[m-rev.] for review: require distinct type variables in existentially quantified data ctors

[Julien Fischer]

On Mon, 22 Jul 2024 at 17:50, Zoltan Somogyi <zoltan.somogyi at runbox.com> wrote:
>
>
>
> On Mon, 22 Jul 2024 16:27:16 +1000, Julien Fischer <jfischer at opturion.com> wrote:
> > For review by anyone.
> >
> > There should be a change to the reference manual as well, but as the
> > compiler seems to be quite inconsistent about whether variables in
> > quantifiers must be distinct or not, I'll leave that until I have
> > looked into the other cases.
>
> That's fine. My review is attached.

Updated diff attached.  Note that the new diff  no longer uses parse_vars.
(That predicate appears to be dead code and can probably be removed.)

Julien.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DIFF.reqtv2
Type: application/octet-stream
Size: 6611 bytes
Desc: not available
URL: <http://lists.mercurylang.org/archives/reviews/attachments/20240722/7292af0f/attachment.obj>