[m-rev.] for review: allow for only a single positive leap second

[Julien Fischer]

On Tue, 14 Apr 2026 at 22:48, Zoltan Somogyi <zoltan.somogyi at runbox.com> wrote:

>
> On Tue, 14 Apr 2026 21:58:28 +1000, Julien Fischer <jfischer at opturion.com> wrote:
> > diff --git a/NEWS.md b/NEWS.md
> > index d70465ef1..3e78b3c8c 100644
> > --- a/NEWS.md
> > +++ b/NEWS.md
> > @@ -246,6 +246,9 @@ Changes to the Mercury standard library
> >      - func `det_date_from_string/1` (replacement:
> > `det_date_time_from_string/1`)
> >      - func `date_to_string/1`       (replacement: `date_time_to_string/1`)
> >
> > +* `date_time/0` and `duration/0` values now only allow for a single positive
> > +  leap second.
> > +
> >  * The following functions and predicates have been added:
> >
> >      - pred `init_date_time/8`
> > @@ -1424,6 +1427,11 @@ Changes to the Mercury standard library
> >      - pred `spawn_native_joinable/5`
> >      - pred `join_thread/4`
> >
> > +### Changes to the `time` module
> > +
> > +*  The `tm_sec` field of the `tm/0` type now only allows for a single positive
> > +   leap second.
>
> Both of these sentences are misleading without the qualifier "for any given minute",
> since we still allow, as we have to, more than a single leap second per year.

Added.

Thanks for the review.

Julien.