cxf-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Benson Margulies <bimargul...@apache.org>
Subject Re: cxf - aegis service customization
Date Mon, 18 May 2009 22:03:50 GMT
OK, now I am tuned into the full muddle.

Number is an abstract class. So, how do you expect Aegis to pick a concrete
type?

On Mon, May 18, 2009 at 6:02 PM, Benson Margulies <bimargulies@apache.org>wrote:

> Aha, the service explodes. I'm working on it.
>
>
> On Mon, May 18, 2009 at 5:30 PM, Benson Margulies <bimargulies@apache.org>wrote:
>
>> It turns out to seem to me to work fine if you just specify the typeName.
>> I'm checking in a unit test.
>>
>> On Mon, May 18, 2009 at 12:06 PM, Attila Balogh <attila.balogh@gmail.com>wrote:
>>
>>> dear Benson,
>>>
>>> at this URL:
>>> https://issues.apache.org/jira/browse/CXF-1729?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
>>> you wrote that in CXF 2.2 service customization with Aegis is possible
>>> again, however i've been struggling with it for a half day now, and
>>> can't get it working.
>>>
>>> i am using CXF 2.2.1, i have a simple interface and an according
>>> implementing class with a method that is expecting two
>>> java.lang.Number parameters, which i want to receive as xsd:long
>>> types, i created a ..Impl.aegis.xml:
>>>
>>> <mappings xmlns:xsd="http://www.w3.org/2001/XMLSchema">
>>>    <mapping>
>>>        <method name="findContents">
>>>            <parameter index="1" componentType="java.lang.Number"
>>>                typeName="xsd:long" />
>>>            <parameter index="2" componentType="java.lang.Number"
>>>                typeName="xsd:long" />
>>>        </method>
>>>    </mapping>
>>> </mappings>
>>>
>>> But unfortunately it doesn't seem to work. (The same thing (with
>>> type="java.lang.Number") works for a bean that has also a Number
>>> property, so in the worst case i'll create a parameter bean for all
>>> service methods, but that wouldn't be the nicest solution)
>>>
>>> Am i doing something completely wrong and this mapping shouldn't be
>>> there/ should be in some different format?
>>>
>>> If you are not the person to ask this from then please tell me where
>>> to turn and apologies for disturbing you.
>>> Thanks a lot in advance,
>>>
>>> Attila Balogh
>>>
>>
>>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message