Hi Doug:

>From dougo@ccs.neu.edu Fri Aug  1 17:50:06 1997
>From: Doug Orleans <dougo@ccs.neu.edu>
>To: Karl Lieberherr <lieber@ccs.neu.edu>
>Cc: dem@ccs.neu.edu
>Subject: Re: compatibility strategies/class graphs
>
>Karl Lieberherr writes:
> > Proposal (Boaz):
> > A strategy graph S is compatible with a class graph G under the default
> > name map if for all strategy graph edges e of S there is a path in G
> > from Source(e) to Target(e).
> > 
> > How do we deal with the edge constraints attached to strategy edges,
> > i.e., the negative path constraints?
>
>A strategy graph S is compatible with a class graph G, a name map N,
>and edge constraints B if for all strategy graph edges e of S there is
>a path P in G from s = N(Source(e)) to t = N(Target(e)) such that for
>every element [node or edge] x in P except for s and t, (B(e))(x) is
>true. [B(e) is the edge constraints of the strategy edge e, i.e. a
>predicate function mapping path elements to booleans.]
>

This is a good definition. It means that we can have 
bypassing ->X,y,Y, although X and Y are not in the class graph.
This could be dangerous, however, since the user might have intended
to bypass edge: ->X,y,Z in the class graph.

> > What about only-through?
> > What is your definition of only-through? You said it is the negation
> > of bypassing. Please can you refresh my memory.
>
>The constraint "bypassing -> *,x,*" is a predicate that returns false
>for any construction edge labelled "x", and true for everything else.
>The constraint "only-through -> *,x,*" is a predicate that returns
>true for any constraint edge labelled "x", and false for everything else.
	      ^^^^^^^^^^
              construction is what you meant.
>
Very elegant and symmetric. It must be right.

>--Doug
>

--Karl