Checking start/end dates are done very, very often in date calculations (often looping through every day in a date range). So I'd prefer more opaque code but less overhead.
since we don't use `file.add` to add inverse items and we use `file.create_entity` it was duplicating new elements, added optional parameter that makes it possible to track already added relationships between `append_asset`s.
issue occurred with ifcpatch ExtractElements - it was adding project from other ifc file (with the contexts) and then adding some elements from it using `append_asset`. During `append_asset` it was considering existing context with the duplicated one and was partially purging it.
as probably in the most cases user would want 1 style per 1 representation item
After this it's finally possible in BBIM to assign styles to representation items through the materials: https://imgur.com/a/cgiJ1KV
If multiple object placements were reusing the same IfcAxis2Placement then editing one of them would remove IfcAxis2Placement ignoring it's inverse references making all other object placements invalid.
When PredefinedType is not found in the related enum reassign_class is considering it USERDEFINED and trying to set .ObjectType. The problem was it was doing for type classes also, when it should have set .ElementType instead.
Why it crashed BlenderBIM - in IfcClassData.data["has_entity"] we store currently active IFC entity which is recreated during `ifcopenshell.util.schema.reassign_class` making old entity invalid. Since `root.reassign_class` was failing in the process, related BBIM operator was failing too, IfcClassData wasn't updated and removed entity was accessed from UI leading to crash.
- error in settings poll message for bim.assign_selected_as_product
- test that drawing.assign_product won't create new assignments for the same product
replaced shapebuilder with general ifcopenshell code in `test_add_boolean` (the only place besides `test_shape_builder` where it was used), that way mathutils dependency in tests is localized only to `test_shape_builder.py`.
added test-safe option to makefile, so github workflow would ignore `test_shape_builder` tests for now and it wouldn't interrupt the entire workflow
added test that model.get_manual_booleans will ignore IfcBooleanClippings until we find some other solution to distinguish automatic and manual clippings. And some refactor along the way.
Noticed a couple hundreds of orphaned IfcPropertyEnumeration in my .ifc like below 😁
IFCPROPERTYENUMERATION('Status',(IFCLABEL('NEW'),IFCLABEL('EXISTING'),IFCLABEL('DEMOLISH'),IFCLABEL('TEMPORARY'),IFCLABEL('OTHER'),IFCLABEL('NOTKNOWN'),IFCLABEL('UNSET')),$);
It was producing orphaned IfcLocalPlacement if placement was parent to some other one. During edit_object_placement we were replacing all parent references with the new placement but wasn't removing the old one.
previously it wasn't deleting object placement when you removed the objects (and related elements like IfcDirections etc) and working on project long enough you might had a bunch of orphaned data in it
Andrej730