jwg12 IfcRevolvedAreaSolid AxisStartInXY

[aothms]

Where rule fix due to more general point type usage in IfcPlacement

[aothms]

--- tmp/a/IFC.exp	2022-08-26 23:08:09.300513800 +0200
+++ tmp/b/IFC.exp	2022-09-05 20:18:40.046645700 +0200
@@ -10094,21 +10094,21 @@
  SUPERTYPE OF (ONEOF
 	(IfcRevolvedAreaSolidTapered))
  SUBTYPE OF (IfcSweptAreaSolid);
 	Axis : IfcAxis1Placement;
 	Angle : IfcPlaneAngleMeasure;
  DERIVE
 	 AxisLine : IfcLine := IfcRepresentationItem() || IfcGeometricRepresentationItem () || IfcCurve() || IfcLine(Axis.Location, 
   IfcRepresentationItem() || IfcGeometricRepresentationItem () || IfcVector(Axis.Z,1.0));
  WHERE
 	AxisDirectionInXY : Axis.Z.DirectionRatios[3] = 0.0;
-	AxisStartInXY : Axis.Location.Coordinates[3] = 0.0;
+	AxisStartInXY : ('IFC4X3_DEV.IFCCARTESIANPOINT' IN TYPEOF(Axis.Location)) AND (Axis.Location\IfcCartesianPoint.Coordinates[3] = 0.0);
 END_ENTITY;
 
 ENTITY IfcRevolvedAreaSolidTapered
  SUBTYPE OF (IfcRevolvedAreaSolid);
 	EndSweptArea : IfcProfileDef;
  WHERE
 	CorrectProfileAssignment : IfcTaperedSweptAreaProfiles(SELF\IfcSweptAreaSolid.SweptArea, SELF.EndSweptArea);
 END_ENTITY;
 
 ENTITY IfcRightCircularCone

[SergejMuhic]

Okay, I am afraid I experienced a serious brain fart here. I cannot seem to get my head around the extra check. If this check is required, would it not make more sense to just restrict IfcAxis1Placement.Location in the first place? It would be consistent with IfcAxis2Placement2/3D.

[aothms]

It seems LocationIsCP already exists on IfcAxis1Placement, so the first part is indeed redundant. I've created an issue so can we can address that in the next tc buildingSMART/IFC4.3.x-development#535

[SergejMuhic]

Well then, this is great. No need for such a check.