How requirements on enumerated properties should be interpreted about ids

[andyward]

Isn't it asking new a value of "NEW" in the Enumerated Value property?

             <requirements>
                <property measure="IfcLabel" minOccurs="1" maxOccurs="1">
                    <propertySet>
                        <simpleValue>Pset_WallCommon</simpleValue>
                    </propertySet>
                    <name>
                        <simpleValue>Status</simpleValue>
                    </name>
                    <value>
                        <simpleValue>NEW</simpleValue>
                    </value>
                </property>
            </requirements>

But the entity actually contains "EXISTING" and "DEMOLISH", so you'd expect it to fail?

#11=IFCPROPERTYENUMERATEDVALUE('Status',$,(IFCLABEL('EXISTING'),IFCLABEL('DEMOLISH')),#10);

[aothms]

The ids file referenced by @pasi-paasiala is indeed intended to fail as indicated by the name *fail*-any_matching_value_in_an_enumerated_property_will_pass_3_3.ids. Maybe the intended ids file to referenced was this one: pass-any_matching_value_in_an_enumerated_property_will_pass_1_3.ids

I think the current behaviour in the test cases is consistently that for any multi-valued construct (layered material, bounded prop, list prop, enum prop, ...) that at least one of these elements then must match the constraint.

It's been a while, but I also remember it has been discussed that for applicability the behaviour could be to have any of the values match: "A concrete wall is a wall with at least one layer of concrete", but for the requirement part all should match: e.g as you said: "existing or demolish, but nothing else". I think this was scrapped for being too magical and/or confusing.

So I don't think there is currently a way to specify that.

[CBenghi]

Make sure the test suite covers passes and failures appropriately. This specific file seems to be ok.