Re: Real continuations

[Stefan Monnier]

> btw, would it be practical to support real first class continuations in
> Elisp?

Real first-class continuations require reifying the stack as an object
reachable from the heap and that can be copied.  There are many
different ways to do that to try and keep the benefits of the normal
representation of the stack, but I think it could prove difficult to
implement something like call/cc given the current code base.

> When you exit a recursive edit, isn't what you get (internally) more or
> less already the execution of some kind of continuation?

Indeed catch/throw do provide some of the functionality of call/cc, as
do threads.  Together they cover a large part of what call/cc is
typically used for, but they still don't provide call/cc.


        Stefan