04-27-07 12:11 AM
Michael,
I've changed my mind, and now I think we should fix the XSD, having strict
checks and lose flexibility, instead of having the jdo implementation do mor
e
checking. It's just too long to implement/maintain/test, prone to errors and
costs in performance.
I propose the extension element to be placed as last child of the parent
element, and applications upgrading from JDO 2 to JDO 2.1 could use a PMF
vendor property to disable XSD validation if their schema is not valid
By the way, the DTD in the JDO 2 spec document refers to JDO 1 DTD instead o
f
JDO 2 DTD.
Regards,
Quoting Michael Bouschen <mbo.tech@spree.de>:
> Hi,
>
> as Craig recently pointed out the JDO XSD files are not correct and
> cause failures when using the feature schema-full-checking during XML
> parsing (see Craig's email with subject "XSD woes" and the JIRA issue
> http://issues.apache.org/jira/browse/JDO-472).
>
> Here is a proposal of what should be changed from my point of view:
> - Loosen the restrictions on the placement of extension elements,
> meaning that the jdo implementation needs to do more checking. I sent
> around an updated version of the JDO XSD file (subject "Re: XSD woes")
> and did not receive any objections.
> - Update the version entry in the XML Schema file to 2.1 and rename the
> xsd files accordingly: jdo_2_1.xsd, orm_2_1.xsd and jdoquery_2_1.xsd.
> - For consistency rename the DTD files to include the version number 2.1
> (e.g. jdo_2_1.dtd) and adapt the standard PUBLIC and SYSTEM DOCTYPE.
>
> Regards Michael
>
> --
> Tech@Spree Engineering GmbH Tel.: +49/(0)30/235 520-33
> Buelowstr. 66 Fax.: +49/(0)30/217 520-12
> 10783 Berlin mailto:mbo.tech@spree.de
>
> Geschaeftsfuehrung: Dr. Gerhard Mueller-Proefrock
> Anna-Kristin Proefrock
> Sitz Berlin, Amtsgericht Charlottenburg, HRB 564 52
>
>
[ Post a follow-up to this message ]
| 
