Thanks for the quick response. I'm unsure how TBC handles sh:node, since my toy example validates without errors, while my real code hangs forever. What is your intended TBC behavior? (Then I will dig in and try to figure out which of my examples doesn't work that way).
Steve On Fri, Jul 2, 2021 at 11:10 AM Irene Polikoff <[email protected]> wrote: > https://www.w3.org/TR/shacl/#shapes-recursion > > On Jul 2, 2021, at 10:50 AM, Steve Ray <[email protected]> wrote: > > I believe that I now understand everything you say regarding sh:node. I > recognize that any instance that satisfies the constraints of the NodeShape > will pass the sh:node constraint, thus not error is raised. > > All was good until I found that I was hanging TBC. My hypothesis is that > in determining satisfaction of the sh:node constraint, the engine had to > examine all the properties of the NodeShape. Since some of those properties > in turn contained sh:node constraints, it had to recurse in to check if > those properties were satisfied. Since some of the instances pointed to > each other, I figured I had sent TBC into an infinite recursion. > > To check this, I built a toy example (attached) with a "sibling" relation > with the same pattern. However, this example seems to validate just fine, > without hanging. I'm wondering why the same infinite recursion is not > happening. > > Steve > > > > > On Wed, Jun 23, 2021 at 3:09 PM Irene Polikoff <[email protected]> > wrote: > >> >> Not sure I understand the question. >> >> sh:node :Connectable >> >> Simply means that the values need to satisfy the constraints defined >> for :Connectable. If you have a type constraint defined for :Connectable >> (i.e., a property have that rdf:type must be some specific class), then it >> will be evaluated, but if you don’t, then, it does not matter what is the >> type of the value resource. >> >> >> Given your model, there are two ways to identify nodes to validate >> against :Connectable - 1)values of :connectsFrom for instances of >> SomeOtherClass and instances of Device and other target classes. This >> includes instances of subclass. >> >> >> On Jun 23, 2021, at 3:00 PM, Steve Ray <[email protected]> wrote: >> >> I forgot one detail, which is that I have a subClassOf hierarchy under >> Device, and the instances of any of those subClasses do not seem to >> validate against the Connectable shape. Does sh:node traverse the >> subClassOf hierarchy of the sh:targetClass values? >> >> Steve >> >> >> >> >> On Wed, Jun 23, 2021 at 11:47 AM Irene Polikoff <[email protected]> >> wrote: >> >>> What behaviour did you expect? >>> >>> Since Connectable is just a shape and does not have any instances, then >>> you can’t use it in sh:class. This constraint says that the value must be >>> an instance of Connectable, which will always be wrong. >>> >>> When you say sh:node s223:Connectable, it means that values should >>> validate against the constraints defined for Connectable. For this scenario >>> targets associated with Connectable are ignored since the constraint is >>> already targeting - all values of s223:connectedFrom that belong to members >>> of SomeOtherClass will be validated against the shape. >>> >>> All instances of Device, DomainSpace and Sensor will also be validated >>> against the Connectable shape - because of the target statements. >>> >>> https://www.w3.org/TR/shacl/#NodeConstraintComponent >>> >>> https://www.w3.org/TR/shacl/#focusNodes >>> >>> On Jun 23, 2021, at 1:19 PM, Steve Ray <[email protected]> wrote: >>> >>> The discussion you have been having regarding sh:targetClass reminds me >>> of a problem I am having, as follows. >>> >>> I'm defining an abstract "mixin" shape, Connectable, that has various >>> properties used by a number of target classes. I do not want to allow >>> instantiation of Connectable, so I defined it only as a NodeShape, but not >>> an owl:Class. (I cannot use the dash:abstract property without stepping >>> outside strict SHACL). Then, I need other classes to have the equivalent of >>> a sh:class constraint on properties that point to instances of all those >>> target classes. I assume I cannot use sh:class as in the example below, >>> since Connectable is not a class. I tried sh:node, since the documentation >>> implies that's what I need, but I'm not getting the expected behavior. >>> >>> Am I misunderstanding the use of sh:node? >>> >>> s223:Connectable >>> a sh:NodeShape ; >>> ...various properties >>> sh:targetClass s223:Device ; >>> sh:targetClass s223:DomainSpace ; >>> sh:targetClass s223:Sensor ; >>> . >>> >>> s223:Device >>> a owl:Class ; >>> a sh:NodeShape ; >>> ... >>> . >>> >>> # Presumed invalid case >>> s223:SomeOtherClass >>> a owl:Class ; >>> a sh:NodeShape ; >>> sh:property [ >>> sh:path s223:connectsFrom ; >>> sh:class s223:Connectable ; >>> ... >>> >>> #Expected valid case >>> s223:SomeOtherClass >>> a owl:Class ; >>> a sh:NodeShape ; >>> sh:property [ >>> sh:path s223:connectsFrom ; >>> sh:node s223:Connectable ; >>> ... >>> >>> >>> Steve >>> >>> >>> >>> -- >>> You received this message because you are subscribed to the Google >>> Groups "TopBraid Suite Users" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to [email protected]. >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/topbraid-users/CAGUep86aRz_enWzvhfMkT5pW0SvNA3XYdbdB%2BAsSXP%3DfEFAusQ%40mail.gmail.com >>> <https://groups.google.com/d/msgid/topbraid-users/CAGUep86aRz_enWzvhfMkT5pW0SvNA3XYdbdB%2BAsSXP%3DfEFAusQ%40mail.gmail.com?utm_medium=email&utm_source=footer> >>> . >>> >>> >>> >>> -- >>> You received this message because you are subscribed to the Google >>> Groups "TopBraid Suite Users" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to [email protected]. >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/topbraid-users/3A11B544-71EA-48BD-B068-0BA71EBD5E1D%40topquadrant.com >>> <https://groups.google.com/d/msgid/topbraid-users/3A11B544-71EA-48BD-B068-0BA71EBD5E1D%40topquadrant.com?utm_medium=email&utm_source=footer> >>> . >>> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "TopBraid Suite Users" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/topbraid-users/CAGUep86S_teBHHHDmhb13U1sQ4ctx197%3D2LLb60p8BtwiRK69A%40mail.gmail.com >> <https://groups.google.com/d/msgid/topbraid-users/CAGUep86S_teBHHHDmhb13U1sQ4ctx197%3D2LLb60p8BtwiRK69A%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "TopBraid Suite Users" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/topbraid-users/BD2E5E3F-BD00-43F3-B46F-8FC49117B2AF%40topquadrant.com >> <https://groups.google.com/d/msgid/topbraid-users/BD2E5E3F-BD00-43F3-B46F-8FC49117B2AF%40topquadrant.com?utm_medium=email&utm_source=footer> >> . >> > > -- > You received this message because you are subscribed to the Google Groups > "TopBraid Suite Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/topbraid-users/CAGUep86o-GWEYks1zFRu0KnN_QpkLE03yvdWMgmJnV3CmA_dnw%40mail.gmail.com > <https://groups.google.com/d/msgid/topbraid-users/CAGUep86o-GWEYks1zFRu0KnN_QpkLE03yvdWMgmJnV3CmA_dnw%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > <TestShNode.shapes.ttl> > > > -- > You received this message because you are subscribed to the Google Groups > "TopBraid Suite Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/topbraid-users/09E3A686-8FF7-449D-B639-5B906C0631AE%40topquadrant.com > <https://groups.google.com/d/msgid/topbraid-users/09E3A686-8FF7-449D-B639-5B906C0631AE%40topquadrant.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "TopBraid Suite Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/topbraid-users/CAGUep86ri3JpAarLqTT%3DutDBkS7k8hU0N20N%2BpCdroDg-sufgA%40mail.gmail.com.
