MD_Marx |
05-09-2012 12:46 PM |
NullReference Exceptions
Quote:
Originally Posted by JG52Uther
(Post 422159)
Intersting Macro. I play with grass and roads off and am having CTD's with the new patch, with the same 'NullReferenceException: Object reference not set to an instance of an object.' error in the logfile.
|
Sorry for sounding 'Techy', but I call myself a 'software engineer', and I too have had CTDs with this given as a reason. Either the report is incorrect (which doesn't make it very helpful for debugging), or it's true. If true, then something is very, very seriously wrong with the initialisation code of objects, because this should NEVER be seen. There are 2 levels of object initialisation - at start-up e.g. objects within the cockpit, and the more demanding level - dynamically i.e. in-flight. I wonder if the dynamic initialisation is falling behind, out of sequence when the method on the object is called? To me, this sounds like a basic threading error; probably difficult to reproduce, but a closer look in the code design/architecture at anything that affects the completion of the initialisation thread seems to be in order. I would have thought that test code could readily be instrumented to log when an object's method is called before the initialisation is complete? It certainly suggests to me that 'defensive' programming isn't being used............
Marx
|