[gme-users] GME4 metainterpreter issues

[Akos Ledeczi]

The new metainterpreter with GME4 has a much simplified
method of picking the "dominant" object of a SameAspect,
SameFolder or Equivalence relationship. It is always
the object that connects to the operator with the "Right"
connection, i.e. SameAspectRight. This is usually on
the left hand side of the operator for clarity :), but
the connection comes out the right hand side of the object.
Note that you can shoot yourself in the foot if you
use the hotspots when connecting to these operators. In
this case you have to check the connection with the tooltip
(by moving the cursor on top of the connection).

The good part is that now you do not have to guess which
aspect object (that is, name) will be selected. The bad part
is that exisitng metamodels may result in a different
paradigm definition (xmp file) because aspect names (or
even object kindnames in case of equivalence) may have
changed. In this case you have to go back to your
metamodel and change these operators. Note that aspect
ordering may also be effected...

Further bad news is that there is a bug in the aspect mapping.
It may pick up default aspect names that are not valid
in a given context. In this case when you try to open
an existing model with new newly interpreted paradigm, GME
may crash. This has been fixed and will be available with
a new release in the near future.

Yet another problem is that the previous values for aspect
mapping is not being picked up by the new metainterpreter.
This means that the reinterpretation of an existing metamodel
that used aspect mapping before will result in a different
paradigm where the aspects are not mapped correctly. This
will also be fixed shortly. In the meantime the problem can
be avoided by respecifying the aspect mappings.

A good news is that the aspect mapping dialog will only show
valid aspect combination unlike the previous one. If there is
no aspect mapping possible then the dialog box won't even appear.

Sorry for any inconvenience,
Akos