NumberFormatException prevents the model from loading

JSN

I have a large (>1GB) IFC which won’t open in Solibri.

I guess this is not the only issue on the model, and the file might not be correct but at the same time the error message leaves some questions behind.

The relevant line from the file:

#432303 = IFCBUILDINGELEMENTPROXY('33Ty12J7LCuR_J0JVCcNjC', #8, 'Fugend\X\FCbel1:Fugenduebel1:14589475', $, 'Fugend\X\FCbel1:Fugenduebel1', #20273960, #432304, '14589475', .37.);

There might be truly an issue with file as the “.37.” is indeed a bit unusual formatting, but without knowing how this happens in first place I have to add that the model opens fine in other viewers. I wanted to ask again about the general exception policy or if this is indeed something Solibri has to adress?

Btw. correcting the .37. fixes the inital error, but the next one is not far.

Not sure if the authors of the model have much influence on this either.

manuel.furia

We can easily make that parse, I can make an issue about it. But again, it is not STEP compliant and an obvious bug in the authoring tool, so it would be good to report it as a bug to whatever company makes the tool.

manuel.furia

Of course, this also creates an issue of ambiguity. Did the tool mean .37 (as of 0.37) or 37. (as of 37.0)? We can make an assumption to always take the first interpretation, but we don’t really know, so it may lead to wrong data.

EDIT:
The value in that place is not even supposed to be a number, but one of the building proxy type enumeration. See:

JSN

Indeed, something is not quite right, and I suspect those lines are not the only questionable aspects of the file. Still they load with some third-party tools. However I also find it problematic when they open load without any indication of existing errors. That said, completely blocking the file from loading is not an ideal solution either. If exceptions are implemented, at least a disclaimer would be very much appreciated.