From: Eric B. <er...@go...> - 2004-05-31 19:01:32
|
Colin Paul Adams wrote: > Eric> The reason why you don't get it in ISE is probably because > Eric> it is an invariant violated while checking another > Eric> assertion. > > I see. > What I can't find at the moment is the invariant clause that is failing. From what I could see in the SE exception trace, it looks like you have a decimal object which is not special (special = 0) and which has a Void `coefficient'. This violates one of the invariants of MA_DECIMAL. Now I don't know why we get such a inconsistent decimal. > For the same test case, VE reports: > > ------------------------------------------------------------------------------- > Program terminated by exception > > Routine: is_equal > Message: Loop variant violation > Code: 10004 > Object: MA_DECIMAL_COEFFICIENT_IMP > ------------------------------------------------------------------------------- This looks like another error. But it is perhaps a consequence of the same bug, but caught later if the level of nested checked assertions is different. -- Eric Bezault mailto:er...@go... http://www.gobosoft.com |