> > Right now this method is prepare to handle StmtShadowMatches and
> > BodyShadowMatches but I guess your shadow match is neither of those,
> > right? Therefore you might have to subclass ShadowPointsSetter. The
> > points is that you need to have created a correct version of
> > ShadowPoints with the correct begin/end Jimple statements.
>
>
> Out of source-view i extend with my BlockShadowMatch the
> StmtShadowMatch.
This is why you get the behaviour you are seeing right now. A
StmtShadowMatch is a represents a shadow for a *single* statement.
Therefore you should not subclass StmtShadowMatch but rather
ShadowMatch.
> The stmt of my BlockShadowMatch ist the beginning of
> my "source-code"-block. Additional to this stmt my ShadowMatch
> includes a second statement being the end statement of the
> "source-code"-block. When i got it right i have to provide an own
> Weaver and overrride the Weaver.inlineConstructors() method in order
> to instantiate my own ShadowPointsSetter.
>
> I copy a lot of code from the original ShadowPointsSetter into my own
> an add a simple routine which handles my BlockShadowMatches.
Yes to me this sounds like the correct way to do it.
Eric
-- Eric Bodden Sable Research Group McGill University, Montréal, CanadaReceived on Mon Apr 21 2008 - 15:01:02 BST
This archive was generated by hypermail 2.2.0 : Tue Apr 22 2008 - 10:40:07 BST