[gambit-list] Marc, does Gambit evaluate left-to-right order always or only generally?

Adam adam.mlmb at gmail.com
Sat Nov 18 03:32:15 EST 2017


Evaluation is always separated between procedure calls though, so the
undefined order only regards specifically the order in which the operator
and arguments are evaluated, right?

(
E.g.:

For the code (A (B C D) (E F G)), let's call (B C D) X1 and (E F G) X2,
here: While the evaluation order within the block B C D, and within the
block E F G, and within the block A X1 X2, is undefined, the internal
evaluation of X1 and X2 is never mixed so that you'd get an evaluation
order B to F to D, right?

Any mix where A, X1, X2 are kept separate could happen thoug, e.g. A, C, B,
D, F, E, G or F, E, G, D, B, C, A.
)

2017-11-14 10:03 GMT+08:00 Marc Feeley <feeley at iro.umontreal.ca>:

> Left as an exercice to the reader…
>
> Marc
>
>
>
> > On Nov 13, 2017, at 10:37 AM, Adam <adam.mlmb at gmail.com> wrote:
> >
> > Wow.
> >
> > Can you give an actual code exampe where compiled (or interpreted) code
> not does left to right?
> >
> > I think all I observed in practice in the past was left to right.
> >
> > Adam
> >
> >
> > 2017-11-13 21:45 GMT+08:00 Marc Feeley <feeley at iro.umontreal.ca>:
> > Execution order is arbitrary.  The interpreter and compiler don’t use
> the same ordering.  The interpreter tends to be mostly left-to-right, but
> don’t count on it.
> >
> > Marc
> >
> >
> >
> > > On Nov 12, 2017, at 4:19 PM, Adam <adam.mlmb at gmail.com> wrote:
> > >
> > > Hi Marc!
> > >
> > > Does Gambit evaluate left-to-right order always or only generally?
> > >
> > > Thanks,
> > > Adam
> >
> >
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.iro.umontreal.ca/pipermail/gambit-list/attachments/20171118/b909ecc1/attachment.htm>


More information about the Gambit-list mailing list