[prev in list] [next in list] [prev in thread] [next in thread] 

List:       esb-java-dev
Subject:    Re: [Carbon-dev] WSDL types with no targetNamespace
From:       Anjana Fernando <anjana () wso2 ! com>
Date:       2010-03-28 13:57:08
Message-ID: 2e8e13db1003280645t559dcbc3rf22ee3cda32ef1d0 () mail ! gmail ! com
[Download RAW message or body]

Hi,

Yeah, what I was referring to was the targetNamespace of the types
section, so I guess, the current behavior in data-services where a
targetNamespace is always applied is correct, and no need to change
the implementation.

Thanks,
Anjana.

On Sun, Mar 28, 2010 at 1:39 PM, Amila Suriarachchi <amila@wso2.com> wrote:
>
>
> On Sun, Mar 28, 2010 at 10:29 AM, Senaka Fernando <senaka@wso2.com> wrote:
>>
>>
>> On Sun, Mar 28, 2010 at 9:35 AM, Anjana Fernando <anjana@wso2.com> wrote:
>>>
>>> Hi,
>>>
>>> In data-services, we have a small issue, in finding out if we can
>>> create a service with no targetNamespace for the result. That is, if
>>> we can, not provide a targetNamespace in the types section in the
>>> WSDL.
>>>
>>> First, I simply tried it by not giving a targetNamespace to the
>>> XMLSchema and, the types section were generated properly, but in the
>>> "message" sections in the WSDL, the part element which reference a
>>> schema type, it always outputs with the format "prefix:localName", so
>>> since the targetNamespace is not available it prints as
>>> "null:localName", thus it's invalid. I patched Axis2 and tried it
>>> again, now it just prints as "localName".
>>>
>>> So now if I deploy the data-service in question and code generate a
>>> client using its WSDL (which has no targetNamespace in its types), the
>>> client works fine, and you could get the service response properly.
>>> But with the try-it tool it does not work properly, as in, it doesn't
>>> return the result as it should. And by giving the WSDL to the
>>> WSDLValidator, it states that the WSDL is invalid, giving a reason
>>> related to having no targetNamespace. I've attached the WSDL in
>>> question here, does anyone have an idea whether, this is actually a
>>> valid WSDL, and if not giving a targetNamespace in the types is
>>> possible? ..
>>
>> According to [1], a valid WSDL can exist without a target namespace, but
>> must not have any fragment identifiers. However, according to [2], which is
>> the WS-I Basic Profile 1.1, a WSDL 1.1 is treated in the same manner as an
>> XML Schema. It also states that an XML Schema must always have a target
>> namespace. Therefore, a WSDL without a target namespace is not WS-I valid
>> (not interoperable).
>
> Thanks for pointing out the basic profile.
>
> Here the problem is not with the wsdl targetnamespace. But with the
> targetnamespace of the Xml Schema elements within it. This part[1] of the
> basic profile states that is not allowed.
>
> thanks,
> Amila.
>
> [1]
> http://www.ws-i.org/Profiles/BasicProfile-1.1-2004-08-24.html#Schema_targetNamespace_Structure
>
>>
>> [1] http://www.w3.org/TR/wsdl11elementidentifiers/#frag-ids
>> [2] http://www.ws-i.org/Profiles/BasicProfile-1.1-2004-08-24.html
>>
>> Thanks,
>> Senaka.
>>>
>>> Cheers,
>>> Anjana.
>>>
>>> _______________________________________________
>>> Carbon-dev mailing list
>>> Carbon-dev@wso2.org
>>> https://mail.wso2.org/cgi-bin/mailman/listinfo/carbon-dev
>>>
>>
>>
>>
>> --
>> Senaka Fernando
>> Software Engineer
>> WSO2 Inc.
>> E-mail: senaka AT wso2.com;  Mobile: +94 77 322 1818
>>
>> http://www.wso2.com/ - "Lean . Enterprise . Middleware"
>>
>> _______________________________________________
>> Carbon-dev mailing list
>> Carbon-dev@wso2.org
>> https://mail.wso2.org/cgi-bin/mailman/listinfo/carbon-dev
>>
>
>
> _______________________________________________
> Carbon-dev mailing list
> Carbon-dev@wso2.org
> https://mail.wso2.org/cgi-bin/mailman/listinfo/carbon-dev
>
>

_______________________________________________
Carbon-dev mailing list
Carbon-dev@wso2.org
https://mail.wso2.org/cgi-bin/mailman/listinfo/carbon-dev

[prev in list] [next in list] [prev in thread] [next in thread]