I ran into something while checking the georeferencing of a project and wanted to ask before assuming it’s a misunderstanding on our end.
First the context:
https://standards.buildingsmart.org/IFC/RELEASE/IFC4/ADD1/HTML/schema/ifcmeasureresource/lexical/ifccompoundplaneanglemeasure.htm
IfcCompoundPlaneAngleMeasure (used for IfcSite.RefLatitude/RefLongitude) has 4 values - degrees, minutes, seconds, and a fourth value for the millionth-seconds of arc, which should always be divided by 1,000,000.
Example:
IFCSITE(...,(48,7,30,493000),(14,53,0,34000),274.86,...)I expected: 493000/1,000,000 = 0.493″, 34000/1,000,000 = 0.034″ so 48° 07′ 30.493″ N | 14° 53′ 00.034″ E
But Solibri’s native display shows: 48°7’30.493000" | 14°53’0.34000" - the latitude is correct, but the longitude is off by a factor of 10 (0.34 instead of 0.034).
Is there some rounding/formatting logic on your side that could explain this, or does this look like a bug?
One (good) thing noted: I tested by pulling the raw values using the API directly, and it already returns the full, correct raw value (LON -> deg=14 min=53 sec=0 micro=34000) exactly as in the native file. So the issue seems to probaly sit in the display/formatting step, not the data access.
Maybe I am a bit confused already by all the numbers and transformations but is this an issue?
EDIT: It looks like this only happens when the fourth value is below 1000.