New subject: What's a sensible definition of FAMIXMethod>>incomingAccesses?

12 Jan 2011

Take a look at the following definitions:

FAMIXClass>>qIncomingAccesses
	^ self attributes flatCollect: [ :each | each incomingAccesses ]

FAMIXScopingEntity>>qIncomingAccesses  "Package, Namespace"
	^ self classes flatCollect: [ :c | c qIncomingAccesses ]

Now does it make sense to have a similar definition for FAMIXMethod?

FAMIXMethod>>qIncomingAccesses
	^ (self parameters collect: #incomingAccesses)
		addAll: (self localVariables collect: #incomingAccesses);
		yourself

We can't really say there are 'incoming accesses' on attributes in a method.
On the contrary, methods are the ones which make (outgoing) accesses to attributes. On the
other hand, incoming accesses on parameters and local variables are necessarily from the
lexical scope, the method itself: so not a very interesting information. And it's not
consistent with the FamixClass and FamixScopingEntity, which collect all access to
attributes within their scope.

In other words, we can say:
aClass qIncomingInvocaions = aClass methods flatCollect: #qIncomingInvocations

but:
aClass qIncomingAccesses ~= aClass methods flatCollect: #qIncomingAccesses

I am wondering if it would not be better to just say:

FAMIXMethod>>qIncomingAccesses

	^ Array new

Or maybe I'm just trying to split hairs

--
Simon