I think I agree with you. I'm in no position to test with the IBM JDK. Could
you tack a patch onto a JIRA and I'll apply it.
On Mon, Sep 29, 2008 at 7:47 AM, Benson Margulies <[EMAIL PROTECTED]>wrote:
> It may take me a day or two to get to this, but I will.
>
>
> On Sun, Sep 28, 2008 at 10:52 PM, Peter Jones <[EMAIL PROTECTED]>wrote:
>
>>
>> Hi there,
>>
>> Think there's a small discrepancy in the rt/databinding/aegis
>> JaxbTypeTest.
>> Note the type which is checked for when the element QName is
>> "elementProperty" in the two checks below:
>>
>> AnnotatedTypeInfo info = new AnnotatedTypeInfo(tm, JaxbBean1.class,
>> "urn:foo", new TypeCreationOptions());
>> ...
>> QName element = (QName)elements.next();
>> ...
>> Type custom = info.getType(element);
>> if ("bogusProperty".equals(element.getLocalPart())) {
>> assertTrue(custom instanceof StringType);
>> } else if ("elementProperty".equals(element.getLocalPart())) {
>> assertTrue(custom instanceof CustomStringType);
>> } else {
>> fail("Unexpected element name: " + element.getLocalPart());
>> }
>> element = (QName)elements.next();
>> ...
>> custom = info.getType(element);
>> if ("bogusProperty".equals(element.getLocalPart())) {
>> assertTrue(custom instanceof StringType);
>> } else if ("elementProperty".equals(element.getLocalPart())) {
>> assertTrue(custom instanceof StringType);
>> } else {
>> fail("Unexpected element name: " + element.getLocalPart());
>> }
>>
>> If the "elementProperty" is the first QName, we check that the type is an
>> instance of CustomStringType, but if its the second Qname we check that
>> its an instance of StringType.
>>
>> This is the annotation in the JaxbBean1 class:
>>
>> @XmlElement(type = CustomStringType.class)
>> public String getElementProperty() {
>>
>> Would that suggest the type should be a CustomStringType? In the test,
>> it's
>> always a StringType.
>>
>> The test passes with the sun jdk as "elementProperty" is always the second
>> QName returned, but if you run this test with the ibm jdk, it fails since
>> "elementProperty" is the first QName returned.
>>
>> If the test is wrong and it should check for StringType both times, fair
>> enough, otherwise, I can file a jira. Let me know if you have any
>> thoughts.
>>
>> Cheers,
>> Peter
>>
>> --
>> Peter Jones
>> Progress Software
>> E-Mail: mailto:[EMAIL PROTECTED]
>> Tel: 709-738-3725 x321 | Fax: 709-738-3745
>> 570 Newfoundland Drive, St. John's, NL, Canada A1A 5B1
>>
>
>
