007 11:30 cc
AM
Subject
Re: [dita] Groups - DITA 1.1 XSDs
(dita1.1_refact_b.zip) uploaded
There appears to be a major bug in this refactoring, Eric.
It seems that the mixed=true attribute must be redeclared on the
complexContent elements used for the class attribute extension.
Otherwise mixed content is eliminated from elements that should have it.
At least, that's what happened in the commercial editor I used for testing.
--Dana
esirois@ca.ibm.com wrote:
Found a couple of small bugs...with the class attribute values in
taskMod.xsd (chrow and stepxmp).
It doesn't affect validation of files, only the processing.
Kind regards,
Eric
-- Mr Eric Sirois
The document revision named DITA 1.1 XSDs (dita1.1_refact_b.zip) has
been
submitted by Mr Eric Sirois to the OASIS Darwin Information Typing
Architecture (DITA) TC document repository. This document is
revision #10
of dita_xsd_1.1.zip.
Document Description:
The implementation of the DITA 1.1 XML Schemas.
View Document Details:
http://www.oasis-open.org/apps/org/workgroup/dita/document.php?document_id=22329
Download Document:
http://www.oasis-open.org/apps/org/workgroup/dita/download.php/22329/dita1.1_refact_b.zip
Revision:
This document is revision #10 of dita_xsd_1.1.zip. The document
details
page referenced above will show the complete revision history.
PLEASE NOTE: If the above links do not work for you, your email
application
may be breaking the link into two pieces. You may be able to copy
and paste
the entire link address into the address field of your web browser.
-OASIS Open Administration