Leo~ On Tue, 16 Nov 2004 18:32:07 +0100, Leopold Toetsch <[EMAIL PROTECTED]> wrote: > Matt Fowles wrote: > > > > I disagree with that analysis. Let us consider the actual effect of > > such an implementation. > > > > First iteration > > > > i = 0; > > foo(); #at this point a continuation created capturing i=0, promoted > > by Jens and stuff happens > > #eventually it is invoked, restoring i=0 > > i++; #i = 1 > > foo(); #at this point a NEW return continuation is created capturing > > That would work if there is a one to one representation of the invoation > of foo() an it's continuation. But no one guarantees that.
I suppose that what I am arguing is that anyone who does not maintain such a one-to-one representation (at least from the perspective of code calling foo()); full well deserves what they get. They are restoring the execution to an earlier state by invoking an old continuation. If the earlier state called them again the first time, they should probably expect the earlier state to call them again the second time. Unless they have some specific knowledge that the earlier state will change it behavior (because things in the heap have changed), there should be no expectation for it to. Matt -- "Computer Science is merely the post-Turing Decline of Formal Systems Theory." -???
