You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cxf.apache.org by gilgal <gi...@gmail.com> on 2010/08/27 16:38:29 UTC
WADL generation question
When cxf generates my WADL I get references to objects in the grammar
section, but if it is a primitive type it has a comment of what the
primitive type is but no element attribute set. Is there a way to get what
is in the comment into the element attribute of the representation tag?
Here are two examples.
References the object in the grammar:
<resource path="/optout/individual/{individualId}">
<method name="GET">
<request/>
<response>
<representation mediaType="application/xml" element="prefix1:indOptOut"/>
<representation mediaType="application/json"/>
</response>
</method>
</resource>
Has only a comment that it is xs:boolean:
<resource path="/ministering/remove/{unitNbr}/{individualId}">
<method name="DELETE">
<request/>
<response>
<!-- Primitive type : xs:boolean -->
<representation mediaType="application/xml"/>
<!-- Primitive type : xs:boolean -->
<representation mediaType="application/json"/>
</response>
</method>
</resource>
Thanks,
Gil
--
View this message in context: http://cxf.547215.n5.nabble.com/WADL-generation-question-tp2740129p2740129.html
Sent from the cxf-user mailing list archive at Nabble.com.
Re: WADL generation question
Posted by Sergey Beryozkin <sb...@gmail.com>.
Hi
Just FYI, this improvement has been implemented
thanks for your help
Sergey
[1] https://issues.apache.org/jira/browse/CXF-3242
On Thu, Sep 9, 2010 at 4:38 PM, Gilgal Jones <gi...@gmail.com> wrote:
> Sergey,
>
> Here is the response I got on users@wadl.dev.java.net from Marc Hadley:
>
> I assume you are trying to describe a representation that contains a
> Boolean
> – right ?
>
> I think a better way to represent this would as follows:
>
> <representation mediaType=”text/plain”>
> <param name=”result” type=”xsd:Boolean” style=”plain”/>
> </representation>
>
> It might be worth considering switching from text/plain to something with a
> bit more structure. E.g. you could use application/x-www-form-urlencoded:
> instead of a body of “true” you’d send back “result=true”. That little bit
> of extra structure will allow you to add additional information in the
> future with less chance of breaking (well written) clients. In that case
> you’d use style=”query” in the param element. Personally I’d use JSON or
> XML
> but I understand the urge to keep things simple.
>
> Marc.
>
> On Thu, Sep 2, 2010 at 11:21 AM, Sergey Beryozkin <sberyozkin@gmail.com
> >wrote:
>
> > Hi
> >
> > sorry for a delay
> >
> > On Fri, Aug 27, 2010 at 3:38 PM, gilgal <gi...@gmail.com> wrote:
> >
> > >
> > > When cxf generates my WADL I get references to objects in the grammar
> > > section, but if it is a primitive type it has a comment of what the
> > > primitive type is but no element attribute set. Is there a way to get
> > what
> > > is in the comment into the element attribute of the representation tag?
> > >
> > Here are two examples.
> > > References the object in the grammar:
> > > <resource path="/optout/individual/{individualId}">
> > >
> > > <method name="GET">
> > > <request/>
> > > <response>
> > > <representation mediaType="application/xml"
> element="prefix1:indOptOut"/>
> > > <representation mediaType="application/json"/>
> > > </response>
> > > </method>
> > > </resource>
> > > Has only a comment that it is xs:boolean:
> > > <resource path="/ministering/remove/{unitNbr}/{individualId}">
> > >
> > >
> > > <method name="DELETE">
> > > <request/>
> > > <response>
> > > <!-- Primitive type : xs:boolean -->
> > > <representation mediaType="application/xml"/>
> > > <!-- Primitive type : xs:boolean -->
> > > <representation mediaType="application/json"/>
> > > </response>
> > > </method>
> > > </resource>
> > >
> > > Thanks,
> > > Gil
> > >
> > >
> > This might be done in principle. I'm not sure though what it will mean.
> XML
> > schema does have datatypes such as xs:boolean, etc. But those datatypes
> are
> > meant to be used by schemas defining new custom (complex or simple) types
> > and elements, the XML ones. So if we have say a text/plain
> representation,
> > then saying that a it's of type xsd:
> > integer for example, using @element attribute, seems a bit strange. But
> > may
> > be it could be just to do with reusing existing well-known datatypes, so
> >
> > <representation mediaType="text/plain" element="xsd:boolean"/>
> >
> > is kind of ok...
> > I'll think about it.
> > In meantime, if you could ask for some clarifications on the WADL list
> then
> > it would really help. Ask if using @element="xsd:boolean" for non xml
> > representations is acceptable
> >
> > thanks, Sergey
> >
> >
> >
> > --
> > > View this message in context:
> > >
> >
> http://cxf.547215.n5.nabble.com/WADL-generation-question-tp2740129p2740129.html
> > > Sent from the cxf-user mailing list archive at Nabble.com.
> > >
> >
>
Re: WADL generation question
Posted by Gilgal Jones <gi...@gmail.com>.
Sergey,
Here is the response I got on users@wadl.dev.java.net from Marc Hadley:
I assume you are trying to describe a representation that contains a Boolean
– right ?
I think a better way to represent this would as follows:
<representation mediaType=”text/plain”>
<param name=”result” type=”xsd:Boolean” style=”plain”/>
</representation>
It might be worth considering switching from text/plain to something with a
bit more structure. E.g. you could use application/x-www-form-urlencoded:
instead of a body of “true” you’d send back “result=true”. That little bit
of extra structure will allow you to add additional information in the
future with less chance of breaking (well written) clients. In that case
you’d use style=”query” in the param element. Personally I’d use JSON or XML
but I understand the urge to keep things simple.
Marc.
On Thu, Sep 2, 2010 at 11:21 AM, Sergey Beryozkin <sb...@gmail.com>wrote:
> Hi
>
> sorry for a delay
>
> On Fri, Aug 27, 2010 at 3:38 PM, gilgal <gi...@gmail.com> wrote:
>
> >
> > When cxf generates my WADL I get references to objects in the grammar
> > section, but if it is a primitive type it has a comment of what the
> > primitive type is but no element attribute set. Is there a way to get
> what
> > is in the comment into the element attribute of the representation tag?
> >
> Here are two examples.
> > References the object in the grammar:
> > <resource path="/optout/individual/{individualId}">
> >
> > <method name="GET">
> > <request/>
> > <response>
> > <representation mediaType="application/xml" element="prefix1:indOptOut"/>
> > <representation mediaType="application/json"/>
> > </response>
> > </method>
> > </resource>
> > Has only a comment that it is xs:boolean:
> > <resource path="/ministering/remove/{unitNbr}/{individualId}">
> >
> >
> > <method name="DELETE">
> > <request/>
> > <response>
> > <!-- Primitive type : xs:boolean -->
> > <representation mediaType="application/xml"/>
> > <!-- Primitive type : xs:boolean -->
> > <representation mediaType="application/json"/>
> > </response>
> > </method>
> > </resource>
> >
> > Thanks,
> > Gil
> >
> >
> This might be done in principle. I'm not sure though what it will mean. XML
> schema does have datatypes such as xs:boolean, etc. But those datatypes are
> meant to be used by schemas defining new custom (complex or simple) types
> and elements, the XML ones. So if we have say a text/plain representation,
> then saying that a it's of type xsd:
> integer for example, using @element attribute, seems a bit strange. But
> may
> be it could be just to do with reusing existing well-known datatypes, so
>
> <representation mediaType="text/plain" element="xsd:boolean"/>
>
> is kind of ok...
> I'll think about it.
> In meantime, if you could ask for some clarifications on the WADL list then
> it would really help. Ask if using @element="xsd:boolean" for non xml
> representations is acceptable
>
> thanks, Sergey
>
>
>
> --
> > View this message in context:
> >
> http://cxf.547215.n5.nabble.com/WADL-generation-question-tp2740129p2740129.html
> > Sent from the cxf-user mailing list archive at Nabble.com.
> >
>
Re: WADL generation question
Posted by Sergey Beryozkin <sb...@gmail.com>.
Hi
sorry for a delay
On Fri, Aug 27, 2010 at 3:38 PM, gilgal <gi...@gmail.com> wrote:
>
> When cxf generates my WADL I get references to objects in the grammar
> section, but if it is a primitive type it has a comment of what the
> primitive type is but no element attribute set. Is there a way to get what
> is in the comment into the element attribute of the representation tag?
>
Here are two examples.
> References the object in the grammar:
> <resource path="/optout/individual/{individualId}">
>
> <method name="GET">
> <request/>
> <response>
> <representation mediaType="application/xml" element="prefix1:indOptOut"/>
> <representation mediaType="application/json"/>
> </response>
> </method>
> </resource>
> Has only a comment that it is xs:boolean:
> <resource path="/ministering/remove/{unitNbr}/{individualId}">
>
>
> <method name="DELETE">
> <request/>
> <response>
> <!-- Primitive type : xs:boolean -->
> <representation mediaType="application/xml"/>
> <!-- Primitive type : xs:boolean -->
> <representation mediaType="application/json"/>
> </response>
> </method>
> </resource>
>
> Thanks,
> Gil
>
>
This might be done in principle. I'm not sure though what it will mean. XML
schema does have datatypes such as xs:boolean, etc. But those datatypes are
meant to be used by schemas defining new custom (complex or simple) types
and elements, the XML ones. So if we have say a text/plain representation,
then saying that a it's of type xsd:
integer for example, using @element attribute, seems a bit strange. But may
be it could be just to do with reusing existing well-known datatypes, so
<representation mediaType="text/plain" element="xsd:boolean"/>
is kind of ok...
I'll think about it.
In meantime, if you could ask for some clarifications on the WADL list then
it would really help. Ask if using @element="xsd:boolean" for non xml
representations is acceptable
thanks, Sergey
--
> View this message in context:
> http://cxf.547215.n5.nabble.com/WADL-generation-question-tp2740129p2740129.html
> Sent from the cxf-user mailing list archive at Nabble.com.
>