Re: [ebxmlms-develop] Using Sun's JAXM implementation: bugs?

[Ng C. Y. [Cyng] <cy...@cs...>]

Hello,

> <problem item="1">
> In the com.sun.xml.messaging.jaxm.ebxml.EbXMLMessageImpl class, the
> namespace
>
> 	NS_URI = "http://www.ebxml.org/namespaces/messageHeader";
>
> is being used, and therefore none of the ebXML tags are found, because the
> namespaces don't match. If I change this to
>
> 	NS_URI =
> "http://www.oasis-open.org/committees/ebxml-msg/schema/msg-header-2_0.xsd";
>
> in Sun's EbXMLMessageImpl.java, the ebXML tags are found and parsed. Why
> would Sun's code be using the www.ebxml.org namespace? Is it from an older
> version of the specification?
> </problem>

        EbXMLMessageImpl provided in JAXM is quite preliminary and has not
been updated for sometimes. I don't know if this class is released as
a prototype or demonstation only. Anyway, the eb: namespace is still
ebMS 1.0 obviously.


> <problem item="2">
> Also in the com.sun.xml.messaging.jaxm.ebxml.EbXMLMessageImpl class, the
> private timeStamp field is declared as long, and the timestamp is parsed as
>
> 	this.timeStamp = Long.parseLong(elem.getValue());
>
> Since the timestamp in the msh-header-2_0.xsd schema is an ISO8601 string
> (<element name="Timestamp" type="dateTime"/>), rather than a long, this
> throws an exception. This code must be changed to parse an ISO8601 string.
> </problem>

        Yes, we do our own parsing and validity check in Hermes in order
to be compliant with the XML Schema and ebMS Specification. long is
obviously not the choice.


> Should these problems be reported to Sun as bugs?

        Yes, you may consider to do so.

Regards,
CY

----------------------------------------------------------------------------
Ng Chi Yuen, CY.       cy...@ce...       http://www.cecid.hku.hk/
Technology Officer,
Centre for E-Commerce Infrastructure Development,
The University of Hong Kong
----------------------------------------------------------------------------