Community
Participate
Working Groups
I have this simple aspect and class that gives a spurious error in the editor: public aspect RR { public R<?> R.c() { return null; } after() : execution(R.new()) { } } class R<T> { } Delete the '<?>' or the after advice and the spurious error disappears.
It's the question mark. Parser/scanner is thinking it is part of a ? : operator; Solution is to keep track of type parameter nesting.
Fixed with regression tests. Will be available in next dev build. Also, addressed situations where there are nested type parameters, inner classes and inner aspects that use '?'.
