https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81093
--- Comment #4 from janus at gcc dot gnu.org --- (In reply to Dominique d'Humieres from comment #3) > > ! wrong result > > write(*,*) gen_data%mydata(:)%data1(1), gen_data%mydata(:)%data1(2) > > As said before I am not sure this is valid (a data_type2 object printed a > data_type1 one). I don't see why it would not be valid. data_type2 is an extension of data_type1, so it certainly has both components, data1 and data2. For an object which is declared as "class(data_type1)", you can always access the data1 component without any problem, no matter if the dynamic type is data_type1 or data_type2. For the data2 component this is not the case. It is only present if the dynamic type is data_type2 (or an extension of it), and you need a SELECT TYPE statement to print it. Does that alleviate your concerns or did I just not get your point?
