Closed toru-ikeda-iij closed 2 weeks ago
I don't think that constraint works the way you are proposing, it looks like you want to have a way to 'template' the policies.
The simplest, you could build a SHACL that checks for grammar and not provide the "rightOperand", and your system highlights the error in validation:
"constraint": [{
"leftOperand": "spatial",
"operator": "neq", // change to the operand you require
// the SHACL will not let this be an Agreement
}]
You could also expand the model and provide a property for "the stage that builds" an odrl:Agreement
to fill it in:
"profile": "http://ex.com/odrl:profile:toru-profile", // profile:toru-profile defines what "templateValue" is
"constraint": [{
"leftOperand": "spatial",
"operator": "neq", // change to the operand you require
"rightOperand": null,
"templateValue": true // it is left to the next step to "fill in the value"
}]
In the SHACL, the feedback might be more generic than that which could be provided by way of the profile (e.g. the profile could have a link to a taxonomy of what values are allowed between the 9898.movie
and spatial
).
A possible issue with this for your users: if they want to "test" your rules before releasing them, you have paths to incomplete/untestable constraints (although this could be "filled in" by the development environment, but i don't know your end goal).
I've opened 65 - but it is a proposal to deal with templates in a different way.
As Josh indicated, your proposal would not be semantically correct.
However, another option is to consider using rightOperandReference.
"constraint": [{
"leftOperand": "spatial",
"operator": "eq",
"rightOperand": "http://example.com/getSpatial?asset=http://example.com/asset:9898.movie"
Thank you for your reply. We have found that there is more than one available path. I will mainly consider the method of using Profile, which is appropriate for expressing ODRL, and I will consider whether there is a method that better suits the use case by using rightOperandReference
together.
This issue will be closed as the question has been resolved.
Thank you.
Hello, we are thinking of publishing ODRL to a data catalog like CKAN. Specific restrictions may not have been determined at the time of catalog publication. Is it possible to express just that there is restriction in such a case?
For example, we suspect that we can indicate that there is a spatial constraint by writing
is not null
as shown below.Is this method correct? Please let me know if another method is correct. Thank you.