> Hi Deepal:
>
> Deepal Jayasinghe wrote:
>>>>> The important thing seems to me to decide exactly what you get if you
>>>>> do a registry.get("/foo;comments"). I would think you'd get a
>>>>> collection Resource which contains the comments as individual Atom
>>>>> entries.
>>>> In this case the return type is Resource and that resource contains
>>>> links to the comments
>>>>> On the wire, this is EXACTLY the same as when you do
>>>>> registry.getComments("/foo").
>>>> In this case the return type is set of comment objects.
>>> What I'm saying is that there should be absolutely no difference at
>>> the protocol level. Both of these calls should generate
>> There is a difference , the difference is in one case we call
>> registry.get whereas in other we call registry.getComments and the
>> results are two different. And I can implement this if that is only for
>> comments , but the same thing is there for tags, versions as well.
>
> Hm, that doesn't sound right to me. See below.
>
>> The problem is this , when we call "get" method what we get is Resource
>> with its child list having paths , for example if we call "get" method
>> with "/c1/c2;comments" then the resultant Resource will have child list
>> that does not have information about the comments. Therefore as I can
>> see calling "/c1/c2;comments" using "get" API wrong to me , I think that
>> should be changed to "/c1/c2;commentPaths" since what actually we get is
>> path , not actual resource.
>
> Well, comments are just Atom entries, aren't they? As far as I can
> see there is NO reason to ever return a feed of "just paths" to
> comments - you'd return the comments themselves.
The thing the API gave me a resource having comment paths , not actual
comments. Then I have to go though each path and get the actual comment
and return them.
-Deepal
_______________________________________________
Registry-dev mailing list
[email protected]
http://wso2.org/cgi-bin/mailman/listinfo/registry-dev
