Re: [weirds] Required values in JSON responses

[Andy Newton <[e-mail]>]

On 10/24/14, 11:22 AM, "Audric Schiltknecht"
<[e-mail]> wrote:

>Le 2014-10-24 11:11, Maarten Bosteels a écrit :
>> 
>> 
>> On 24/10/14 16:02, "Andy Newton" <[e-mail]> wrote:
>> 
>>> On 10/24/14, 9:43 AM, "Audric Schiltknecht"
>>> <[e-mail]> wrote:
>>>
>>>> Hello,
>>>>
>>>> I have not followed all previous discussions about this, so I may be
>>>> missing some information regarding this issue.
>>>>
>>>> In current JSON response draft (version 10):
>>>>
>>>> * Section 2.1:
>>>>   Clients processing JSON responses need to be prepared for values
>>>>   specified in this document to be absent from a response, as no JSON
>>>>   value listed is required to appear in a response.  In other words,
>>>>   servers are free to not included values based on their own policies.
>>>>
>>>> I understand this statement as the fact that there is no REQUIRED
>>>>values
>>>> defined in all the document (and that in fact, even an empty JSON
>>>> response is a valid response).
>>>>
>>>> (Also, there is a small nit: "servers are free to not included
>>>>values" ->
>>>> "servers are free to not include values").
>>>>
>>>>
>>>> However,
>>>> * Section 4.9
>>>>   An objectClassName is REQUIRED in all RDAP
>>>>   response objects so that the type of the object can be interpreted.
>>>>
>>>> This sections states that the 'objectClassName' JSON member is
>>>>REQUIRED.
>>>>
>>>> Furthermore,
>>>> * Section 4.2:
>>>>
>>>>   The JSON name/values of "rel", "href", "hreflang", "title", "media",
>>>>   and "type" correspond to values found in Section 5 of [RFC5988].
>>>>The
>>>>   "value" JSON value is the context URI as described by [RFC5988].
>>>>The
>>>>   "href" JSON value MUST be specified.  All other JSON values are
>>>>   OPTIONAL.
>>>>
>>>> Here, the 'href' value is a MUST for 'links' array.
>>>>
>>>>
>>>> Long story short, sections 4.2 and 4.9 both specify required values
>>>>for
>>>> JSON objects whereas Section 2.1 states that there is no required
>>>>value.
>>>>
>>>> Did I miss something about that?
>>>
>>> No, I think you found a wording glitch. I propose the following change:
>>>
>>> OLD:
>>>
>>>   Clients processing JSON responses need to be prepared for values
>>>   specified in this document to be absent from a response, as no JSON
>>>   value listed is required to appear in a response.  In other words,
>>>   servers are free to not included values based on their own policies.
>>>
>>> NEW:
>>>
>>>
>>>   Clients processing JSON responses need to be prepared for values
>>>   representing registration data specified in this document to be
>>>   absent from a response.  In other words, servers are free to not
>>>   included JSON values containing registration data based on their
>>>   own policies.
>> 
>> NEWER: 
>> 
>> 
>> Clients processing JSON responses need to be prepared for values
>>    representing registration data specified in this document to be
>>    absent from a response.  In other words, servers are free to not
>>    include JSON values containing registration data based on their
>>    own policies.
>> 
>> Replaced "free to not included" by "free to not include"
>
>That looks better, indeed.
>
>
>Related to the 'objectClassName' requirement, Section 5 may also need to
>be changed:
>
>   In addition to the "links" array with a self link, servers SHOULD
>   provide an "objectClassName" (see Section 4.9) string in each object.
>
>I think it is possible to completely remove this sentence, since it is
>already expressed in Section 4.9 (as a REQUIRED value).

I agree.

-andy