Strictly speaking, the only properties that should exist in a pset should be those in a pset template. However, there are situations where this is not the case, such as when the pset template has since been modified, or if there is invalid data coming from other software, or migrating between schemas. In this case, we should still load the property, and give the user the option to edit it (or null it).
a bit related to #4704
1) fixed util.cost.get_total_quantity
2) fixed similar issue in cost.data that calculates the final value that user will see in UI
3) changed UI, "-" is shown when there are no quantities and "0" is when quantities are there but they just equal to zero.
Before - https://i.imgur.com/EO53DhM.png
After - https://i.imgur.com/H6rK4sP.png
fyi @myoualid
Previously, we were clever in detecting types and occurrences because if you edited a type, you wanted all occurrences to be updated. Now, we no longer need to be clever because like any other UI element, the annotations shown in the active view are updated after every IFC operation.
It was failing with an error below if you'd try to edit any property since `primary_measure_type()` returned `[]` because it was executed before `pset_templates()` and `IfcStore.pset_template_file` wasn't yet set.
```python
Error: Python: Traceback (most recent call last):
File "\blenderbim\bim\module\pset_template\operator.py", line 146, in execute
props.active_prop_template.primary_measure_type = template.PrimaryMeasureType
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: bpy_struct: item.attr = val: enum "IfcLabel" not found in ()
```
1) description to emphasize that active object will be the one that's cutting
2) skip non-ifc elements, previously they failed to process with error `AttributeError: 'NoneType' object has no attribute 'RepresentationType'`. Also skip objects without representations.
3) add an info message at the end
Here's a short gif in case if anyone doesn't about that feature: https://imgur.com/a/338w7jx
Occurred only with only_assigned_to_faces = True
A bit related to #4675
Error was
styles = [style for style, usage in zip(styles, usage_count, strict=True) if usage > 0]
ValueError: zip() argument 2 is longer than argument 1
On large projects predict dense mesh stage can take 20s+ and reusing attribute value can save up to half of this time.
Using indices also helps but it's not that significant and sometimes it's the same time as using attribute names.
Andrej730
