Re: shapes-ISSUE-187 (severity and message): sh:severity and sh:message are not defined as shapes graph properties

On 10/10/2016 11:30, Karen Coyle wrote:
> I think that the definition of sh:severity must be in the section on 
> constraints if sh:severity is a constraint property. Then, the section 
> on validation would explain the use of sh:severity in the validation 
> report.

Ok, I have moved that content out of the results section. It is now 2.3.6:

https://github.com/w3c/data-shapes/commit/6bb4fa1ce5d08e84b5e95f9fd61288a4e611a361

Does this address your suggestion?

>
> It's true that the values of sh:message are not mandated, but that's 
> not my question. My question is whether sh:message can be included in 
> a constraints graph as a way for developers to determine the message 
> that will be returned to users? If not, then is there any 
> SHACL-defined use of sh:message? If the answer to that is "no" then 
> I'm not convinced that sh:message can be considered a SHACL property 
> if its object is entirely undefined.

Note to Arnaud: I'd appreciate having ISSUE-178 on the agenda soon. I 
believe it would help resolving this issue here, and is probably not 
controversial either.

I had already answered Karen's question, sh:message is used in 
constraint components and sh:sparql constraints.

Holger



>
> kc
>
> On 10/9/16 4:17 PM, Holger Knublauch wrote:
>> On the use of sh:message, I had raised ISSUE-178. In addition to this,
>> sh:message can be used in constraint components and sh:sparql
>> constraints (SHACL Full), both in the shapes graph. Section 3.4 also
>> explicitly states
>>
>> "The values of |sh:message| are not mandated by SHACL and are considered
>> implementation-specific."
>>
>> What additional aspects of sh:message are you referring to here?
>>
>> On sh:severity, section 3.4.9 explains how it is used in shapes graphs.
>>
>> "Constraints may specify their severity level in the shapes graph
>> <#dfn-shapes-graph> using the property |sh:severity|,..."
>>
>> I do not understand what you think is missing.
>>
>> Holger
>>
>>
>> On 9/10/2016 12:46, RDF Data Shapes Working Group Issue Tracker wrote:
>>> shapes-ISSUE-187 (severity and message): sh:severity and sh:message 
>>> are not defined as shapes graph properties
>>>
>>> http://www.w3.org/2014/data-shapes/track/issues/187
>>>
>>> Raised by: Karen Coyle
>>> On product:
>>>
>>> sh:severity and sh:message are described in the validation report, 
>>> but are not included as properties in the shapes graph even though 
>>> they are intended to be controlled by the shapes graph. It seems 
>>> that they would be coded with the specific constraints to which they 
>>> apply.
>>>
>>> There is a default for sh:severity, but the validation report 
>>> section (3.) does not give any information about the source of the 
>>> sh:message information. Presumably the only source of messages is 
>>> the constraints graph. If there is any other source of messages, it 
>>> is not stated here.
>>>
>>>
>>>
>>
>

Received on Monday, 10 October 2016 01:56:20 UTC