GEM056 No duplicate geometric contexts

[Moult]

I've come across a vendor that currently generates a new geometric representation context per shape representation. So we have 500,000 Model / Body / MODEL_VIEW contexts. Yikes.

It's not illegal, but I'm almost certain it should be, given the semantic value of switching contexts and the usefulness of the RepresentationsInContext inverse.

[pjanck]

I'm not sure, what you mean with

It's not illegal

considering this:

The use of one instance of IfcGeometricRepresentationContext to represent the model (3D) view is mandatory, the use of a second instance of IfcGeometricRepresentationContext to represent the plan (2D) view is optional (but needs to be given, if there are scale dependent plan views), the additional scale or view dependent contexts need to be handled by using the subtype IfcGeometricRepresentationSubContext pointing to the model view (or the plan view) as the ParentContext.

(own emphasize)

[Moult]

Thanks @pjanck , I agree with the intention, but, it seems to not be spelled out so explicitly for subcontexts, nor to prohibit more than one. It's also subject to vague interpretation (are you allowed to have zero instances for non-geometric IFC datasets?)

Perhaps with a validation rule the validator can add some certainty.

[aothms]

Do you have any proposal for something semi formal to test against? I agree with the premise. Sent from a mobile device, excuse my brevity. Kind regards, Thomas Op wo 9 apr 2025 12:11 schreef Dion Moult ***@***.***>:

…

Thanks @pjanck <https://github.com/pjanck> , I agree with the intention, but, it seems to not be spelled out so explicitly for subcontexts, nor to prohibit more than one. It's also subject to vague interpretation (are you allowed to have zero instances for non-geometric IFC datasets?) Perhaps with a validation rule the validator can add some certainty. — Reply to this email directly, view it on GitHub <#397 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAILWVZGQZQPAJXQWW3LPPT2YTW5LAVCNFSM6AAAAAB2YQ455OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOOBZGE2DGNBVGQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***> *Moult* left a comment (buildingSMART/ifc-gherkin-rules#397) <#397 (comment)> Thanks @pjanck <https://github.com/pjanck> , I agree with the intention, but, it seems to not be spelled out so explicitly for subcontexts, nor to prohibit more than one. It's also subject to vague interpretation (are you allowed to have zero instances for non-geometric IFC datasets?) Perhaps with a validation rule the validator can add some certainty. — Reply to this email directly, view it on GitHub <#397 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAILWVZGQZQPAJXQWW3LPPT2YTW5LAVCNFSM6AAAAAB2YQ455OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOOBZGE2DGNBVGQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[Moult]

How about a check that there exists no two IfcRepresentationContext (incl. subcontext) with the same values of ContextIdentifier, ContextType, TargetScale, TargetView, and UserDefinedTargetView?

[aothms]

Sounds good to me. This would be rather similar to our rule requiring identical values on the geospatial coordinate op. Therefore easy to implement. Multiple project libraries would complicate things though... We'll add it to the backlog. Sent from a mobile device, excuse my brevity. Kind regards, Thomas Op do 10 apr 2025 01:55 schreef Dion Moult ***@***.***>:

…

How about a check that there exists no two IfcRepresentationContext (incl. subcontext) with the same values of ContextIdentifier, ContextType, TargetScale, TargetView, and UserDefinedTargetView? — Reply to this email directly, view it on GitHub <#397 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAILWV4WEPEWXVDM5LLGWE32YWXQZAVCNFSM6AAAAAB2YQ455OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOOJRGIYTMNZSHE> . You are receiving this because you commented.Message ID: ***@***.***> *Moult* left a comment (buildingSMART/ifc-gherkin-rules#397) <#397 (comment)> How about a check that there exists no two IfcRepresentationContext (incl. subcontext) with the same values of ContextIdentifier, ContextType, TargetScale, TargetView, and UserDefinedTargetView? — Reply to this email directly, view it on GitHub <#397 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAILWV4WEPEWXVDM5LLGWE32YWXQZAVCNFSM6AAAAAB2YQ455OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDOOJRGIYTMNZSHE> . You are receiving this because you commented.Message ID: ***@***.***>