Community
Participate
Working Groups
In [let pOtherTypes : OrderedSet(Type) = pTypes - pUMLTypes - pTypesTypes - pExpressionsTypes] where the inputs are all OrderedSets there is no diagnosis of the problem assigning a Set to an OrderedSet; just a silent non-execution of the invalid let.
An Acceleo "let" block has the meaning of a Java "instanceof" operator. In the current case, the "let" is just not executed since the types don't correspond when checked at runtime.
The report was about a missing diagnosis not an incorrect functionality. It is customary for tools to give warnings about dead code. This is a case for just such a warning.