Add ResourceService::supportedInteractionModels method #95
Conversation
ajs6f
left a comment
ajs6f
left a comment
There was a problem hiding this comment.
Beautiful, one little style thing (in three or four places) but this is the right idea.
| allModes.add(ACL.Read); | ||
| allModes.add(ACL.Write); | ||
|
|
||
| allInteractionModels.add(LDP.Resource); |
There was a problem hiding this comment.
If we have Guava on line (I think we do) maybe use something from Sets here to construct and fill the set?
There was a problem hiding this comment.
Guava isn't a compile dependency, but I have no problem with adding it as a test dependency (if it's not already there).
| allInteractionModels.add(LDP.Container); | ||
| allInteractionModels.add(LDP.BasicContainer); | ||
| allInteractionModels.add(LDP.DirectContainer); | ||
| allInteractionModels.add(LDP.IndirectContainer); |
| allInteractionModels.add(LDP.Container); | ||
| allInteractionModels.add(LDP.BasicContainer); | ||
| allInteractionModels.add(LDP.DirectContainer); | ||
| allInteractionModels.add(LDP.IndirectContainer); |
|
I think the Trellis ontology should change |
|
Just waiting for Travis et al to complete |
|
Good? |
|
@ajs6f yes, feel free to squash/merge etc whenever you'd like |
Resolves #92
Once we have agreement on this approach, I can apply it to #88, allowing the test suite to adjust to varying levels of LDP container support.
I decided not to include a default implementation in the interface -- if we want implementations to be clear about what is supported, then I thought it makes more sense to simply require an implementing class to implement the method.
I am also making use of a new named individual in the Trellis ontology
trellis:InvalidInteractionModel, which is used with theldp:constrainedBylink header in the case of an error condition.